technological innovation in the energy sector: case of the
TRANSCRIPT
Technological innovation in the energy sector:
case of the organic Rankine cycle
Thesis submitted to obtain the degree of Doctor in Applied Economics by
Sanne Lemmens
Supervisors: Prof. dr. em. Aviel Verbruggen Prof. dr. Johan Braet
March 2017
Supervisors
Prof. dr. em. Aviel Verbruggen (Universiteit Antwerpen)
Prof. dr. Johan Braet (Universiteit Antwerpen)
Members of the doctoral jury
Prof. dr. ir. Michel De Paepe (Universiteit Gent)
Prof. dr. H. Martin Junginger (Universiteit Utrecht)
Dr. Gerrit Jan Schaeffer (Business4Good BV)
Prof. dr. Johan Springael (Universiteit Antwerpen)
Prof. dr. ir. Steven Van Passel (Universiteit Antwerpen)
Nederlandse titel
Technologische innovatie in de energiesector: studie van de organische Rankinecyclus
Cover design Universiteit Antwerpen
Printing and binding Universitas
Cover illustration Femke Lemmens
ISBN 978-90-8994-160-2
© 2017 Sanne Lemmens.
i
Publications
The following works have been published in the period 2012-2017. Some are (partly) included in
this thesis; others are the results of additional research activities.
A1 Journal publications
Lemmens, S., & Lecompte, S., 2017, Case study of an organic Rankine cycle applied for excess
heat recovery: Technical, economic and policy matters, Energy Conversion and Management
138: 670-685.
Lemmens, S., 2016, Cost Engineering Techniques and Their Applicability for Cost Estimation of
Organic Rankine Cycle Systems, Energies 9 (7): 485-503.
Lecompte, S., Lemmens, S., Huisseune, H., van den Broeck, M., De Paepe, M., 2015, Multi-
Objective Thermo-Economic Optimization Strategy for ORCs Applied to Subcritical and
Transcritical Cycles for Waste Heat Recovery, Energies 8: 2714-2741.
Verbruggen, A., Laes, E., Lemmens, S., 2014, Assessment of the actual sustainability of nuclear
fission power, Renewable and Sustainable Energy Reviews 32:16-28.
P1 Conference proceedings
Lecompte, S., Lemmens, S., Verbruggen, A., van den Broek, M., De Paepe, M., 2014, Thermo-
economic comparison of advanced Organic Rankine Cycles, Energy Procedia 61: 71-74.
P3 Conference proceedings
Lemmens, S., 2015, A perspective on costs and cost estimation techniques for organic Rankine
cycle systems, 3rd International Seminar on ORC Power Systems (ASME-ORC2015), 12-14
October 2015, Brussels, Belgium.
Lemmens, S., 2015, The impact of energy policy on energy efficiency in Europe : a case study of
waste heat recovery with an organic Rankine cycle in Flanders, Proceedings of the 28th
International Conference on Efficiency, Cost, Optimization, Simulation and Environmental
Impact of Energy Systems (ECOS 2015), June 30-July 3, 2015, Pau, France.
Lemmens, S., 2015, Waste heat recovery: potential, policy challenges and technology choices,
38th IAEE International Conference, 25-27 May 2015, Antalya, Turkey.
ii
Scientific reports
Lemmens, S., Verbruggen, A., Couder, J., 2016, Excess heat recovery with organic Rankine Cycles
(ORC) in Flanders: policy setting, Report no. ORCNext-D6.1 for the ORCNext project (IWT SBO-
110006), Antwerp, 39pp.
Lemmens, S., 2016, Modelling tools for assessing the economic and financial feasibility of ORC
applications, Report no. ORCNext-D6.2 & D6.4 for the ORCNext project (IWT SBO-110006),
Antwerp, 44pp.
Lemmens, S., 2016, Integration of ORC applications in the energy-economy of functional
production processes and industrial plants, Report no. ORCNext-D6.3 for the ORCNext project
(IWT SBO-110006), Antwerp, 20pp.
Lemmens, S., Verbruggen, A., 2016, Learning processes for energy efficiency technologies: case
study organic Rankine cycles, Report no. ORCNext-D6.5 for the ORCNext project (IWT SBO-
110006), Antwerp, 32pp.
Verbruggen, A., Lemmens, S., Lecompte, S., 2016, Organic Rankine cycle investment as a real
option, Report no. ORCNext-D6.6 for the ORCNext project (IWT SBO-110006), Antwerp, 24pp.
iii
Acknowledgements
Writing a word of thanks after the enduring years of PhD research is customary. Having reached
this point myself, I understand the true value of a supportive environment – for any project, not
just a PhD. Explicit acts of positive appreciation are perhaps not as common as they could, or
should, be, but are nonetheless an important driver of fulfilling personal development and
valuable individual relations. I would therefore like to take this opportunity to express my
sincere feelings of gratitude to all the people who surrounded me in this PhD project.
First of all, I would like to thank my supervisor, Aviel Verbruggen. Your interesting lectures and
ideas have brought out my latent interests for environmental matters and your lessons in
critical thinking will remain relevant in my future endeavours. Thank you for offering me the
opportunity to pursue this PhD. These years have opened an entirely new world for me, in
which I learned about so many new things and met a lot of interesting people. Also thanks to my
supervisor Johan Braet, for welcoming me into his research group for the final years of my
research. Your practice-oriented perspective has more than once been a source of valuable
insights.
I would also like to thank the members of my PhD jury for their valuable comments on my
thesis. Your critical insights and suggestions have undoubtedly improved the quality of this
work.
Thanks to the partners of the ORCNext project, for welcoming me into the project but also for
all the interesting meetings and insights (and a look into the complex world of energy
engineering). Special thanks to Steven, and to Michel, for being patient tutors when I had yet
another technical question.
Furthermore, I would like to thank Johan Couder for all the valuable support in these years. I
really appreciate all the motivational speeches and the time you took to discuss my research
quests together. I also had the exceptional honour of accompanying you on multiple, delightful
trips into the tangles of Belgium’s public transport network. And thank you for introducing me in
the highly esteemed societies of the Cake and Coffee Conference. Erik, your ever philosophical
contributions to these symposia were without exception a welcome source of mental
dispersion.
I have had the honour and pleasure to spend the majority of these past years with my
colleagues from the fifth floor. Thank you all for the interesting discussions during the coffee
breaks and beyond. I am leaving my desk in B.503, but the good memories I take with me are
iv
uncountable. Johan S, thank you for sharing some of your interesting music explorations with
me. But also for being an honest listener and someone who is always there for a good talk.
Trijntje, for all the interesting discussions we had, on all imaginable topics, and for broadening
my cultural knowledge on a regular basis. Johan B, for bringing a practical schwung to the fifth
floor. Kenneth, always in for a cool story during the coffee break, thanks for being an
appreciative work leader. David, for introducing me to the fascinating world of data mining, but
also for all the fun coffee break stories.
Also thanks to all current and former fellow PhD students on the fifth floor: Kevin, Floor, Tyché,
Sofie, Stiene, Ellen, Luca, Jochen, Daniel, Annelies, Christine, Renata, Enric, Iolanda, Sandy, Julie,
Christof, Babiche, Jasmine, Florian, Nicholas, Marco, Dorien, Jaco, Alexander, Jelle, Marija, Jellis.
And to Heidi and Adriaan, for being the wise youngsters. Kevin, after four and a half years
together on 20 m² our ways have departed. I think I can state honestly that our bond after these
years has become closer than we both could ever have anticipated. We had so many interesting
discussions on numerous topics and I think we both broadened our perspective and learned a
lot from each other. I want to thank you truly for being such an involved officemate. You always
noticed when my stress levels were peaking or something else was going on and your
motivational speech is one of the best.
Floor, thank you for brightening our side of the hallway with your sunshine and never-ending
enthusiasm. Tyché, for ensuring our side of the hallway did not become too light-footed. Julie,
thank you for being always so enthusiastic and for your honest support in difficult times.
Annelies, for being the very best personification of girl-geek-chic, but also for all the fun
moments I enjoyed with you and for your creative contributions to the hallway’s gossip and
fashion advice. Christine, for sharing a profound passion for food and, even better, for sharing
the results of your cooking activities with us. Christof, for enlightening the PhD days for all of us
with cultural endeavours and for showing us the Gaston-side of you. Daniel, thank you for all
the supportive speeches in the final weeks, but especially for the uncountable good talks we
had and all the very fun moments during and after office hours. Renata, for your good vibe and
creative, critical perspective in interesting discussions. Jelle, thank you for introducing me into
the fascinating world of board games (didn’t expect I would ever play a game related to
Battlestar Galactica…), for all the fun moments and for your honest involvement in multiple
discussions and projects. Also eternal thanks for the arrangement with my final-year funding,
this has been a true PhD-saver action. Florian, for our entertaining Deutsch-Nederlands
practices. Jaco, thank you for the very most entertaining e-mail conversations. Thanks to you, I
explored the World Wide Web for the best funny cat compilations for the very first time.
Finally, thank you An for bringing structure when I was in chaos and Esmeralda, for being an
involved mater familias in our PhD track.
v
Verder is dit het geschikte moment om mijn ouders te bedanken. Mama, dankjewel om er altijd
te zijn met je vrolijke insteek. Papa, dankjewel om er altijd te zijn met je warme inborst. Jullie
hebben me allebei op jullie eigen manier gesteund om mijn eigen ding te doen en daar ben ik
jullie dankbaar voor. Gon, dankjewel om er altijd te zijn met gezelligheid en Spaanse
vakantiesfeer. Femke, dankjewel om er altijd te zijn met je zusterlijk en creatief enthousiasme.
Ben, thanks for your insightful advice on key moments and for leading by example on how to
follow different dreams.
Dankjewel aan mijn lieve vriendinnen Chloë, Jelke, Francine, Marjolein en Jessie. Onze heerlijke
relaxweekends waren er meer dan eens op het moment dat ik ze wel kon gebruiken. Maar
vooral voor onze goede en eerlijke babbels die me altijd weer hielpen om de zaken in
perspectief te plaatsen. Jullie doen me elke keer waarderen wat belangrijk is in het leven.
En Jessie, proficiat!!
Dankjewel aan mijn lieve vriend(inn)en Karen, Marleen, Aurélie, Julie, Isabelle, Karolien, Sophie,
Lieselotte, Christijn, Jan, Katrin en Laurie. Jullie oprechte aanmoedigingen tijdens al mijn
universiteitsavonturen zijn altijd een echte steun geweest. Bedankt voor al de lieve
aanmoedigingen tijdens de laatste weken. En om mijn dagen te verrijken met al onze goede
babbels, uitstapjes, en fijne momenten samen. Ik kijk er oprecht naar uit om weer meer tijd met
jullie door te brengen.
Dankjewel Kim om zo’n goede vriendin te zijn, die me aanzet om mijn creativiteit de vrije loop te
laten en me geregeld een eerlijke spiegel voorhoudt. En Thijs, om zo vaak zo lekker te koken
wanneer ik weer eens gezellig de voeten onder tafel kwam steken bij jullie.
Heleen, dankjewel voor al de fijne lunches samen. Jij maakte van elke week een betere week
met je klare kijk op de dingen. Ik ben blij dat deze periode ons dichter bij elkaar heeft gebracht.
Imke, dankjewel om al vele jaren mijn vriendin te zijn en voor al de zotte dingen die we samen
hebben uitgestoken. Maar ook om zo een sterke steun te zijn tijdens mijn studiejaren. Fré,
dankjewel voor al de eet- en kookinspiratie en Katrien, om altijd zo enthousiast jezelf te zijn.
Dankjewel Ludo en mijn andere partners-in-crime van de academie, om mij elke week een
enorme uitlaatklep te geven en om van ont-plooi-ing een levenswijze te maken.
Het beste hou ik altijd graag als laatst. Dankjewel mijn lieve Jente om er te zijn, samen, in
chaotische en in groene dagen. De beste dagen komen er nog aan.
Sanne Lemmens
Antwerp, March 2017
vii
Nederlandse samenvatting
Het uitgangspunt van dit proefschrift zit vervat in de titel: technologische innovatie in de
energiesector. Technologische innnovatie omvat veranderingen, klein of groot, die invloed
hebben op onze dagelijkse gewoontes en de manier waarop de maatschappij is ingericht.
Energie is de hoeksteen van onze huidige maatschappij en noodzakelijk voor economische
ontwikkeling. In het licht van hedendaagse bedenkingen m.b.t. de duurzaamheid van
energieproductie en –gebruik, vertrekt deze thesis vanuit de interactie tussen technologische
innovatie en deze energie-uitdagingen. De aanpak in dit werk is technologie-specifiek, met focus
op de organische Rankinecyclus (ORC). De ORC is conceptueel gelijkaardig aan de conventionele
stoomcyclus, maar gebruikt alternatieve werkingsmedia in plaats van water. Door het gebruik
van deze alternatieve fluida kan de ORC gebruikt worden om elektriciteit op te wekken uit
energie-bronnen met een lagere temperatuur dan mogelijk met de klassieke stoomcyclus. ORC
technologie wordt dan ook meestal gelinkt aan hernieuwbare energiebronnen zoals biomassa,
geothermie of zonne-energie, maar ook aan industriële restwarmte. Met andere woorden, de
technologie heeft potentieel om bij te dragen aan wereldwijde doelstellingen op het vlak van
hernieuwbare energie en energie efficiëntie. De algemene opzet van deze thesis bestaat erin
om inzicht te krijgen in het potentieel van ORC technologie, waarbij het potentieel betrekking
heeft op zowel het fysieke, het technische als het economische aspect.
De verspreiding van een opkomende technologie wordt beïnvloed door verschillende factoren,
waarvan de technische haalbaarheid er slechts één is. Andere factoren zijn de kosten van de
technologie, de interactie met andere technologieën en tussen de verschillende betrokken
actoren of de impact van beleidsmakers op onderzoek en ontwikkeling en op de marktformatie.
In de academische literatuur bestaan er ten minste twee kaderwerken die een structuur bieden
om het ontwikkelingstraject van een opkomende technologie te bestuderen: strategic niche
management en technological innovation systems. Elk van deze kaderwerken bestudeert de
essentiële drijfveren en processen die een nieuwe technologie, of zelfs een nieuw technologisch
systeem, sturen. The technische aspecten van ORC technologie worden intensief onderzocht,
maar de andere perspectieven zijn tot op heden grotendeels onontgonnen terrein. Een
volledige analyse van deze dimensies valt dan ook buiten het bereik van deze thesis. Dit
proefschrift vangt aan met het bestuderen van de basisinzichten m.b.t. ORC technologie en
verbreedt de scope van de analyse gaandeweg bij het bereiken van nieuwe inzichten. Centraal
in dit onderzoek is de technologie zelf. De focus verbreedt dan tot de basisdimensies van de
kosten, de invloed van de markt en de economische context, van het bestaande beleidskader
en, ten slotte, de diffusie van de technologie en de impact ervan.
viii
De eerste onderzoeksvraag in deze thesis luidt: “Wat zijn de verdiensten van ORC technologie?
En wat zijn de state-of-the-art inzichten omtrent de economische aspecten?”. Deze essentiële
vragen worden beantwoord in hoofdstuk 2 van de thesis. De verdiensten van ORC technologie
worden onder het algemeen verstaan als de mogelijkheid om energiebronnen met een kleinere
capaciteit en een lagere temperatuur te verwerken dan doorgaans mogelijk met een
stoomcyclus. Hierdoor wordt de ORC doorgaans geassocieerd met duurzame energiebronnen
zoals biomassa, geothermie en zonne-energie, of met industriële restwarmtestromen. Ondanks
het gepercipieerde potentieel van de technologie, bestaat er tot op heden geen verregaand
overzicht van de economische aspecten van ORC technologie. De bijdrage van hoofdstuk 2
bestaat dan ook uit een grondige studie van de literatuur. De belangrijkste bevindingen van
deze studie zijn de volgende. Ten eerste, de praktische context en implementatie van ORC
systemen varieert sterk naargelang het type warmtebron dat wordt gebruikt. De opdeling van
applicaties volgens de vier meest voorkomende warmtebronnen (restwarmte, biomassa,
geothermie en zonne-energie) is dus zinvol. Ten tweede, het technische perspectief domineert
de literatuur en de studies die de economische aspecten mee opnemen, leggen de nadruk
hierbij voornamelijk op de kapitaalkosten van het bestudeerde systeem. De gepubliceerde
kosten verschillen sterk tussen en binnen de vier categorieën. Dit is te wijten aan het feit dat er
zeer weinig informatie beschikbaar is over de werkelijke kosten van geïnstalleerde ORC
systemen. Een groot deel van de literatuur bespreekt kapitaalkosten die werden geschat,
gebaseerd op nieuw ontworpen ORC designs. Er bestaan verschillende technieken om de kosten
van industriële processen te schatten, gaande van schaalmethoden tot correlaties voor elk van
de specifieke componenten. De resultaten van deze schattingen zijn zeer divers, maar bijna
geen enkele van de studies bespreekt in welke mate deze schattingen representatief zijn voor
de werkelijke ORC kosten. Ten slotte, verschillende technologieën voor elektriciteitsproductie
worden doorgaans onderling te vergeleken aan de hand van de levelized cost of electricity
(LCOE). De LCOE geeft aan tegen welke kosten het systeem elektriciteit kan produceren. Er zijn
slechts enkele onderzoeken die de LCOE voor ORC technologie bestuderen, opnieuw met sterk
uiteenlopende resultaten.
Om tegemoet te komen aan de vragen omtrent het gebruik van geschatte kapitaalkosten,
onderzoekt hoofdstuk 3 de principes van cost engineering. Het doel van dit hoofdstuk is om
inzicht te krijgen in de methodes die zo wijdverspreid worden gebruikt om ORC kosten te
berekenen, maar zelf zelden kritisch worden geëvalueerd. De leidende vraag in dit hoofdstuk
luidt: “In welke mate kunnen cost engineering technieken gebruikt worden om inzicht te
verkrijgen in de kosten van ORC systemen?”. Doorheen de jaren zijn er verschillende methoden
ontwikkeld om de kosten van industriële projecten ex ante in te schatten. Eenvoudige
technieken vragen minder inspanning, maar leiden tot resultaten met een lagere
nauwkeurigheid. Hogere nauwkeurigheden zijn mogelijk met definitive en detailed schattingen,
maar deze vragen een verregaand overzicht van het systeemontwerp. Kostenschattingen in het
ix
kader van onderzoek hebben over het algemeen nauwkeurigheden in de grootteorde van order-
of-magnitude (-50 % to +100 %), study (-30 % to +50 %) of preliminary (-20 % to +30 %)
schattingen. Dit hoofdstuk onderzoekt verschillende schattingsmethoden door ze toe te passen
om de kosten van een case study te schatten. De resultaten variëren sterk naargelang de
methode die werd gebruikt, maar ook in vergelijking met de werkelijke kosten. Een deel van
deze variatie is te verklaren door de inherente beperkingen aan de nauwkeurigheid van de
schattingsmethoden, maar ook het feit dat de gebruikte schaalfactoren, correlaties en andere
factoren werden vastgesteld voor industriële processen in het algemeen en niet voor ORC
technologie specifiek. Eén manier om deze nauwkeurigheden te verhogen ligt dus in het
uitbreiden van de inzichten m.b.t. de werkelijke kosten van ORC technologie, zodat technologie-
specifieke factoren kunnen worden vastgesteld.
De diffusie van een opkomende technologie wordt niet enkel beïnvloed door deze techno-
economische basis, maar ook door externe factoren zoals de huidige marktcondities of de mate
van beleidsinterventie. Hoofdstuk 4 komt tegemoet aan de behoefte naar inzichten in de
economische aspecten van reële ORC systemen en verbreedt de scope van het onderzoek tot de
kwestie van financiële analyse, de impact van de verschillende parameters en de invloed van
overheidsbeleid. De centrale vraag in dit hoofdstuk luidt: “Welke factoren beïnvloeden de
financiële haalbaarheid van een investering in ORC technologie?”. De vraag wordt beantwoord
aan de hand van een diepgaande case study, betreffende een ORC geïnstalleerd in Vlaanderen
voor restwarmterecuperatie uit een industriële oven. De belangrijkste resultaten van dit
hoofdstuk zijn de volgende. Ten eerste, het investeringsproject werd positief beoordeeld met
een interne rendementsgraad van 12.6 %. Al nuanceert de manager verantwoordelijk voor de
ORC deze theoretische resultaten, want door wederkerende problemen met de productie van
de ORC bedraagt het werkelijke rendement ongeveer 8 %. Deze positieve evaluatie is het
resultaat van de voordelige impact van enkele overheidssteunmaatregelen. In een scenario
waar de subsidies niet mee in rekening worden gebracht, maar winstbelasting wel aanwezig is,
daalt het rendement nl. tot 5.4 %. Merk op dat hoewel dit resultaat aan de lage kant van de
verwachtingen ligt in sommige industriële sectoren, deze investering toch de moeite is gezien
de huidige lage interestvoeten. Ten slotte, de sensitiviteitsanalyse toont aan dat de
belangrijkste impact op de financiële resultaten voortkomt uit het behalen van voldoende
draaiuren en vanwege de elektriciteitsprijs op de markt.
Ten slotte, een centraal proces in de ontwikkeling van opkomende technologieën is het opdoen
van ervaring met deze technologie, om bij te kunnen leren over de praktische aanpak en
opbouw, maar ook de werking ervan in de praktijk. Om de diffusie van ORC technologie en de
bijbehorende leereffecten te bestuderen, verbreedt hoofdstuk 5 de scope van de studie rond de
vraag: “Hoe ziet het diffusieproces van ORC technologie er uit en hoe beïnvloedt het gebruik van
de technologie de kosten?”. De bijdrage van dit hoofdstuk bestaat uit drie delen. Ten eerste, de
x
kennis omtrent de kosten van reële ORC systemen wordt drastisch uitgebreid door een
database samen te stellen met meer dan 100 relevante cases. Ten tweede, de diffusie van de
technologie wordt bestudeerd aan de hand van een tweede database, waarin bij benadering 95
% van al de ORC systemen wereldwijd werden opgenomen. Bij weten van de auteur is een
dergelijke hoeveelheid data nog nooit verzameld voor ORC technologie. Ten derde, de
combinatie van deze twee databases laat toe om inzichten te verkrijgen in de evolutie van de
kosten van ORC technologie. De belangrijkste bevindingen van dit hoofdstuk zijn de volgende.
De verspreiding van ORC technologie nam een sterke vlucht na de start van het nieuwe
millennium, een gelijkaardig patroon als genoteerd voor verschillende hernieuwbare energie-
technologieën. Deze interesse voor duurzame technologieën kan gerelateerd worden aan de
sterke stijging van de elektriciteitsprijzen en een toegenomen bewustwording van energie-
kwesties en de weergave hiervan in energiebeleid, voornamelijk in Europa. Ten tweede, ORC
systemen bestaan in elke regio van de wereld, maar het grootste deel staat in Europa en
biomassa is de meest gebruikte energiebron. De ORC producentenmarkt is redelijk
oligopolistisch: twee bedrijven, Ormat en Turboden, domineren de markt aangaande het totaal
geïnstalleerd vermogen en het aantal referenties. De intensieve onderzoeks- en ontwikkelings-
activiteiten die voorafgaan aan het design van een werkend prototype en de hoge kosten om de
eerste unit te produceren, maken het moeilijk om als producent de markt te betreden. Toch
neemt het aantal producenten gestaag toe en zijn er nieuwe producenten die een toenemend
marktaandeel voor zich nemen. Ten slotte wordt er onderzocht in welke mate de kosten van
ORC systemen onderhevig zijn aan schaal- of leereffecten. De schaaleffecten voor ORC
technologie zijn significant, en het sterkst gemeten voor de ORC’s die werken op biomassa. De
invloed van toenemende ervaring op de kosten van de technologie is ambigu.
xi
Contents
1. Introductory chapter: technological innovation and challenges for energy sustainability ............. 1
1.1. Introduction .................................................................................................................................................. 3
1.2. Technological change and system transitions .............................................................................................. 4
1.2.1. A taxonomy of technology and innovation .......................................................................................... 4
1.2.2. Frameworks in transition research ...................................................................................................... 6
1.3. Energy sustainability and the role of technological change ....................................................................... 10
1.3.1. Current challenges for energy sustainability ..................................................................................... 10
1.3.2. Technology focus: the organic Rankine cycle .................................................................................... 13
1.4. The potential of ORC technology: framework and methodology............................................................... 16
1.4.1. Technology and economics ................................................................................................................ 18
The organic Rankine cycle: technical characteristics, applications and economics (Chapter 2) ......................... 18
Cost engineering techniques and their application for ORC technology (Chapter 3) .......................................... 20
1.4.2. Markets and economic context & policy and institutions ................................................................. 21
Case study of an organic Rankine cycle applied for excess heat recovery (Chapter 4)........................................ 21
1.4.3. Technology and innovation ................................................................................................................ 21
The dynamics of ORC technology: technological innovation, economies of scale and learning by doing (Chapter
5) ......................................................................................................................................................................... 22
Bibliography ............................................................................................................................................................. 23
2. The organic Rankine cycle: technical characteristics, applications and economics ..................... 25
2.1. Introduction ................................................................................................................................................ 27
2.2. The organic Rankine cycle .......................................................................................................................... 27
2.2.1. Working fluids .................................................................................................................................... 28
2.2.2. Cycle components and architectures ................................................................................................. 29
2.2.3. Areas of application ........................................................................................................................... 30
Energy efficiency improvement: excess heat recovery ........................................................................................ 30
Renewable power generation: biomass energy .................................................................................................. 33
Renewable power generation: geothermal energy ............................................................................................. 33
Renewable power generation: solar energy ........................................................................................................ 34
2.3. The costs of ORC technology: a literature review ...................................................................................... 34
2.3.1. Methodology...................................................................................................................................... 35
2.3.2. The costs of ORC technology: heat recovery applications ................................................................. 35
xii
Heat recovery ORC applications: capital costs of real systems ........................................................................... 36
Heat recovery ORC applications: representative manufacturer’s quotes ............................................................ 36
Heat recovery ORC applications: estimated capital costs ................................................................................... 38
Heat recovery ORC applications: assumed capital costs ..................................................................................... 43
Heat recovery ORC applications: an overview of capital costs ............................................................................ 43
2.3.3. The costs of ORC technology: biomass applications .......................................................................... 45
Biomass ORC applications: capital costs of real systems .................................................................................... 46
Biomass ORC applications: representative manufacturer’s quotes..................................................................... 47
Biomass ORC applications: estimated capital costs ............................................................................................ 47
Biomass ORC applications: assumed capital costs .............................................................................................. 49
Biomass ORC applications: an overview of capital costs ..................................................................................... 50
2.3.4. The costs of ORC technology: geothermal applications .................................................................... 51
Geothermal ORC applications: capital costs of real systems ............................................................................... 52
Geothermal ORC applications: representative manufacturer’s quotes ............................................................... 52
Geothermal ORC applications: estimated capital costs ...................................................................................... 53
Geothermal ORC applications: assumed capital costs ........................................................................................ 55
Geothermal ORC applications: an overview of capital costs ............................................................................... 56
2.3.5. The costs of ORC technology: solar applications ............................................................................... 56
Solar ORC applications: capital costs of real systems .......................................................................................... 57
Solar ORC applications: representative manufacturer’s quotes .......................................................................... 57
Solar ORC applications: estimated capital costs ................................................................................................. 58
Solar ORC applications: assumed capital costs ................................................................................................... 59
Solar ORC applications: an overview of capital costs .......................................................................................... 60
2.4. Discussion of the results ............................................................................................................................. 61
2.5. Chapter conclusions ................................................................................................................................... 62
Bibliography ............................................................................................................................................................. 64
3. Cost engineering techniques and their application for ORC technology ....................................... 75
3.1. Introduction ................................................................................................................................................ 77
3.2. Cost estimation for industrial plants .......................................................................................................... 77
3.3. Comparing the estimated and actual costs of a heat recovery ORC system .............................................. 81
3.3.1. Case study: ORC for industrial heat recovery .................................................................................... 81
3.3.2. Using cost data of other systems: the capacity exponent ratio method ........................................... 82
3.3.3. Using technical parameters of the system: factorial estimation techniques .................................... 84
Estimating the purchased equipment costs......................................................................................................... 84
xiii
Estimating the total investment costs: multiplication factors ............................................................................. 86
Estimating the total investment costs: percentages of delivered equipment costs ............................................ 88
3.4. Discussion of the results ............................................................................................................................. 90
3.5. Chapter conclusions ................................................................................................................................... 92
Bibliography ............................................................................................................................................................. 94
4. Case study of an organic Rankine cycle applied for excess heat recovery .................................... 97
4.1. Introduction ................................................................................................................................................ 99
4.2. An ORC excess heat recovery project in Belgium ..................................................................................... 100
4.2.1. Technical setup ................................................................................................................................ 101
4.2.2. Investment and annual costs ........................................................................................................... 101
4.2.3. Public policy ..................................................................................................................................... 103
4.3. Financial appraisal of the excess heat recovery project ........................................................................... 105
4.3.1. Metrics for financial appraisal ......................................................................................................... 105
4.3.2. Assumptions for the financial appraisal ........................................................................................... 107
4.3.3. Results of the financial appraisal ..................................................................................................... 110
4.4. The impact of public policy ....................................................................................................................... 112
4.4.1. The impact of public policy: changing the extent of government intervention .............................. 112
4.4.2. The impact of public policy: changing the impact of corporate income taxation ........................... 114
4.5. Parameter sensitivity analysis .................................................................................................................. 115
4.5.1. Sensitivity of the results: ceteris paribus analysis ........................................................................... 116
Sensitivity analysis: ORC net power and capital investment ............................................................................. 116
Sensitivity analysis: annual O&M costs ............................................................................................................. 117
Sensitivity analysis: load hours .......................................................................................................................... 117
Sensitivity analysis: operating years .................................................................................................................. 117
Sensitivity analysis: electricity prices ................................................................................................................. 119
Sensitivity analysis: inflation ............................................................................................................................. 119
Sensitivity analysis: discount rate ...................................................................................................................... 120
Sensitivity analysis: summary ............................................................................................................................ 121
4.5.2. Sensitivity of the results: Monte Carlo simulation ........................................................................... 122
4.6. Discussion of the results ........................................................................................................................... 125
4.7. Chapter conclusions ................................................................................................................................. 129
Bibliography ........................................................................................................................................................... 131
5. The dynamics of ORC technology: technological innovation, economies of scale and learning by
doing ............................................................................................................................................. 133
xiv
5.1. Introduction .............................................................................................................................................. 135
5.2. The innovation path of ORC technology ................................................................................................... 136
5.2.1. The technology life-cycle ................................................................................................................. 136
5.2.2. The technology life-cycle: use and interpretation ........................................................................... 138
5.2.3. The innovation path of ORC technology: invention, research and development ............................ 140
Academic publications ....................................................................................................................................... 140
5.2.4. The innovation path of ORC technology: technology diffusion and market formation ................... 142
Worldwide installed ORC capacity: historical trend .......................................................................................... 142
Worldwide installed ORC capacity: distribution according to heat source........................................................ 144
Worldwide installed ORC capacity: regional distribution .................................................................................. 146
Worldwide installed ORC capacity: manufacturers’ market share .................................................................... 148
5.3. Cost development of ORC technology ...................................................................................................... 150
5.3.1. Data collection ................................................................................................................................. 150
Methodology ..................................................................................................................................................... 151
Data prospection ............................................................................................................................................... 151
5.3.2. Cost development of ORC technology: economies of scale ............................................................ 154
The economies of scale for ORC technology ...................................................................................................... 155
5.3.3. Cost development of ORC technology: the impact of experience ................................................... 165
Learning by doing and the experience curve ..................................................................................................... 165
Experience curves: types, variable selection and interpretation ....................................................................... 168
Experience curves for ORC-generated electricity ............................................................................................... 170
5.4. Discussion of the results ........................................................................................................................... 177
5.4.1. ORC technology diffusion and the distribution of market shares .................................................... 177
Factors influencing the diffusion of ORC technology ......................................................................................... 179
5.4.2. Scale economies for ORC technology .............................................................................................. 183
5.4.3. On learning-by-doing in the ORC market ......................................................................................... 186
Factors influencing the underlying cost dynamics ............................................................................................. 189
Remarks on the experience curves for ORC-generated electricity ..................................................................... 192
5.5. Chapter conclusions ................................................................................................................................. 193
Bibliography ........................................................................................................................................................... 196
6. Concluding chapter: synthesis, implications and recommendations ........................................ 199
6.1. Recapitulation of the key results .............................................................................................................. 201
6.2. Implications and recommendations ......................................................................................................... 204
Appendix A. Summary of the literature review on the economics of ORC technology ................... 207
xv
Appendix B. Numerical results for the financial appraisal of the case study ................................... 229
Appendix C. Survey from ORC end-users ....................................................................................... 233
xvii
Tables
Table 1.1. Challenges and Work Packages in the ORCNext-project. ............................................................................ 15
Table 1.2. Three criteria defined to structure the literature on ORC economics. ....................................................... 19
Table 2.1. Characteristics of suitable ORC working fluids. ........................................................................................... 29
Table 2.2. Amount of heat recovery ORC publications according to system scope and economic scope. .................. 36
Table 2.3. Amount of biomass ORC publications according to system and economic scope. ..................................... 46
Table 2.4. Amount of geothermal ORC publications according to system and economic scope. ................................ 52
Table 2.5. Amount of solar ORC publications according to system and economic scope............................................ 57
Table 3.1. Classification of capital cost estimates for process industry projects. ........................................................ 79
Table 3.2. Cost estimation using the capacity exponent ratio method. ...................................................................... 83
Table 3.3. Coefficients and correlations for estimation of the purchased equipment costs. ...................................... 85
Table 3.4. Factorial estimation of the purchased equipment costs. ............................................................................ 85
Table 3.5. Total investment costs estimation with the Module Costing Technique. ................................................... 88
Table 3.6. Total investment costs estimation using percentages of delivered equipment costs. ............................... 89
Table 4.1. Policies for heat recovery at the European level, from the Belgian federal government and the Flanders
Region. ...................................................................................................................................................... 105
Table 4.2. Project parameters for assessment of the excess heat recovery ORC project.......................................... 109
Table 4.3. Government interventions of the excess heat recovery ORC project. ...................................................... 109
Table 4.4. Probability distribution assumptions for the input parameters for the Monte Carlo simulation. ............ 123
Table 4.5. Parameter contributions to the variance in NPV, established by the Monte Carlo simulation. ............... 124
Table 5.1. Economies of scale for ORC projects and modules: complete dataset. .................................................... 157
Table 5.2. Economies of scale for ORC projects and modules: heat recovery subset. .............................................. 160
Table 5.3. Economies of scale for ORC projects and modules: biomass subset. ....................................................... 161
Table 5.4. Economies of scale for ORC projects and modules: geothermal subset. .................................................. 162
Table 5.5. Economies of scale for ORC projects and modules: Turboden subset. ..................................................... 164
Table 5.6. Experience curve fitting for ORC-generated electricity: complete dataset. .............................................. 171
Table 5.7. Experience curve fitting for ORC-generated electricity: heat recovery subsample. ................................. 172
Table 5.8. Experience curve fitting for ORC-generated electricity: biomass subsample. .......................................... 173
Table 5.9. Experience curve fitting for ORC-generated electricity: geothermal subsample. ..................................... 175
Table 5.10. Experience curve fitting for ORC-generated electricity: Turboden subsample. ...................................... 176
Table 5.11. Suitable temperature range of the heat source, capacity range of installed systems and percentage of
applications in each heat source category for each manufacturer in the database. ................................ 179
Table 5.12. Overview of economies of scale of ORC projects and modules, estimated for different categories. ..... 184
Table 5.13. Progress rates for ORC projects and modules, estimated for different categories................................. 186
Table 5.14. Progress rates for the adjusted ORC projects and modules, for different categories............................. 189
xix
Figures
Figure 1.1. The multi-level perspective on socio-technical transitions. ......................................................................... 8
Figure 1.2. The transition management cycle. ............................................................................................................... 9
Figure 1.3. World total primary energy supply (a) in 2014, according to energy source and (b) from 1970 to 2014. 11
Figure 1.4. (a) Component layout and (b) T-s diagram of the basic subcritical ORC. ................................................... 15
Figure 1.5. Schematic overview of the research framework. ...................................................................................... 17
Figure 1.6. Terminology for delineation for ORC system scopes: ORC modules, projects, integration costs and
project costs. ............................................................................................................................................... 20
Figure 2.1. Investment costs for heat recovery ORC systems, as published in the literature...................................... 45
Figure 2.2. Investment costs for biomass ORC systems, as published in the literature............................................... 51
Figure 2.3. Investment costs for geothermal ORC systems, as published in the literature. ........................................ 56
Figure 2.4. Investment costs for solar ORC systems, as published in the literature. ................................................... 61
Figure 3.1. Diagram of the investment costs of the ORC case study. .......................................................................... 81
Figure 3.2. (a) Real and (b) estimated purchased equipment costs of the case study. ............................................... 86
Figure 3.3. Summary of the real and the estimated investment costs for the case study. ......................................... 91
Figure 4.1. Diagram of the case study’s total capital investment. ............................................................................. 102
Figure 4.2. Cumulative non-discounted cash flow of the ORC project for all discount rate scenarios. ..................... 111
Figure 4.3. NPV, IRR, PP, LCOE and policy LCOE of the ORC project for high, medium and low discount rates. ....... 112
Figure 4.4. Cumulative cash flow for five different policy scenarios. ........................................................................ 113
Figure 4.5. Comparison of the NPV for three different discount rates and five different policy scenarios. .............. 114
Figure 4.6. Cumulative cash flow for four different policy scenarios. ....................................................................... 115
Figure 4.7. Comparison of the NPV for three different discount rates and four different policy scenarios. ............. 115
Figure 4.8. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the ORC module costs for different
discount rates. ........................................................................................................................................... 116
Figure 4.9. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the annual O&M costs for different
discount rates. ........................................................................................................................................... 117
Figure 4.10. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the load hours for different discount
rates. ......................................................................................................................................................... 118
Figure 4.11. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the operating hours for different
discount rates. ........................................................................................................................................... 118
Figure 4.12. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the electricity price for different
discount rates. ........................................................................................................................................... 119
Figure 4.13. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the general inflation rate for
different discount rates. ............................................................................................................................ 120
Figure 4.14. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the electricity price inflation rate for
different discount rates. ............................................................................................................................ 120
Figure 4.15. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the real discount rate. ................ 121
Figure 4.16. Sensitivity of the NPV to changes in various project parameters, for a real discount rate of 6 %. ........ 122
Figure 4.17. Sensitivity of the NPV to changes in various project parameters, for a real discount rate of 6 %. ........ 122
Figure 4.18. Frequency distribution of the NPV, established by the Monte Carlo simulation................................... 124
Figure 4.19. NPV, IRR, PP, LCOE and policy LCOE of the ORC project for different nominal discount rates.............. 126
Figure 4.20. LCOE for utility-scale renewable power technologies. .......................................................................... 127
xx
Figure 5.1. (a) Technology adoption and (b) cumulative technology adoption, over time........................................ 137
Figure 5.2. Annual count of ORC-related publications in the Web of Science. .......................................................... 141
Figure 5.3. Annual citation count of ORC-related publications in the Web of Science. ............................................. 141
Figure 5.4. Cumulative annual growth of ORC references according to thermal source. .......................................... 143
Figure 5.5. Cumulative annual growth of installed capacity of ORC references according to thermal source. ......... 144
Figure 5.6. (a) Number of ORC references per thermal source; (b) total installed capacity of ORC references per
thermal source. ......................................................................................................................................... 145
Figure 5.7. Number of ORC references for different size ranges, per thermal source. ............................................. 145
Figure 5.8. Total installed capacity of ORC references for different size ranges. ...................................................... 146
Figure 5.9. (a) Number of ORC references per region; (b) total installed capacity of ORC references per region. ... 147
Figure 5.10. Number of ORC references per region, according to thermal source. .................................................. 147
Figure 5.11. Total installed capacity of ORC references per region, according to thermal source. ........................... 148
Figure 5.12. (a) Number of ORC references per manufacturer; (b) total installed capacity of ORC references per
manufacturer. ........................................................................................................................................... 149
Figure 5.13. Number of ORC references per manufacturer, according to thermal source........................................ 149
Figure 5.14. Cumulative annual growth of ORC references according to manufacturer. .......................................... 150
Figure 5.15. ORC cost database prospection: distribution of the references according to thermal source. ............. 153
Figure 5.16. ORC cost database prospection: references according to start year. .................................................... 153
Figure 5.17. ORC cost database prospection: references according to installed capacity, in ranges. ....................... 154
Figure 5.18. ORC cost database prospection: references according to specific investment costs, in ranges. ........... 154
Figure 5.19. Specific investment costs of ORC systems: complete dataset. .............................................................. 156
Figure 5.20. Cumulative frequency of ORC specific investment costs: complete dataset. ........................................ 156
Figure 5.21. Specific investment costs as a function of installed capacity: complete dataset per heat source. ....... 158
Figure 5.22. Specific investment costs as a function of installed capacity: heat recovery subset. ............................ 159
Figure 5.23. Cumulative frequency of ORC specific investment costs: heat recovery subset. .................................. 159
Figure 5.24. Specific investment costs as a function of installed capacity: biomass subset. ..................................... 160
Figure 5.25. Cumulative frequency of ORC specific investment costs: biomass subset. ........................................... 161
Figure 5.26. Specific investment costs as a function of installed capacity: geothermal subset................................. 162
Figure 5.27. Economies of scale for ORC technology: cumulative frequency curves for the geothermal subset. .... 162
Figure 5.28. Specific investment costs as a function of installed capacity: Turboden subset. .................................. 163
Figure 5.29. Economies of scale for ORC technology: cumulative frequency curves for the Turboden subset. ....... 164
Figure 5.30. The one-factor learning curve on (a) a normal scale and (b) a log-log scale.......................................... 166
Figure 5.31. Experience curves for ORC-generated electricity: complete dataset. ................................................... 171
Figure 5.32. Experience curves for ORC-generated electricity: heat recovery subsample. ....................................... 172
Figure 5.33. Experience curves for ORC-generated electricity: biomass subsample. ................................................ 173
Figure 5.34. Experience curves for ORC-generated electricity: geothermal subsample. ........................................... 174
Figure 5.35. Experience curves for ORC-generated electricity: Turboden subsample. .............................................. 176
Figure 5.36. Electricity prices in European countries for industrial consumers. ........................................................ 182
Figure 5.37. Average capacity [MW] of ORC systems installed.................................................................................. 185
Figure 5.38. Average capacity [MW] ORC systems installed after 2000, for all systems and for Turboden. ............. 185
Figure 5.39. Relation between the price and costs of a new product. ...................................................................... 190
Figure 5.40. Experience curves for ORC-generated electricity: Turboden subsample compared to the rest of the
data. .......................................................................................................................................................... 190
xxi
Abbreviations
AFUDC allowance for funds used during construction Bio Biomass energy CEPCI Chemical Engineering Plant Cost Index CER capacity exponent ratio CHP combined heat and power CRS central receiver system CSP concentrating solar power DFCI direct fixed-capital investment EC European Commission ECU European Currency Unit EEG Erneuerbare-Energien-Gesetz (Renewable Energy Act Germany) EGS Enhanced/Engineered Geothermal Systems EP-PLUS ecology premium in Flanders ETC evacuated-tube collector ETS Emissions Trading Scheme EU European Union FCI fixed-capital investment Geo geothermal energy GHG greenhouse gas GWP global warming potential HR heat recovery HVAC heating, ventilation and air conditioning IFCI indirect Fixed-Capital Investments IPCC Intergovernmental Panel on Climate Change IRR internal rate of return ISO International Organization for Standardization LCOE levelized cost of electricity LFR linear Fresnel reflector LR learning rate M module MCT Module Costing Technique MED multi effect distillation MENA Middle East and Northern Africa MLP multi-level perspective on socio-technical transitions NPI net power output index NPV net present value ODP ozone depletion potential OFSC offsite costs O&M operation & maintenance
xxii
ONSC onsite costs ORC organic Rankine cycle P project PEC purchased-equipment cost PDEC percentages of delivered equipment costs PEX pressure exchanger PP payback period PR progress ratio PT parabolic trough PTC parabolic trough collector PV photovoltaic PWT Pelton wheel turbine RO reverse osmosis SIC specific investment costs SCORC subcritical organic Rankine cycle SNM strategic niche management Sol solar energy TCORC transcritical organic Rankine cycle TIS technological innovation systems TGC tradable green certificates TM transition management TPES Total Primary Energy Supply UA product of the heat transfer coefficient and the heat transfer area UNFCCC United Nations Framework Convention on Climate Change VAT Value Added Tax WACC weighted average cost of capital WP work package
xxiii
Nomenclature
Units
Gt giga tonnes Mtoe mega tonnes of oil equivalent J Joule kWe kilowatt electric kWth kilowatt thermal kWnet net power output kWgross gross power output / installed capacity
Monetary units
€ Euro £ Great British Pound NOK Norwegian Krone $ United States Dollar (unless stated otherwise)
Prefixes
P Peta 1015 T tera 1012 G giga 109 M mega 106 k kilo 103
Symbols
A the equipment attribute Aa the equipment cost attribute of the required component Ab the equipment cost attribute of the known component ca the purchase costs of the required component cb the purchase costs of the known component ci the costs in year i cj the costs in year j
ct the net cash flows in year t 𝐶0 the capital investment 𝐶𝐶𝑢𝑚 the cost per unit; 𝐶𝑒
0 the cost of the first produced unit;
xxiv
Cp0 the purchased equipment costs at ambient pressure and using carbon steel
CBM the bare module costs CTM the project costs (referred to as the total module costs in Turton et al. (2013)) CF,t the fuel costs in year t CS the investment support Cum the total cumulative production Dt the depreciation in year t Et the electricity production at time t FBM the bare module cost factor Ii the cost index for year i Ij the cost index for year j
It the investment deduction in year t; K1, K2 & K3 the coefficients determined for the type of equipment in the Module Costing
Technique LR the learning rate; m the experience parameter (the learning elasticity) n the exponent used to correlate the costs in the scaling method n the economic lifetime of the project in project appraisal NM the amount of ORC module references NP the amount of ORC project references NTOT the total amount of ORC references O&Mt the operation and maintenance costs in year t; PR the progress rate. r the discount rate S the salvage value at the end of the project lifetime t the point in time (year) T the applicable tax rate
1. Introductory chapter:
technological innovation and
challenges for energy sustainability
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
3
1.1. Introduction
The starting point of this work is encompassed in the title: technological innovation in the
energy sector. Technological innovation involves changes, small or large, that influence daily
habits and the way in which societies are organized. Technologies that, at the moment of their
invention, may not have been appreciated to their full extent can be at the origin of true shifts
in societal customs. Steam power, for instance, originally invented and applied in the form of a
steam engine for water pumping, is today the driver behind the large majority of electricity
generation worldwide. Energy is the cornerstone of today’s societies and key for economic
development. The way in which resources are extracted or transformed into usable forms of
energy and the conventions by which energy is used pervades society. Energy is at the nexus of
the human and the natural environment, relating to the drivers of societal interaction and
development as well as impacting natural cycles. Today, energy sectors worldwide are
confronted with questions about sustainability.
The interaction between technological innovation and energy sustainability lies at the core of
this thesis. This first chapter presents the rationale for my research and discusses the questions
of relevance. Section 1.2 introduces the concepts technology and innovation. It explores the
literature to get insight into the origins of research on technological innovation as well as the
contemporary frameworks conceptualized to structure technological innovation and
corresponding system changes. These frameworks offer guidance to structure the course of my
work. Section 1.3 covers the topic of energy sustainability. It stipulates what a sustainable
energy system would look like and summarizes current challenges in energy supply and use.
Attaining energy sustainability requires action on multiple fronts: from respecting the resilience
of the local and global environment to ensuring affordability and the independence and
accountability of regulatory institutions. Many researchers couple the insights from the study of
technological innovation to the challenges in the energy sector, pursuing an answer to the
question whether - and how - it is possible to induce a system transition towards enhanced
sustainability. There are two general approaches: either a more aggregate perspective on the
transition and its drivers or a more bottom-up perspective with focus on an innovation which is
perceived to have potential to contribute to the transition. This thesis takes the latter approach.
More specifically, the focus lies on the potential of one technology in particular: the organic
Rankine cycle. Section 1.3.2 introduces the working principle of this technology and discusses its
merits in light of the current quest for energy sustainability. The organic Rankine cycle’s
characteristics make it suitable to harness several renewable energy sources and to improve
industrial energy efficiency by putting unused excess thermal energy flows to work. Despite the
fact that the technology has proven its merits and technical viability in multiple applications,
several questions remain unsolved. For instance, there is basically no published information
CHAPTER ONE
4
about how often the technology is used today, about why end-users engage in projects with this
technology, or what its prospects for future deployment are. This thesis tries to answer these
and other questions, approaching the issue from an economic point of view.
The research focusses on the potential of organic Rankine cycle technology. There is special
attention for the economics, but the matter is approached from several angles. The framework
applied in this thesis is inspired by the approaches put forward by strategic niche management
and technological innovation systems studies, but there is no exhaustive overlap with either of
their structures. Section 1.4 presents which research questions are addressed in this study, sets
out the framework and scope that delineate it and discusses which methodologies are used.
1.2. Technological change and system transitions
The role of technology in shaping the course of economic and social development is substantial.
Human well-being has been altered through technological developments of all kinds. For
instance, the development and adaptation of the steam engine to more applicable layouts in
the late 18th century implied unprecedented opportunities for industrial development. Similarly,
the introduction of electricity for commercial uses marks the start of an entirely new way of
living, with applications such as light bulbs, electric motors and telegraphs. More recently,
societies are strongly influenced by technological innovations in the field of information and
communication. Measuring the extent to which the aforementioned innovations have altered
the life-style in many societies is almost impossible. At the same time, technological advances in
an increasingly industrialized society affect the natural environment. For example the use of
insecticides in agriculture, like DDT used to control typhus and malaria in the mid-20th century,
but also one of the first chemicals worldwide applied and praised for its insecticidal properties.
The publication of Rachel Carson’s ‘Silent Spring’ in 1962 exposed the adverse effects of
pesticides – not just DDT – on not only the undesired pests but also on ecosystem and human
health (Carson & Darling, 1962). Similarly, changing energy use since the beginning of the
industrial revolution has improved human wealth and health, but is also responsible for
contemporary questions of resource limits, impacts by pollution, and climate stability. The
relation between the environment and technology is double: technology is the cause of and the
cure for numerous environmental issues (Gray, 1989).
1.2.1. A taxonomy of technology and innovation
The study of technological change - from an economic perspective - dates back to more than a
century ago. Technological change is by many understood as the main driver behind long-term
economic development (Rosenberg, 1996). Important contributions to the study of
technological change are accredited to Thorstein Veblen and Joseph A. Schumpeter. Although
earlier authors such as Adam Smith and Karl Marx were active observers in the Industrial
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
5
Revolution, Veblen was the first to emphasise the interactive relation between technology and
social relations (A. Grübler, 1998). Technology was previously considered as the domain of
inventors, engineers and entrepreneurs. Veblen discussed the importance of interactions and
the circular character of these: technology was developed by social actors but at the same time
influencing these actors themselves (see e.g., Veblen (1904, 1921)) (A. Grübler, 1998).
Schumpeter is often perceived as the father of the technological innovation research field. He
stressed the importance of new combinations, particularly those that are radically different
from the current standard, and focussed on the pivotal role of the entrepreneur in achieving
technological innovation (A. Grübler, 1998). His views have been adopted widely by scholars
and have had a dominant influence on the technological innovation discourse (see e.g.,
Schumpeter (1942)). The study of technological innovation gained renewed interest in the
1980s, a period marked by severe economic recessions. Scholars such as C. Freeman, who
specialized in the economics of innovation, technical change and business cycles, induced
increasing interest for technological innovation studies (see e.g., Chris Freeman and Soete
(1997)).
Technology refers to physical artefacts, such as machinery, but also to the knowledge required
to produce and operate these. Technology is not static but continuously subject to change. This
change is caused primarily by the replacement of aging apparatus, but also by new inventions.
(Arnulf Grübler, 1998) An innovation is defined as the introduction of something new: a new
idea, method or device (Merriam-Webster Dictionary, 2016). Christopher Freeman and Perez
(1988) distinguish between incremental and radical innovation, and related changes in
technology systems and changes in techno-economic paradigms. Incremental innovation occurs
rather gradually and continuously. It is often the result of improvements in the production
process, originating from engineers’ or production workers’ suggestions, or end-users’
feedback. Incremental innovations are frequently embodied in upscaling of facilities, and quality
improvements of products and services. Although an incremental innovation may not be
noticed by itself, their combined effect strongly influences productivity growth. (Christopher
Freeman & Perez, 1988) Radical innovations are discontinuous events. Unlike incremental
innovation, radical innovation is often the result of intentional research and development. A
radical innovation could not have been achieved as the result of incremental improvements.
Radical innovations have a more dramatic effect in the sense that they can induce, or be the
trigger for, new investments and new markets. (Christopher Freeman & Perez, 1988) A change
in a technology system is induced by a combination of radical and incremental innovations,
together with organisational and managerial innovations. It entails a far-reaching change in
technology, which influences multiple sectors and even causes the emergence of new sectors.
An entire technological cluster is affected by the innovations. (Christopher Freeman & Perez,
1988) Finally, changes in techno-economic paradigms are marked as a technological revolution.
It occurs as a combination of many radical and incremental innovation clusters and can contain
CHAPTER ONE
6
new technology systems. The changes in the technology system are so far-reaching that the
behaviour of the entire economy is pervasively affected. It entails new products, services,
systems and industries, but also most other branches of the economy are affected. Not only the
technology’s engineering trajectory is affected, but also the relative cost structure of its inputs.
(Christopher Freeman & Perez, 1988) Such techno-economic paradigms are associated with
long-wave Kondratiev cycles. Kondratiev cycles characterize sinusoidal waves of economic
prosperity, recession, depression and improvement, typically observed in periods of about 50
years. In the phase of recession, the new paradigm develops within the old and displays its
advantages. But to become established as the new dominant technological regime, there have
to be structural adjustments throughout the entire economy, including social and institutional
changes. (Christopher Freeman & Perez, 1988) An example of such a shift is the transition to
steam power and railways around 1830, marked as the start of the second Kondratiev wave.
(Christopher Freeman & Perez, 1988) Today, such structural adjustments are often discussed in
terms of transitions. A transition is defined as “a change from one state or condition to another”
(Merriam-Webster Dictionary, 2016) or, more specific for the research field, as “a gradual,
continuous process of change where the structural character of a society (or a complex sub-
system of society) transforms” (Rotmans, Kemp, & van Asselt, 2001, p. 16).
1.2.2. Frameworks in transition research
The study of transitions is a prosperous research field. Many scholars study the characteristics
of transitions in order to understand its components and drivers, but also with the aim to
govern it into desired directions, for instance towards improved sustainability. Markard, Raven,
and Truffer (2012) identify at least four standing theoretical frameworks in transition studies:
the multi-level perspective on socio-technical transitions, transition management, strategic
niche management and technological innovation systems. All these frameworks adopt systemic
views of far-reaching transformation processes of socio-technical systems (Markard et al.,
2012).
The strategic niche management (SNM) approach departs from the existence of technological
regimes, which are defined as “the whole complex of scientific knowledge, engineering practices,
production process technologies, product characteristics, skills and procedures, and institutions
and infrastructures that make up the totality of a technology” (Kemp, Schot, & Hoogma, 1998, p.
182). In other words, technological regimes represent the currently prevailing track of
technological change. Technological changes tend to depart from this regime and aim mostly for
regime optimization rather than transformation, which clarifies why most innovations are
incremental instead of radical (Kemp et al., 1998). When the dominant regime has undesirable
characteristics, for instance because it involves unsustainable practices, the question is how the
regime can be shifted to an improved one. The SNM perspective seeks the answer in the
deployment of new, better technologies which offer an alternative for the dominant
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
7
technological regime. The creation and support of niches is crucial for the deployment of such
new technologies (Markard et al., 2012). The niche is a protected space, where a promising
technology can be developed and used without direct competition with the prevailing regime.
Hence, strategic niche management is the deliberate creation of such niches, with the goal to
learn about the technology and its desirability and to boost the development and diffusion of
the technology. (Kemp et al., 1998)
The multi-level perspective on socio-technical transitions (MLP) expands the insights from the
SNM perspective and sees transitions as the result of the interplay among three levels: the
niches, the socio-technical regime and the landscape (see Figure 1.1). The core of the
framework is again the regime, situated at the meso-level and representing the established,
stable technology structure. However, the original concept of the regime is broadened by
recognizing the importance of the technology itself and its developers, but also that of the
sociological rules that shape it. These rules impact not only the engineering community, but also
other social groups such as policy makers, end-users, suppliers, scientists, capital institutions,
etc. (Frank W. Geels, 2002). This expanded interpretation was entitled the socio-technical
regime to make the distinction with the original technological regime. The literature contains
different definitions of the concept socio-technical regime (see Markard and Truffer (2008)), but
the idea is that of a set of rules, knowledge, common practices, skills, institutions and
infrastructures which apply to not only engineers or scientists, but also to end-users, policy
makers, businesses, societal organisations, etc. (Markard & Truffer, 2008). Generally, the regime
represents the existing technological equilibrium, supporting innovation but mostly of an
incremental nature (Frank W. Geels, 2002), along established pathways of development
(Markard et al., 2012). The landscape is the overarching macro-level and signifies more general
structural trends or the context in which the actors interact but which is external to the actors
or the technology itself (Frank W. Geels, 2002). The niche is to be situated at the micro-level,
acting as an ‘incubation room’ where radical innovations can deploy, protected from the
prevailing selection in the market. Niches provide the space to develop a technology, but also to
allow for technological learning and to build the network to support the innovation. (Frank W.
Geels, 2002) The multi-level perspective emphasizes the importance of the ‘nested character’ of
these three levels. Niches are embedded within the regimes and the regimes are embedded
within the landscapes. The niche is the place where a radical innovation can develop, but a
transition involves multiple technologies as well as the creation of a new socio-technical regime.
The landscape may change as a result of this new regime. Hence, the successful deployment of a
technological transition depends on processes at all three levels (Frank W. Geels, 2002) and
involves changes from one socio-technical system to another (F. W. Geels, 2005).
CHAPTER ONE
8
Figure 1.1. The multi-level perspective on socio-technical transitions.
Legend: Illustration of the interaction between the niches, the regimes and the landscape in the multi-level perspective on socio-
technical transitions. Source: Frank W. Geels (2002).
Alternatively, the transition management (TM) framework was developed from a more practical
approach. The transition management framework finds its origins in a governance experiment
in The Netherlands, conducted to encourage a transition to address several environmental
challenges (Loorbach, 2009; VROM, 2001). Transition management is a governance approach
“based on insights from governance and complex systems theory as much as upon practical
experiment and experience” (Loorbach, 2009), which bundles knowledge about transitions into
a management strategy for both public and private actors (Rotmans et al., 2001). The goal is to
influence governance activities in order to induce a faster change in the desired direction,
mostly towards better sustainability (Loorbach & Rotmans, 2010). The operational approach of
transition management is conceptualized in a four-step transition management process cycle
(Figure 1.2): (1) problem structuring (strategic governance); (2) agenda development (tactical
governance); (3) mobilization and experimentation (operational governance); (4) evaluation,
monitoring and learning (reflexive governance) (Loorbach, 2009). The transition management
approach has been applied for several regional and national policy projects (see e.g., Loorbach
and Rotmans (2010), but it is uncertain which role it will play in actual policy making in the
future (Markard et al., 2012).
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
9
Figure 1.2. The transition management cycle.
Legend: Illustration of the four-step cycle adopted in the transition management approach. Source: Loorbach (2009).
Finally, the technological innovation systems (TIS) framework takes the perspective of novel
technologies and their emergence and considers the institutional and organizational changes
that correspond with this technology development (Markard et al., 2012). The framework is
built around the concept of an innovation system, defined as “the network of institutions in the
public and private sectors whose activities and interactions initiate, import, modify, and diffuse
new technologies” (C. Freeman, 1987). Central to the TIS approach is the idea that certain
essential processes, the so-called functions, should perform well as a prerequisite for the
effectiveness of the innovation system itself (Markard et al., 2012). Most commonly, the TIS
framework recognizes seven central functions in successful technological innovation systems
(Bergek, Jacobsson, Carlsson, Lindmark, & Rickne, 2008; Hekkert, Suurs, Negro, Kuhlmann, &
Smits, 2007): (1) entrepreneurial activity and experimentation; (2) knowledge development; (3)
knowledge diffusion; (4) influence on the direction of search; (5) market formation; (6) resource
mobilization; (7) creation of legitimacy and development of positive externalities. Knowing the
activities that encourage or impede innovation allows to intentionally steer the process (Hekkert
et al., 2007). Considering all these functions, i.e. the complete innovation system, has
broadened the perspective from only the influence of market conditions or failures to that of
the entire system on the technology’s development (Markard et al., 2012).
These four frameworks (MLP, SNM, TM and TIS), suggest that there are two common
approaches to study the change processes associated with a transition. The first method takes a
more aggregated perspective on the transition processes and their drivers (Markard & Truffer,
CHAPTER ONE
10
2008). For instance, the MLP and the TM framework engage with a focus on the transition
process itself and its driving factors. The second approach starts from the perspective of a
particular innovation, which is thought to have potential to contribute to the desired transition,
and focusses on the conditions required for its successful diffusion. (Markard & Truffer, 2008)
This is the perspective taken by the SNM and the TIS frameworks. In researching transitions, the
type of study and its goals determine which approach is best suited.
1.3. Energy sustainability and the role of technological change
The study of technological change can relate to many different aspects of society. The focus in
this work lies on energy. Energy is the mainstay of society. It comes in many different forms,
from chemical to mechanical and thermal energy, and from diverse sources, stocked in the
Earth’s reserves or in the form of a continuous flux. Its functions in society are even more
diverse. Before the Industrial Revolution people primarily used candles for lighting, hearth-fires
for heating and animal or human power for labour and transport. Now, ‘only’ 250 years later, it
is hard to imagine a world without cars, industrial production or electric light. All of these are
the result of innovations in the harvesting, the conversion and the use of different forms of
energy.
1.3.1. Current challenges for energy sustainability
The energy sector has grown to vast magnitudes. The worldwide total primary energy supply
summed to 13,699 Mtoe in 2014. Deducting use by the energy sector itself, losses, etc., the
total worldwide final energy use in the same year amounted 9,425 Mtoe (OECD/IEA, 2016).
More interesting is the distribution of the energy sources used for this production and its
evolution over time (Figure 1.3). More than 80 % of the world’s energy supply originates from
non-renewable energy sources such as coal, oil and natural gas, and another 5 % is produced by
nuclear processes. Of the total final energy use in 2014, 29 % is for the account of the industry
and 28 % for the transport sector (OECD/IEA, 2016).
Confidence in the infinite character of these traditional energy sources caused copious, supply-
induced electricity systems (Verbruggen, 2008). But the fossil-nuclear oligopoly is challenged by
contemporary issues, including discussions on the limits of energy resources (see e.g.,
Verbruggen and Al Marchohi (2010)), the geopolitics of energy supply (see e.g., Van de Graaf
(2017)) and the impact of energy use on the climate. The latest Intergovernmental Panel on
Climate Change (IPCC) report confirms the continuous increase of anthropogenic greenhouse
gas emissions despite numerous mitigation policies (IPCC, 2014). Fossil fuel combustion and
industrial processes make up the largest contributors (±78 %) (IPCC, 2014).
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
11
Figure 1.3. World total primary energy supply (a) in 2014, according to energy source and (b) from 1970 to 2014.
Legend: Illustration of the share of energy sources in the worldwide total primary energy supply (anno 2014 and as a time trend
from 1970 to 2014). Worldwide energy supply is unambiguously dominated by fossil fuels is. Source: OECD/IEA (2016).
In other words, the contemporary practices in energy supply and use do not comply with
fundamental requirements of sustainable development. The concept sustainable development
was put on the agenda by the so-called Brundtland report in 1987, introducing the definition
“sustainable development is development that meets the needs of the present without
compromising the ability of future generations to meet their own needs” (WCED, 1987). The
focus in this definition is on the concept needs, referring in particular to redistribution and
seeing that those who are less economically developed likewise have access to natural
resources, and the idea of limitation, denoting technical, social and environmental limits in the
ability to meet these needs (WCED, 1987). Sustainable development involves requirements
along the four dimensions environmental/ecological (planet), economics (prosperity), social
(people) and governance (politics), together with the fifth dimension risk, which relates to all
the previous four (Verbruggen, Laes, & Lemmens, 2014). In a context of energy supply and use
this implies respect for limited resources, resilience of the local and global environment and
aiming for efficiency (planet) as well as internalizing of external costs, economic efficiency,
affordability of capital investments and reliability (prosperity). Other key criteria are
affordability of the services for end-users, no shift of external and future costs to developing
countries and future generations, low exposure to hazards and reliable information about those
that remain, and global redistribution of access to and wealth accumulation from natural
resources (people). In terms of governance sustainable development calls for global,
independent agencies that study uncertainties and impacts of power choices, independent and
accountable regulatory institutions, priority of public interest over private profits, and citizen
CHAPTER ONE
12
participation in debates about local energy system deployment and governance (politics).
Finally, the energy system should have an acceptable degree of risk, which requires full
insurability and liability for the operators but also that know-how and technological
development should not be shifted from providing basic societal needs for energy to a race for
superior military apparatus (e.g., in the case of proliferation). (Verbruggen et al., 2014)
The current setup of energy systems worldwide is unsustainable in many ways. This realization
prompts the issue of how to transform the energy system into a more sustainable one, i.e.
whether and how it is possible to induce a full energy transition. The matter of energy or
sustainability transitions has received much attention from scholars of multiple fields. Social
scientists focus on the societal drivers of transitions, the interaction between the involved
parties and the structural approaches that may guide such a transition. Engineers search for
technically usable solutions that bring ideas of more sustainable energy systems into practice.
Commonly, one of two general approaches is adopted: either a more aggregated perspective on
the transition and its processes, or the perspective of a technology with perceived potential to
contribute to the transition (cfr. section 1.2.2). This thesis will follow the latter approach and
focus on the process of technological innovation itself as driver for a larger societal change. One
of the overarching goals of sustainability transitions research entails addressing the discharge of
CO2-emissions. This requires decreasing the carbon intensity of electricity supply and reducing
demand via efficiency improvements and behavioural change (IPCC, 2014). On the supply side,
alternatives exist in the form of renewable energy sources. These are naturally regenerated in
the short term, i.e. at a rate which is faster than the rate of use. The counterpart, reducing the
intensity of our energy demand, includes increasing the efficiency of this use. The industry
sector is responsible for 29 % of final energy use (in 2014), and emitted about 13 Gt CO2 in 2010
(IPCC, 2014; OECD/IEA, 2016). It is estimated that a 25 % reduction in energy intensity of the
industry sector could be achieved directly by wide-scale upgrading, replacement and
deployment of best available technologies (IPCC, 2014). However, the actual implementation of
energy efficiency is hindered mainly by high initial investment costs, lack of information (IPCC,
2014) and low energy prices (Verbruggen, 2003). In other words, the role of technological
change in improving energy sustainability is key. It is the combination of multiple innovations,
incremental and radical, as well as organisational and managerial advances that can lead to a
change in a technology system. But it is this same technological system, or regime, that can be
responsible for hindrances in the transition process. The slow and difficult transition towards
sustainable energy is to a large extent the result of path-dependency and lock-in into the
existing fossil fuel regime (R. Raven, 2012; Unruh, 2000). Technologically superior alternatives
are not necessarily adopted as the dominant design and inferior designs may become locked-in
through path-dependent development processes. Think of the QWERTY-keyboard. Typewriters
could get crammed when two adjoining letters were used quickly after each other, so the
QWERTY-keyboard was designed with the specific purpose to position letters in an illogical
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
13
manner and hence slow typists down. When the issues with keyboards were no longer
prevailing (1930s), an alternative was intensively researched and developed: the Dvorak-
keyboard. This keyboard is superior to the QWERTY-keyboard in many ways, but practically
nobody uses it because vested interests support continuation of the business-as-usual. (Rogers,
2003) Unruh (2000) argues this selection process is largely the result of increasing returns to
scale during the development and commercialization process, which accelerates the relative
improvement compared to competing designs.
1.3.2. Technology focus: the organic Rankine cycle
The existence of such path-dependent lock-in of inferior designs is one of the reasons why
transition studies emphasise the role of experimentation and niches as the origin of innovation
(R. Raven, 2012). All four transition frameworks (section 1.2.2) adopt experimentation in niche
markets as a key activity or function. In the context of energy sustainability endeavours,
practices with technology niche management have focussed on for instance the promotion of
electric vehicles (Schot, Hoogma, & Elzen, 1994), wind energy deployment, especially in
Denmark (Garud & Karnøe, 2003) or the development of biogas plants (R. P. J. M. Raven &
Geels, 2010). The focus of this thesis lies on one technology in particular, with perceived
potential to contribute to energy sustainability: the organic Rankine cycle (ORC). The work
conducted in this study originated as part of a larger research project: The Next Generation
Organic Rankine Cycles (ORCNext) (ORCNext, 2012-2016). The ORCNext-project is a strategic
basic research project funded by the Flemish agency for Innovation by Science and Technology
(IWT-SBO 110006). The starting point of the ORCNext-project was the large amount of excess
heat available in industry, which could be recovered and valorised by means of organic Rankine
cycle technology.
The organic Rankine cycle is a thermodynamic power cycle. A thermodynamic cycle is defined as
a sequence of thermodynamic processes, where a system changes from one state to another,
with the final state identical to the initial state (Çengel & Boles, 2002). The ORC is conceptually
based on the conventional steam Rankine cycle, which converts thermal energy into work. This
cycle was invented in 1859 by William Rankine and since then technologically highly refined and
widely applied in steam cycle power plants. The idea of using organic fluids instead of water was
coined not long after the invention of the conventional steam cycle, but the idea remained
unused, or at least subsurface, until the second half of the 20th century. It was the research by
physicist Harry Zvi Tabor and engineer Lucien Bronicki that lead to the construction of the first
working ORC installation in 1961. Driven by the quest to harness more solar energy, they
developed a cycle suitable of recovering heat at lower temperatures than conventionally. At
that point, the technical feasibility of the technology was demonstrated.
CHAPTER ONE
14
The basic component layout of the ORC is identical to that of the steam cycle and visualized in
Figure 1.4 (a). The key components are an evaporator, expander, condenser and a pump,
completed by the working fluid. The main difference with the steam cycle is that instead of
water an alternative medium is used as working fluid. The alternative fluid is typically a fluid
with a high molecular mass and a boiling point lower than that of water. A temperature-entropy
(T-s) diagram of the cycle is shown in Figure 1.4 (b). The working principle of the cycle goes as
follows. Firstly, the hot working fluid leaving the turbine (1) is condensed in the condenser. The
heat from the condensation process is transferred to a cooling loop (7-8) which typically consists
of water or air. Subsequently, the condensed working fluid enters the pump (2) and is
pressurized (3). Then, the working fluid enters the evaporator and is heated to a superheated
state (4). The temperature of the heat carrier (5-6) is thus gradually reduced. The superheated
vapour enters the turbine in which it is expanded and the cycle is repeated. This basic layout of
the cycle is commonly identified as the subcritical ORC (SCORC).
The working fluid used in the cycle determines its suitability to retrieve thermal energy from an
energy source. Water as working fluid is well apt for thermal sources with higher temperatures,
but is limited for use in smaller systems with lower-temperature thermal sources. Using organic
working fluids allows to adapt the cycle for a better pairing with energy sources of lower
temperature. This implies that changing from the conventional steam cycle to one using organic
working fluids opens the opportunity to utilize ample low-temperature thermal energy sources
that could not be employed previously. The sources associated with ORC technology mostly
include renewable sources such as biomass, geothermal or solar energy or waste sources such
as excess heat from industrial processes. In other words, ORC technology has potential to
contribute to worldwide renewable energy goals and to improve energy efficiency by using
discarded energy flows.
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
15
Figure 1.4. (a) Component layout and (b) T-s diagram of the basic subcritical ORC.
Legend: Illustration of the working principle of the organic Rankine cycle (ORC). (a) The ORC has four essential components (evaporator, turbine, condenser and pump), completed by a generator and an organic working fluid. (b) The temperature-
entropy (T-s) diagram shows the changes in temperature (T) and entropy (s) during the thermodynamic process of the ORC. Copyright: S.Lecompte.
Electricity generation from low temperature heat flows is possible with several technologies,
but the ORC can be considered the most mature technology at the moment (Crook, 1994;
Oluleye, Jobson, Smith, & Perry, 2016). Nevertheless, there is room for improvement in terms
of, for instance, enhancing the efficiency of current ORC cycles or adapting ORC systems to
allow utilization of thermal sources with a lower flow rate. In this background, the ORCNext-
project identified five challenges and attempts to answer these in six different research work
packages (see Table 1.1) (ORCNext, 2011). The research conducted for this thesis originates
from the challenges identified for WP6 of the project, focussing on the economic dimension of
the technology’s development.
Table 1.1. Challenges and Work Packages in the ORCNext-project.
Challenges Work Packages
1. Improving the efficiency
2. Control over the dynamic behavior
3. Reducing the design time
4. Availability of test-infrastructure and
expertise
5. Economic and financial analysis of the
technology
WP1: Cycle architectures, design
methods and optimizations
WP2: Screw expander technology
WP3: Dynamical behavior of ORC cycles
WP4: Supercritical ORC
WP5: Model validation by means of
lab-scale testing
WP6: Economic and financial analysis
WP7: Project management
Source: ORCNext (2011).
2
Condenser
G
Generator
Evaporator
Turbine
Pump
1
3
6
5
7
8
4
5
6
7
8
4
2
3
1T
S
a b
CHAPTER ONE
16
1.4. The potential of ORC technology: framework and methodology
This thesis starts from the interaction between technology-specific innovation and energy
sustainability. The aim is to investigate the potential of ORC technology, with a focus on the
economic aspects. In its most general form, the research question addressed in this thesis takes
the form:
“What is the potential of organic Rankine cycle (ORC) technology to contribute to
renewable energy and energy efficiency goals?”
However, studying the potential of a particular technology entails many aspects. Section 1.2.2
identified two frameworks suitable for studying an emerging technological innovation: strategic
niche management (SNM) and technological innovation systems (TIS). In the context of SNM, a
technological niche is a protected space to learn about a new technology (R. Raven, 2012).
Central themes in the SNM approach are the relation between local experiments and global
niches, the role of expectations, the shaping of new social networks and constituencies, the role
of learning, the protection of the niche and, finally, the relation between the niche and the
regime (R. Raven, 2012). The TIS framework delineates the functions that a successful
innovation system must possess: entrepreneurial activity and experimentation, knowledge
development, knowledge diffusion, influence on the direction of search, market formation,
resource mobilization and creation of legitimacy and development of positive externalities. For
emerging technologies, the starting point is that the TIS still has to be built up (Junginger, van
Sark, & Faaij, 2010).
This thesis draws insights from both approaches to structure the research, but a complete
investigation of all aspects of the technology’s innovation path reaches beyond the scope of this
work. The technical features of ORC technology are a vividly studied domain, but the economic
perspective is a largely unexplored frontier. Hence, this thesis starts at the very basis and
increases the scope of the investigations step by step. The subsequent sections discuss the
structure of the thesis, the research questions of relevance for each part and the applied
methodology. A schematic overview of the research frame is presented in Figure 1.5. The ORC is
central in the research. Its deployment is influenced by its own technological features, but also
by its economic performance, the prevailing conditions in the market and the institutional
framework. Each of these aspects are interlinked and influenced by insights from innovation
studies.
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
17
Figure 1.5. Schematic overview of the research framework.
Legend: Illustration of the different levels of influence considered in this research.
Bio = biomass energy; Geo = geothermal energy; ORC = organic Rankine cycle; Sol = solar energy.
Innovation theory
POLICY & INSTITUTIONS
MARKETS & ECONOMIC CONTEXT
Power & Heat Excess heat, Bio, Geo, Sol
Project
Module
Competing technologies
Europe
Global
Belgium
Innovation & Experience
Cost of power & heat Investment costs
TECHNOLOGY & ECONOMICS
ORC
CHAPTER ONE
18
1.4.1. Technology and economics
In its most general form, the goal of this study is to attain insight in the potential of ORC
technology. Understanding the potential of a technology starts with recognizing its merits. More
specifically, a potential can entail many aspects, but commonly the distinction is made between
the physical the technical, the economic and the market potential (see e.g., Verbruggen et al.
(2010)).
The organic Rankine cycle: technical characteristics, applications and economics (Chapter 2)
Although not the primary goal of this thesis, the first step involves understanding the technical
potential of ORC technology and its merits, which emanate from its technical characteristics.
Insight in the merits clarifies why studying ORC technology is interesting in the first place.
Section 1.3.2 introduced the working principle itself. Chapter 2 builds on these insights and
discusses the state-of-the-art knowledge on ORC technology. The focus of this chapter remains
relatively narrow, with emphasis on the technology itself, its merits and a review of the current
knowledge about the costs of ORC systems. Insight into the costs constitutes the basis to draw
an idea of the technology’s economic potential. Hence, the research question addressed in this
chapter can be formulated as follows:
“What are the merits of ORC technology? And what is the state-of-the art
knowledge about its economics?”
The answers to these questions are sought for by exploring the literature. Although ORC
technology constitutes a very active domain of research, the large majority of efforts are
focussed on technical optimization. A relatively small share of the literature engages in
discussing the economics of ORC technology and the majority of these entail engineering
studies that provide bottom-up cost projections of the technology design under consideration.
By knowledge of the author, no comprehensive review of the economic ORC literature has been
made until today. Hence, the contribution of chapter 2 consists of an extensive insight in the
current knowledge about the costs of ORC systems in different applications.
An important contribution of this review is the framework established to structure the
literature. The literature on ORC systems is very diverse and the scope of the economic
investigation differs strongly. Therefore, this review structures the economic publications
according to three defined criteria: the application scope, the system scope and the economic
scope (see Table 1.2). The application scope refers to the heat source used to operate the ORC
system. The system scope refers to the delineation of the ORC system in the study. The
terminology used in this work is presented in Figure 1.6. A distinction is made between an ORC
module, an ORC project and an ORC system. The term ORC module refers to only the ORC unit
itself. It is the foundation of the ORC system (evaporator, expander, condenser, pump and
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
19
generator). ORC Integration refers to the equipment needed to install the ORC module. The
integration requirements differ strongly per type of heat source and according to the
requirements and limitations on-site. The project costs refer to the additional costs that are
involved in the completion of the ORC project. These include labour costs, costs for start-up,
engineering, etc. The term ORC project concerns the complete ORC power system, including the
necessary piping, heat exchangers, construction labour, start-up, contingencies, etc. These
integration requirements are very site-specific and vary by application. The generic terminology
ORC System is used to refer generally to the ORC power system, when the distinction between
module and project is not relevant. Finally, the economic scope refers to the origin of the cost
data published for the ORC system. The figures can refer to the costs of a real ORC system,
which is the most valuable type of ORC cost information. Alternatively, the costs can be
provided by manufacturers in terms of price quotes. Subsequently, the costs can be estimated
using various techniques. A final option is where the costs of the ORC system are simply
assumed. This delineation of the application, system and economic scope is not only used to
structure the literature review in chapter 2, but remains relevant throughout the thesis.
Table 1.2. Three criteria defined to structure the literature on ORC economics.
Criterion Criterion delineation Categories
Application scope Which heat source is used? Excess heat Biomass Geothermal Solar Hybrid Other Not specified
System scope What is the scope of the ORC costs? ORC module ORC project
Economic scope a. What is the origin of the ORC costs? Real ORC system Manufacturer’s quote Estimated Assumed
b. Does the study perform a financial project appraisal?
Yes No
CHAPTER ONE
20
Figure 1.6. Terminology for delineation for ORC system scopes: ORC modules, projects, integration costs and project costs.
Legend: The scope of the ORC system is defined according to the terminology ORC Module, ORC Integration, Project Costs, ORC
Project and ORC System.
Cost engineering techniques and their application for ORC technology (Chapter 3)
The focus in chapter 2 is the delineation of state-of-the art insights in the merits and costs of
ORC technology. An important finding is that the large majority of the literature on ORC
economics involves costs that have been estimated rather than real system costs. This raises the
question to what extent these estimated costs are representative, i.e. can they be used to make
statements about the economics of ORC technology or should they be interpreted as merely
indicative? Chapter 3 explores the principles of cost engineering and assesses their use for ORC
cost estimation. The aim of this chapter is to get insight in the methodologies that are widely
used in the ORC literature, but not so often critically assessed. The goal is to see whether it is
appropriate to use existing cost engineering practices to attain insight in the economics of ORC
technology. The research question addressed in this chapter can thus be formulated as:
“To what extent can cost engineering techniques be used to get insight in the costs
of ORC systems?”
To answer this question, the first approach is an exploration of the theory of cost engineering.
Many of the available practices and data sources stem from applications in the chemical
engineering industry. These methods are now widely used to estimate the costs of industrial
processes. To investigate their suitability for ORC research, chapter 3 presents the data
retrieved from a commissioned case study and applies multiple cost engineering techniques to
estimate the costs of this system. This allows to evaluate the methodologies applied in each of
the techniques, as well as to compare them mutually and to assess their accuracy in
approaching the costs of the real system.
ORC Module
Equipment such as:
- Heat recovery heat
exchanger;
- Biomass combustion
equipment;
- Geothermal well piping &
pumps;
- Solar installation;
- …
ORC Integration
Non-equipment costs such as:
- Costs for transportation;
- Labor costs for on-site
construction;
- Costs for start-up;
- Costs for engineering or
project planning;
- …
Project costs
ORC Project
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
21
1.4.2. Markets and economic context & policy and institutions
A key facet in the deployment of a technological innovation is the context in which it emerges.
Both the SNM and the TIS framework stress the importance of experimentation and knowledge
development within the niche. But to succeed beyond the stage of experimentation the
technology has to stand ground in the prevailing regime. An important question in this matter is
the extent to which ORC technology is evaluated as an economically viable option by potential
end-users. This issue is addressed in chapter 4 of the thesis.
Case study of an organic Rankine cycle applied for excess heat recovery (Chapter 4)
The literature on ORC technology is abundant, but contains only few insights in the economics
of real, commissioned ORC systems (cfr. Chapter 2). The bulk of research on the economics of
ORC is based on costs that have been estimated and, although some of the cost estimation
methods are very meticulous, they cannot replace insights from real project commissions (cfr.
chapter 3). Therefore, chapter 4 presents and investigates insights from an ORC case study,
commissioned in Flanders, Belgium. The research question at the core of this chapter is:
“Which factors influence the financial feasibility of an investment in ORC
technology?”
This question is answered from different perspectives. First of all, the case study gives insight in
the relative importance of each of the cost factors. The economics ORC systems are typically
characterized by substantial capital investments and relatively low annual costs, dependent on
the type of heat source used. The case study under investigation utilizes industrial excess heat,
which implies there are no costs involved for fuel purchasing. A second aspect is the influence of
public policy. Chapter 4 introduces the extent to which extent heat recovery is part of public
policy in the European Union, and discusses more in detail which policy measures are relevant
for this specific investment in the Flemish context. Finally, the case study is used to get an
understanding of the relative importance of each of the project parameters on the financial
appeal of an ORC investment.
1.4.3. Technology and innovation
Finally, a central role in the study of innovations is assigned to technology learning. Learning
entails the accumulation of knowledge over time. In the context of technological innovation
studies, learning is often understood in terms of the increasing experience gained from
producing and using a particular technology. Gaining experience with a technology is thus
inherently interwoven with its application, i.e. there is no learning-by-doing without doing. It
involves learning related to user preferences, cultural connotations and industrial networks,
about policy and regulation and the societal and environmental impact of the innovation (R.
CHAPTER ONE
22
Raven, 2012). Most often however, learning is studied in a narrower sense and conceptualizes
the impact of accumulating experience with a technology on its costs. The deployment of ORC
technology in the market and the related cost dynamics are the subject of chapter 5.
The dynamics of ORC technology: technological innovation, economies of scale and learning by
doing (Chapter 5)
The deployment of an emerging technology is often conceptualized in terms of a technology life
cycle, where the deployment starts with its initial invention and – in some cases – ends in
market maturity and saturation. The deployment of ORC technology is investigated in chapter 5
by means of an extensive exploration of the ORC market and its development. Moreover, the
analysis expands to the impact of learning-by-doing, but also that of technology and production
scaling, on the costs of ORC technology. The question at the core of this chapter is then:
“How did ORC technology diffuse and how does the deployment of the technology
influence its costs?”
This research question is answered by means of an extensive collection of data on the market
and the economics of ORC technology. First of all, to study the diffusion of the technology a
database was composed, containing approximately 95 % of all ORC systems commissioned or
sold worldwide. This data gives insight in the historical deployment of the technology, as well as
the marker share of each of the ORC manufacturers, the relative use of various energy input
sources and the geographical distribution of the ORC systems. Secondly, a database was
composed with insight in the economics of ORC systems from various origins and in different
applications. This second database is used to analyse the extent to which ORC systems are
subject to economies of scale, i.e. whether an increase in scale corresponds to a reduction in
average costs. Finally, chapter 5 studies the impact of the quickly accumulating experience with
ORC systems on the development of the technology’s costs. By knowledge of the author, the
diffusion and costs of ORC technology have never been studied at such a scale previously.
INTRODUCTORY CHAPTER: TECHNOLOGICAL INNOVATION AND CHALLENGES FOR ENERGY SUSTAINABILITY
23
Bibliography
Bergek, A., Jacobsson, S., Carlsson, B., Lindmark, S., & Rickne, A. (2008). Analyzing the functional dynamics of technological innovation systems: A scheme of analysis. Research Policy, 37(3), 407-429. doi:10.1016/j.respol.2007.12.003
Carson, R., & Darling, L. (1962). Silent spring. Boston, USA: Houghton Mifflin. Çengel, Y. A., & Boles, M. A. (2002). Thermodynamics: an engineering approach (Fourth ed.). New York,
USA: McGraw-Hill. Crook, A. W. (1994). Profiting from Low-Grade Heat. Freeman, C. (1987). Technology Policy and Economic Performance: Lessons from Japan. London, UK:
Pinter Publishers Ltd. Freeman, C., & Perez, C. (1988). Structural crises of adjustment, business cycles and investment
behaviour. In G. Dosi, C. Freeman, R. Nelson, G. Silverberg, & L. Soete (Eds.), Technical Change and Economic Theory (pp. 38-66). London, UK: Pinter.
Freeman, C., & Soete, L. (1997). The Economics of Industrial Innovation (Third ed.). London, UK: Pinter. Garud, R., & Karnøe, P. (2003). Bricolage versus breakthrough: distributed and embedded agency in
technology entrepreneurship. Research Policy, 32(2), 277-300. Geels, F. W. (2002). Technological transitions as evolutionary reconfiguration processes: a multi-level
perspective and a case study. Research Policy, 31, 1257-1274. Geels, F. W. (2005). The dynamics of transitions in socio-technical systems: A multi-level analysis of the
transition pathway from horse-drawn carriages to automobiles (1860–1930). Technology Analysis & Strategic Management, 17(4), 445-476. doi:10.1080/09537320500357319
Gray, P. E. (1989). The Paradox of Technological Development. In J. H. Ausubel & H. E. Sladovich (Eds.), Technology and Environment (pp. 192-204). Washington D.C., USA: National Academy Press.
Grübler, A. (1998). Technology and global change. Cambridge, UK: Cambridge University Press, 452 pp. Hekkert, M. P., Suurs, R. A. A., Negro, S. O., Kuhlmann, S., & Smits, R. E. H. M. (2007). Functions of
innovation systems: A new approach for analysing technological change. Technological Forecasting and Social Change, 74(4), 413-432. doi:10.1016/j.techfore.2006.03.002
IPCC. (2014). Summary for Policymakers. Climate Change 2014, Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, [Edenhofer, O., R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von Stechow, T. Zwickel and J.C. Minx (eds.)]. Cambridge, United Kingdom and New York, NY, USA., Intergovernmental Panel on Climate Change: 33 pp.
Junginger, M., van Sark, W., & Faaij, A. (2010). Technological Learning in the Energy Sector: Lessons for Policy, Industry and Science. Cheltenham, UK: Edward Elgar Publishing Limited.
Kemp, R., Schot, J., & Hoogma, R. (1998). Regime shifts to sustainability through processes of niche formation: The approach of strategic niche management. Technology Analysis & Strategic Management, 10(2), 175-198. doi:10.1080/09537329808524310
Loorbach, D. (2009). Transition Management for Sustainable Development: A Prescriptive Complexity-Based Governance Framework. Governance: An International Journal of Policy, Administration, and Institutions, 23(1), 161-183.
Loorbach, D., & Rotmans, J. (2010). The practice of transition management: Examples and lessons from four distinct cases. Futures, 42(3), 237-246. doi:10.1016/j.futures.2009.11.009
Markard, J., Raven, R., & Truffer, B. (2012). Sustainability transitions: An emerging field of research and its prospects. Research Policy, 41(6), 955-967. doi:10.1016/j.respol.2012.02.013
CHAPTER ONE
24
Markard, J., & Truffer, B. (2008). Technological innovation systems and the multi-level perspective: Towards an integrated framework. Research Policy, 37(4), 596-615. doi:10.1016/j.respol.2008.01.004
Merriam-Webster Dictionary. (2016). Innovation. Retrieved on 16/08/2016 from http://www.merriam-webster.com/dictionary/innovation.
OECD/IEA. (2016). Key world energy statistics. Paris, France, Organisation for Economic Co-operation and Development / International Energy Agency: 80 pp.
Oluleye, G., Jobson, M., Smith, R., & Perry, S. J. (2016). Evaluating the potential of process sites for waste heat recovery. Applied Energy, 161, 627-646. doi:10.1016/j.apenergy.2015.07.011
ORCNext. (2011). IWT Vlaanderen SBO call 2010-2011. Project proposal strategic basic research (primary economical relevance): The next generation Oganic Rankine cycles (ORCNext). Flanders, Belgium, Universiteit Gent, Universiteit Antwerpen, Uliversité de Liège, HoWest, Atlas Copco, SET and Power-Link: 175 pp.
ORCNext. (2012-2016). The Next Generation Organic Rankine Cycles (ORCNext). Raven, R. (2012). Analyzing Emerging Sustainable Energy Niches in Europe: A Strategic Niche
Management Perspective. In G. Verbong & D. Loorbach (Eds.), Governing the Energy Transition: Reality, Illusion or Necessity? (pp. 376). New York, NY, USA: Routledge.
Raven, R. P. J. M., & Geels, F. W. (2010). Socio-cognitive evolution in niche development: Comparative analysis of biogas development in Denmark and the Netherlands (1973–2004). Technovation, 30(2), 87-99. doi:10.1016/j.technovation.2009.08.006
Rogers, E. M. (2003). Diffusion of Innovations (Fifth ed.). New York, NY, USA: Free Press. Rosenberg, N. (1996). Perspectives on Technology. New York, USA: Cambridge University Press. Rotmans, J., Kemp, R., & van Asselt, M. (2001). More evolution than revolution: transition management
in public policy. Foresight, 3(1), 15-31. doi:10.1108/14636680110803003 Schot, J., Hoogma, R., & Elzen, B. (1994). Strategies for shifting technological systems: The case of the
automobile system. Futures, 26(10), 1060-1076. Schumpeter, J. A. (1942). Capitalism, Socialism and Democracy. New York, NY, USA: Harper & Brothers. Unruh, G. C. (2000). Understanding carbon lock-in. Energy Policy, 28(12), 817-830. Van de Graaf, T. (2017). Is OPEC dead? Oil exporters, the Paris agreement and the transition to a post-
carbon world. Energy Research & Social Science, 23, 182-188. doi:10.1016/j.erss.2016.10.005 Veblen, T. (1904). The Theory of the Business Entreprise. New York, NY, USA: Scribner. Veblen, T. (1921). The Engineers and The Price System. New York, NY, USA: Huebsch. Verbruggen, A. (2003). Stalemate in energy markets: supply extension versus demand reduction. Energy
Policy, 31(14), 1431-1440. doi:10.1016/s0301-4215(02)00330-0 Verbruggen, A. (2008). Renewable and nuclear power: A common future? Energy Policy, 36(11), 4036-
4047. doi:10.1016/j.enpol.2008.06.024 Verbruggen, A., & Al Marchohi, M. (2010). Views on peak oil and its relation to climate change policy.
Energy Policy, 38(10), 5572-5581. doi:10.1016/j.enpol.2010.05.002 Verbruggen, A., Fischedick, M., Moomaw, W., Weir, T., Nadaï, A., Nilsson, L. J., . . . Sathaye, J. (2010).
Renewable energy costs, potentials, barriers: Conceptual issues. Energy Policy, 38(2), 850-861. doi:10.1016/j.enpol.2009.10.036
Verbruggen, A., Laes, E., & Lemmens, S. (2014). Assessment of the actual sustainability of nuclear fission power. Renewable and Sustainable Energy Reviews, 32, 16-28. doi:10.1016/j.rser.2014.01.008
VROM. (2001). Nationaal Milieubeleidsplan 4: een wereld en een wil. Den Haag, The Netherlands, Ministerie van volkshuisvesting, ruimtelijke ordening en milieu: 374 pp.
WCED. (1987). Our Common Future. Oxford, UK, World Commission on Environment and Development, United Nations: 383 pp.
2. The organic Rankine cycle:
technical characteristics,
applications and
economics
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
27
2.1. Introduction
The starting point of this work is the quest for energy sustainability and the role of technological
change. The adopted approach is that of an emerging technology with perceived potential to
address energy sustainability challenges: the organic Rankine cycle. The development of an
emerging technology is influenced by many aspects, but at the core of understanding its
potential insight in its functioning and the applications it can be used for. Therefore, this chapter
initiates this research on ORC technology by focussing on the technology itself and its
economics (cfr. Figure 1.5).
The first step involves acquiring insight in the technology’s functioning and its merits. Relevant
issues concern the working principle of the technology, its (dis)advantages, and its potential
applications. The working principle of the ORC was introduced in the first chapter (section
1.3.2). Chapter 2 expands this technology briefing and discusses which energy sources can be
used to generate power with ORC systems. Starting with this very narrow scope, the aim is to
identify why one should be interested in investigating ORC technology. But the main goal of this
chapter is to draw the basis for this economic investigation. Via an extensive review of the
literature, this chapter provides a comprehensive overview of the current knowledge on the
costs of ORC systems. By knowledge of the author, the ORC economics literature has not been
reviewed at such a scale before.
This chapter is organized as follows. Section 2.2 discusses the specificities of ORC technology.
Although an extensive investigation of the cycle’s technical characteristics and variations lies
beyond the scope of this thesis, this section discusses the elementary knowledge on the
technology as a basis for the remainder of the research. Topics include the relation with the
conventional steam cycle and a discussion of the main (dis)advantages in comparison to this
traditional cycle. Moreover, the various application fields for ORC technology, in terms of
thermal energy source, are introduced and the current use of each of them is discussed.
Subsequently, section 2.3 reviews the literature for data on the economics of ORC systems. Each
of the four principal heat sources, excess heat, biomass, geothermal and solar energy, is
considered separately. Finally, section 2.4 discusses the results and section 2.5 draws some
concluding remarks.
2.2. The organic Rankine cycle
The organic Rankine cycle (ORC) is a thermodynamic power cycle, converting thermal energy to
power. The ORC finds its origins in the conventional steam Rankine cycle. The working principle
and the basic component layout are basically the same (cfr. section 1.3.2). The key distinction
between the organic and the steam Rankine cycle is the working fluid used within the cycle.
CHAPTER TWO
28
2.2.1. Working fluids
The conventional Rankine cycle uses water as working fluid, but in an ORC an alternative
working medium is used, typically with a high molecular mass and a boiling point lower than
that of water. A comprehensive review on ORC working fluid selection is discussed in the work
of Sylvain Quoilin, van den Broek, Declaye, Dewallef, and Lemort (2013). The working fluid plays
an important role regarding the efficiency, operation and environmental impact of the system,
and hence its economics (Amini, Mirkhani, Pakjesm Pourfard, Ashjaee, & Khodkar, 2015). Water
has several advantages for use as working fluid, such as very good thermal/chemical stability,
very low viscosity, it is a good energy carrier, non-toxic, non-flammable and environmentally
friendly, cheap and abundant (Tchanche, Lambrinos, Frangoudakis, & Papadakis, 2011). These
characteristics make water very suitable for high temperature applications and large centralized
systems (Tchanche et al., 2011). On the other hand, water requires superheating to prevent
condensation, potentially erodes turbine blades, may cause excess pressure in the evaporator
and requires complex and expensive turbines (Tchanche et al., 2011). These issues can be
alleviated by selecting organic fluids for small and medium scale power plants (Tchanche et al.,
2011). Organic fluids characterized by higher molecular masses and lower evaporation
temperatures than water are suitable for electricity generation from lower temperature
sources. Tchanche et al. (2011) summarize the advantages of ORCs over classic steam plants:
the evaporation process requires less heat, it occurs at lower pressure and temperature,
superheating is not necessary since the expansion process ends in the vapour region, this also
avoids the risk of blades erosion, and finally, simple single stage turbines can be used because
the pressure drop/ratio will be much smaller due to the smaller temperature difference
between evaporation and condensation (Tchanche et al., 2011). The characteristics of an
appropriate fluid for ORC applications are well explained by Vélez et al. (2012), Sylvain Quoilin
et al. (2013) and summarized by Tchanche et al. (2011) (see Table 2.1). The comparison
between organic and steam Rankine cycles is elaborated further by Sylvain Quoilin et al. (2013).
The temperature range handled by the ORC determines the suitability of the type of working
fluid for a certain application (Arvay, Muller, Ramdeen, & Cunningham, 2011). In general,
refrigerants are used for the lowest temperatures (100-180°C), hydrocarbons suit the
intermediate (180-250°C) temperature range and siloxanes fit the higher end (250-400°C) (Vélez
et al., 2012). In practice, the list of usable fluids is rather short. Only few working fluids are
currently available on the market and some of these are being phased out due to their adverse
environmental impacts (Tchanche, Declaye, Quoilin, Papadakis, & Lemort, 2010).
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
29
Table 2.1. Characteristics of suitable ORC working fluids.
Characteristic of suitable ORC working fluids
Isentropic or dry fluids (zero or positive slope vapour saturation curve)
Good thermal and chemical stability (stable at high temperature)
Good heat transfer properties (low viscosity, high thermal conductivity)
Good compatibility with materials (non-corrosive)
Moderate critical parameters (temperature, pressure)
High thermodynamic performance (high energetic/exergetic efficiency)
Acceptable condensing and evaporating pressures (> 1 bar and < 25 bar respectively)
Good safety characteristics (non-toxic, non-flammable)
High density (liquid/vapour phase) Low environmental impacts (low ODP & GWP)
High specific heat Low cost High latent heat of vaporization Good availability
Legend: ODP = ozone depletion potential; GWP = global warming potential. The ODP and GWP of chemicals are defined to assess
its adverse environmental impacts and their use is regulated in the Montreal Protocol. Source: Tchanche et al. (2011).
2.2.2. Cycle components and architectures
The standard layout of the ORC is commonly identified as the subcritical ORC (SCORC). It is
composed of four main components: an evaporator, an expander, a condenser and a pump. The
majority of the components used in an ORC system are standardized equipment. Particularly the
pump, expander and the condenser are common industrial components and mostly acquired
from experienced manufacturers in their category. The expander constitutes the core of the
ORC system. The application specificities, for instance the operating conditions and the size of
the thermal energy flow, determine which expander type is suitable. Whereas larger-scale
applications (> 1 MW) tend towards axial turbines for the expansion, smaller-scale systems
usually operate on radial turbines (ORCNext, 2011). For very small scale applications (of a few
kW), volumetric systems such as scroll and screw expanders are a suitable option (Declaye,
Quoilin, Guillaume, & Lemort, 2013; Papes, Degroote, & Vierendeels, 2015). The development
of suitable volumetric expanders for ORC applications is an interesting line of research, since
these are not affordably available yet. Oftentimes, a compressor is modified and turned into an
expansion device (Declaye et al., 2013). Hence, the expander determines the type of
applications where the ORC system can be utilized. Most ORC manufacturers construct their
systems with one type of expander and adapted cycle design and are thus specialized in a
particular application type.
A major strand of research involves the study of variations to the basic component layout, i.e.
the cycle architecture, in order to improve the cycle’s performance. Alternatives involve for
instance the introduction of a recuperator in the cycle, or regenerative feed heating in cycles
with turbine bleeding, both with the aim to reuse the heat after it leaves the turbine to pre-heat
the working fluid. Other alternatives are an organic flash cycle, in which a flash tank is added to
CHAPTER TWO
30
the cycle to increase the thermal energy uptake from the heat carrier, or a triangular cycle,
where the working fluid is not boiled in the evaporator and fed to the expander directly. Further
cycle variations under investigation in the literature are the use of zeotropic mixtures as
working fluids, composed of components with different boiling points which allows a good
pairing with the evaporator and condenser temperatures, or transcritical cycles, where the
working fluid is brought to a critical state. Lastly, dividing the ORC into different pressure levels
allows to transfer more of the heat to the cycle and yields high thermal efficiencies. A detailed
discussion of these architecture variations and their virtues lies beyond the scope of this work,
but reviews are provided by e.g., Chen, Goswami, and Stefanakos (2010) or Lecompte,
Huisseune, van den Broek, Vanslambrouck, and De Paepe (2015). Although such cycle variations
are a recurring topic in the scientific literature, most commercial applications today are based
on the SCORC setup.
2.2.3. Areas of application
Understanding the nature of ORCs reveals its potential for practical application. ORC systems
can be used to generate electricity in smaller capacities (starting from only a few kW) compared
to steam cycles and from lower-temperature heat sources (from 90 °C onwards). Hence, ORCs
are typically associated with renewable power generation, from sources such as biomass,
geothermal or solar energy. Alternatively, ORC systems are suitable as a measure to improve
the energy use efficiency of a process or site by means of excess heat recuperation and
conversion. Enhanced efficiency of energy use and renewable electricity generation are both
essential in the transition of energy sectors to more streamlined, efficient, secure and climate-
friendly systems.
Energy efficiency improvement: excess heat recovery
Industrial processes often discard heat flows together with their main production activity. This
rejected heat is mostly referred to as waste heat, but to avoid confusion with other types of
waste some prefer the terminology surplus heat (as used by e.g., Hammond and Norman
(2014)) or excess heat (see e.g., Broberg Viklund and Johansson (2014)). In the remainder of this
work the term excess heat will be used.
Industrial excess heat recovery: heat streams that would otherwise be dissipated could be
recovered and employed for heating purposes or electricity generation. The idea is conceptually
simple but requires an a priori definition of excess heat itself. The opposite, useful heat, is
defined by the European Union (EU) as “heat produced in a cogeneration process to satisfy
economically justifiable demand for heating or cooling” (EU, 2012, p. 315/312). Bendig,
Maréchal, and Favrat (2013) identify several levels of detail for defining ‘waste heat’:
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
31
“Waste heat is heat dissipated to the environment”. This definition follows the first law
of thermodynamics and has the energy balance as a basis (the higher the input, the
higher the output). Users of this definition often disregard the temperature of the heat
stream and its potential use and reuse.
“Low grade waste heat is heat that is not viable for heat recovery within the process”.
This definition includes the notion of usefulness, thereby pointing at the possibility of
heat recycling. Process efficiency optimization then occurs in a hierarchical fashion: heat
recycling within the process (reduction of resource use) is first, and then heat recovery
by a secondary process (increasing system boundaries). Still, this definition is not
complete as it ignores the different temperature levels at which the heat may be
available. Also, temperature is not the only criterion to measure the excess heat
potential. Including a total energy balance would provide a better overview on the
match of the heat sink and the heat load.
Extension of the focus beyond the sensible heat leads Bendig et al. (2013) to the
definition “waste heat as the sum of the exergy that is available in a process after pinch
analysis, heat recovery, process integration and energy conversion (utility) integration
with the help of exergy analysis”. To delineate this definition, a distinction is made
between the residual and the avoidable heat. Residual heat is the amount of excess heat
that a process unavoidably must produce under optimal heat recovery conditions; it can
be used freely without having an impact on the process’ energy balance. Avoidable heat
should not be employed in secondary heat recovery systems, as they could become
unnecessary when the main process’s efficiency increases, or even block investments in
energy efficiency. Hence, internal heat recovery and the use of excess heat valorisation
techniques are competing.
The latter definition of excess heat leads to the understanding that excess heat recovery
involves a secondary process or system to valorise the heat for end-uses different from the main
process’s operation. Following this definition, excess heat recovery is a case of waste
management rather than energy efficiency. On the other hand, energy efficiency refers to the
production of a certain useful output with a minimum of energy inputs. A process using less
energy inputs for the same output is considered more efficient. Excess heat recovery does
expand the boundaries of the system by including additional useful applications thereby
improving the plant’s utilization of input energy. This increases the useful output per energy
input so the energy efficiency of the plant as a whole has improved. Hence, excess heat
recovery systems can be interpreted in two ways: 1) as a matter of waste management; 2) as a
means for improving the energy use efficiency at the level for an industrial plant or site (rather
than at the process level). Whereas the IPCC sees energy recovery as a mitigation option in
CHAPTER TWO
32
waste management (IPCC, 2014), most policy documents classify energy recovery under energy
efficiency.
Useful applications for recovered excess heat exist in meeting on-site or off-site heating, cooling
or electricity demands (Oluleye, Jobson, & Smith, 2014). An individual plant could create savings
on the own energy bills, or even an additional source of income, depending on the cost-
effectiveness of the heat recuperation opportunities. In general, utilizing excess heat could
reduce the primary energy requirements for certain industry parks, regions or countries and
improve overall energy efficiency. The potential for industrial excess heat recovery is generally
assumed large, but a detailed study on the available amount of excess heat in Europe is not
available and country estimates are scarce. Pellegrino, Margolis, Justiniano, Miller, and Thedki
(2004) evaluate the energy losses and recovery potential across 15 industrial sectors.
Viswanathan, Davies, and Holbery (2006) build on the previous work and report a 10,500 PJ of
available energy from thermal emissions by the US Industry. Hammond and Norman (2014)
estimate a 37-73PJ/y of surplus heat that is technically recoverable from UK sites participating in
the EU Emissions Trading Scheme (ETS). Schepers and van Lieshout (2011) estimate the total
potential of excess heat in The Netherlands at 102.6 PJ/y. The HREII-project gauged the
potential of heat recovery in the European industry, specifically for utilization in ORC systems.
Heat recovery in the cement, steel and glass industries and from gas compressor stations would
allow installation of 2.5GW ORC gross power (HREII, 2013). A report from the US Department of
Energy assumes industrial excess heat availability could amount 20 to 50% of industrial energy
use (U.S. DoE, 2008), but the origin of this figure is not clear. Direct comparison or addition of
available excess heat figures is difficult due to the varying assumptions and scopes applied for
the estimates and because the quality of the excess heat is not always specified. The quality of
the information depends on the level of detail. Most general are figures on the availability of
excess heat, as published by Schepers and van Lieshout (2011) where excess heat sources are
identified. Such information provides no insight about the quality of the heat or opportunities
for recovery. Considered the technically recoverable potential is, as done by Hammond and
Norman (2014), presumably yields lower numbers, and even less when the economic potential
is measured (Hammond & Norman, 2014). Element Energy Ltd. (2014) made an estimate of the
heat potential in the largest UK industrial sites and discusses the important distinction among
the existing amount of excess heat, the technical, the economic and the commercial potential.
The heat sources would amount 48 TWh/y, the technical potential takes account of industrial
heat demand and supply and is estimated at 11 TWh/y (-78%). The economic potential includes
the projects with a positive net annual benefit-cost ratio and amounts 8 TWh/y using a social
real discount rate (3.5%) and 7 TWh/y using a private real rate (10%). The commercial potential
considers projects with a simple payback period smaller than two years and totals 5 TWh/y.
(Element Energy Ltd., 2014)
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
33
Renewable power generation: biomass energy
In 2013, 13.5% of the world Total Primary Energy Supply (TPES) originated from renewable
energy sources. Nearly three quarters (73.4%) of this renewable energy supply, or more than
10% of total TPES, stems from biofuels and waste. This includes liquid biofuels, renewable
municipal waste, biogas and, for the majority, solid biofuels. This dominance of biomass sources
in the total stems from their extensive use for residential heating and cooking in developing
countries (OECD/IEA, 2015). These renewable sources were responsible for 21.7% of the
worldwide electricity production in 2013. The largest renewable electricity source is hydro
energy (16.3%), followed by biofuels and waste (1.7%). All other renewable energy sources
generate 3.7% of the electricity worldwide (OECD/IEA, 2015).
The European Commission’s (EC) Renewable Energy Directive defines biomass as “the
biodegradable fraction of products, waste and residues from biological origin from agriculture
(including vegetal and animal substances), forestry and related industries including fisheries and
aquaculture, as well as the biodegradable fraction of industrial and municipal waste” (EU, 2009,
p. L 140/127). Biomass is a convenient energy source because of its different shapes (solid,
liquid and gaseous), broad field of application (electricity, transport fuels and heat) and the
possibility for some biomass-sources to be stored until required (OECD/IEA, 2012). This storage
allows balancing of electricity supply from intermittent sources such as wind and solar and
signifies the importance of biomass in the future (OECD/IEA, 2012).
Renewable power generation: geothermal energy
Geothermal energy refers to the thermal energy originating from within the Earth, mostly in the
form of hot water or steam (Barbier, 2002; OECD/IEA, 2013). A geothermal resource is defined
as “thermal energy that could reasonably be extracted at costs competitive with other forms of
energy at some specified future time” (Barbier, 2002; Muffler & Cataldi, 1978). Although
geothermal energy is generally classified as renewable energy, the validity of this definition
depends on the extraction rate: only in case the rate of reservoir extraction does not surpass
the replenishment rate the extraction is renewable (Barbier, 2002). Geothermal energy is
appealing because of its vast quantities, low emissions of greenhouse gases (GHG) and
independence of weather and seasonal conditions (OECD/IEA, 2011).
The worldwide installed capacity for geothermal electricity generation rose from 200 MWe in
1950 to 7974 MWe in 2000 (Barbier, 2002; Huttrer, 2000) and 10.7 GWe in 2009 (OECD/IEA,
2011). In 2013, geothermal energy accounted for 3.6% of the world’s renewables supply and
was in that way the third largest renewable energy source, after the biofuels and hydro power
(OECD/IEA, 2013). In 2013, geothermal energy was responsible for 3.6% of renewable energy
supply in the world (OECD/IEA, 2015). Geothermal systems exist in the form of hydrothermal
systems, with a subsurface water reservoir, and Enhanced/Engineered Geothermal Systems
CHAPTER TWO
34
(EGS), where no water reservoir is at hand but the heat is captured from the rocks (OECD/IEA,
2011). Geothermal energy can be utilized directly for heating purposes or indirectly to generate
electricity. Traditional geothermal power plants operate as dry-steam plants or flash power
plants, where the geothermal brine is used directly to operate the cycle. Binary power plants,
such as ORC systems, utilize a secondary fluid to recover the heat from the geothermal well and
operate the cycle and are suitable for lower temperature wells (< 150 °C) (DiPippo, 1999;
OECD/IEA, 2011; Walraven, 2014).
Renewable power generation: solar energy
With 885 million TWh per year radiating onto the Earth’s surface, solar energy is the most
abundant energy source (OECD/IEA, 2014). Still, it represents only a minor part of energy supply
in the world: 1.3 % of TPES in 2013 originated from geothermal, wind, solar and tidal sources
together; only 2.2 % of the world renewable energy supply is delivered by the combination of
solar and tidal sources (OECD/IEA, 2015). Nevertheless, the growth of solar energy use has been
significant at an average rate of 46.6% for solar photovoltaic (PV) and 12.3% for solar thermal
energy between 1990 and 2013 (OECD/IEA, 2015). Solar thermal electricity generation plants
emit no greenhouse gases and are inherently capable of thermal energy storage. Concentrating
solar power (CSP) technology exists in the form of parabolic troughs (PT), linear Fresnel
reflectors (LFR), central receiver systems (CRS) or towers and parabolic dishes (Mills, 2004;
OECD/IEA, 2014). These systems are commonly coupled to a steam cycle for electricity
production.
2.3. The costs of ORC technology: a literature review
The merits of ORC technology are well-understood: due to its ability to process lower
temperature sources and at a smaller scale, the technology is particularly suitable for use in
renewable energy or energy efficiency applications. Because it is not just the technical but also
the economic feasibility that determines the extent to which a technology is adopted, the focus
in this section lies on the economic characteristics of ORC technology. At the basis of financial-
economic project analyses are the annual cash flows, composed of the capital investment and
annual expenses and revenues. The capital investment is the one-time cost occurring at the
beginning of the project. It includes the costs directly associated with the system (equipment,
materials, labour etc. required for the equipment and the installation thereof), indirect costs
(engineering, construction costs and contingencies) and other outlays (such as start-up costs,
working capital, etc.) (Bejan, Tsatsaronis, & Moran, 1996). At the core of an ORC project’s
capital investment are the components of the ORC module itself: evaporator, expander and
generator, condenser and pump. The costs for integration of the ORC module into an existing
plant (e.g., for heat recovery applications) or with the equipment necessary for the system’s fuel
supply vary according to the type of application. The annual expenses for ORC projects depend
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
35
on the type of application and the location, but are generally of lesser importance than the
investment costs. The main goal of this chapter is therefore to provide a comprehensive
overview of the state-of-the-art knowledge on the economics of ORC systems. The emphasis lies
on the investment costs, or capital investment, because they constitute the core of the ORC
system and are closely associated with the technological setup. The economics of ORC
technology have been reviewed previously by Vélez et al. (2012) and Sylvain Quoilin et al.
(2013), but the origin and scope of the data is not always specified in their work. This chapter
extends their insights on ORC economics by reviewing the literature more comprehensively and
providing a categorization to classify the origin and scope of the published figures.
2.3.1. Methodology
The Web of Science includes 1904 publications on ORC technology (mid-July 2016), of which
only about 6 % investigate the economic aspects. These studies are very diverse because they
concern different technical setups, use various heat sources and apply other scopes to delineate
the system under consideration. This complicates comparison and bears the risk that economic
numbers are misinterpreted. Hence, the review of the literature is structured according to three
defined criteria: the application scope, the system scope and the economic scope (cfr. section
1.4). The application scope refers to the heat source used to operate the ORC system: excess
heat, geothermal, biomass or solar energy. The system scope refers to the delineation of the
ORC system itself, comprising either the ORC module itself or the complete ORC project. Due to
the varying approaches applied in the investigated literature it is not always possible to execute
this system scope classification in a perfectly unambiguous manner. Finally, the economic scope
refers to the origin of the cost data: real ORC costs, price quotes from a manufacturer, costs
estimated using various techniques, or assumed ORC costs. The subsequent sections discuss the
literature on ORC costs, each focussing on one particular heat source.
2.3.2. The costs of ORC technology: heat recovery applications
The advantages of excess heat recovery are increasingly recognized and ORC technology has
proven its suitability in the field: more than one third of the ORC commissioned or under
construction worldwide is applied to recover excess heat, corresponding to 395 MW installed
capacity.1 Reports on the practical implementation of ORC systems for excess heat recovery are
published by e.g., Bronicki (Sine Dato), who discusses a 6.5 MW plant in Gold Creek, Canada and
a 1.5 MW system at the Heidelberger Zement AG Plant in Lengfurt, Germany. Moreover, Brasz,
Biederman, and Holdmann (2005) give insight in three operating plants using an ORC based on
HVAC (heating, ventilation and air conditioning) equipment. From the investigated literature,
there are 48 papers that discuss the economics of ORC technology (Table 2.2). The details of the
1 Data collected in July 2016, more details on the ORC market distribution are discussed in chapter 5.
CHAPTER TWO
36
literature review are discussed in the subsequent sections and a summary is provided in
Appendix A.
Table 2.2. Amount of heat recovery ORC publications according to system scope and economic scope.
System scope Module Project
Economic scope
Real 1 2
Quote 6 4
Estimated 4 27*
Assumed 1 3 Legend: The table summarizes the literature review by providing the number of publications according to the scope of the ORC
system discussed in the publication and the scope of the economic data provided. *Of which 18 give results for the estimated SIC,
4 estimate the costs but give no numerical results and 5 estimate no costs but use heat exchanger area as proxy).
Heat recovery ORC applications: capital costs of real systems
Detailed economic information on commissioned heat recovery ORC systems is fairly scarce.
Leslie, Sweetser, Zimron, and Stovall (2009) discussed the results of a demonstration project
where excess heat is recovered from a gas turbine, near St. Anthony, North Dakota, USA. The
5.5 MWgr ORC uses pentane as working medium and the project had a capital cost of about
2500 $/kW. The annual operating and maintenance costs are projected at about 200 k$. The net
present value (NPV) and internal rate of return (IRR) are calculated for varying contract
durations and costs of capital and range, respectively, between 2 and 12 M$, and between 5%
and 15% (Leslie et al., 2009). Finally, Tumen Ozdil and Segmen (2016) performed an
exergoeconomic analysis for an existing ORC plant in Adana, Turkey, with an installed capacity
of 260.4 kWgr. The ORC module costed 500 k$, but details on the commissioning period of the
system were not revealed. The turbine was responsible for the largest share in the costs (300
k$), followed by the evaporator (100 k$), the condenser (75 k$) and the pump (25 k$) (Tumen
Ozdil & Segmen, 2016). Finally, Tchanche et al. (2010) built a 2 kW ORC prototype for laboratory
experimentation, for total installed cost of 11.55 k€.
Heat recovery ORC applications: representative manufacturer’s quotes
Publications that discuss the costs of real ORC systems remain relatively scarce. The majority of
information stems from budget quotes published by manufacturers of ORC systems. But the
application of ORCs in industrial heat recovery settings has been discussed since the early days
of the technology’s implementation. Gaia and Macchi (1981) experimented with heat recovery
from a ceramic tunnel oven. The ORC system had a designed maximum power output of 40 kW,
but reached only 22 kW in practice. The ORC module costs were estimated at approximately
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
37
500 ECU2/kW, with a 3-year payback period. Shortly thereafter, Angelino, Gaia, and Macchi
(1984) published the experimental results for a 100 kWnet heat recovery ORC. The project costs
are projected at 92,000 ECU, assuming a series production of 10 units per year. More
contemporary manufacturer insights regarding heat recovery ORCs are provided by Vescovo
(2009), who discussed three potential industrial applications and their representative
investment costs. A project in the cement industry would involve recuperating the excess kiln
gas and the clinker cooler air flow. Recovering the heat via thermal oil circuits for delivery to the
ORC process, the total project costs were projected at 4.8 M€ for 1.6 MWe,net power output, of
which 1.8 M€ for the ORC module. In the glass industry, there are opportunities to recover the
heat from the combustion gas exiting the oven. The projected costs are somewhat lower than
for the cement industry project: 2.6 M€ for 1 MWe,net power output, with half of these costs for
the ORC module. Finally, the steel industry has many excess heat sources. The project suggested
for this case was the cheapest, with a total cost of 4.3 M€ (of which 2.4 M€ for the ORC module)
for 2.4 MWe,net output. A similar work was presented by Forni, Vaccari, Di Santo, Rossetti, and
Baresi (2012) for four industrial ORC applications: heat recovery from a gas turbine, from kiln
and clinker cooling gas, from a melting furnace and from a reheating furnace in an iron and steel
rolling mill. The case of the gas turbine has the largest gross power output (5.4 MW) as well as
the lowest specific investment costs (SIC) (2593 €/kW for the complete project and 1148 €/kW
for the ORC module). The kiln and clinker cooling gas project is the most expensive (5.3 MW for
3320 €/kW), followed by the melting furnace project a (1.3 MW for 2920 €/kW) and the iron
and steel rolling mill project (1.1 MW for 2820 €/kW). Although all the previous papers discuss
ex-ante projected costs rather than the actual costs of commissioned systems, they are
considered representative because they are provided by the manufacturer of the system.
Indirect insight in ORC manufacturer costs is provided by e.g., David, Michel, and Sanchez
(2011). They discuss two potential heat recovery projects involving an ORC system. The first
case study concerns a cokes plant emitting flue gas with 2.5 MWth recoverable thermal energy.
The heat could be recovered by two 125 kWe,gr ORC units, for a projected total investment cost
of 1080 k€ or a specific cost of 4320 €/kWe,gr. In the second case the ORC would produce 160
kWe gross from heat recuperated from a biogas engine. This project’s specific investment costs
are projected lower than for the first case: 2594 €/kWe,gr or 540 k€ in total, of which 415 k€
stems from the ORC module. Furthermore, Ghirardo, Santin, Traverso, and Massardo (2011)
obtained a price quote from a manufacturer: approximately 1675 k€ for an ORC module with an
installed capacity of 1115 kW. Finally, Vanslambrouck, Vankeirsbilck, Gusev, and De Paepe
(2011) gathered a vast knowledge base on the practical implementation and economics of ORC
technology. Based on their experience, the authors publish purchase prices of low temperature
2 ECU = European Currency Unit. This unit was the common monetary unit of the European Communities,
composed as a basket of currencies from the Member States. It in use was from 1979 until 1999, when the Euro was adopted, and existed only in electronic form. (Eurostat, 2016)
CHAPTER TWO
38
ORC modules in the range between 1350 €/kWe for 250 kW capacity and 2200 €/kWe for 50
kWe. High temperature ORC modules have prices between 1000 €/kWe for 2 MW output, 2000
€/kWe for 500 kW and 3000 €/kW for 150 kW of installed capacity. However, the authors
provide no insight in the time at which the data was collected (Vanslambrouck et al., 2011).
Heat recovery ORC applications: estimated capital costs
The large majority of researchers discuss the economics of ORCs, used in heat recovery
applications, on the basis of estimated costs. There are plenty of excess heat sources that can
be used in combination with ORC technology. Moreover, the literature review shows that often
similar approaches are used to estimate the costs of the system under consideration. Therefore,
the subsequent overview is grouped according to the cost estimation approach applied.
Scaling methods
One approach is to estimate the costs of an ORC system by applying a scaling factor to the
known costs of another system. This method is utilized by e.g., Tchanche et al. (2010). They
constructed a small-scale ORC prototype in their laboratory, with 2 kW capacity and a total
installed cost summed at 11.55 k€. The costs of the components were obtained from suppliers
and used to estimate the costs of a 50 kWnet ORC system via extrapolation, resulting at 3034
€/kW (Tchanche et al., 2010). Similarly, Ghirardo et al. (2011) compared different technologies
to recover heat from an on-board solid oxide fuel cell. The price quote for a 1115 kW module
lies at the basis of the cost estimate. The costs for integration of the module are added, and the
costs of a 35 kW project are estimated via scaling, at 117 k€ (Ghirardo et al., 2011). Finally,
Walsh and Thornley (2013) compare an ORC system and a condensing boiler in terms of lifecycle
greenhouse gas reductions and discounted payback period. The costs of the 2 MWnet ORC
project have been estimated at 2023 €/kW (including installation, engineering and material).
The authors used a power law relationship for the estimate, but the details were not specified
(Walsh & Thornley, 2013).
Component correlations: ORC module costs
Another approach consists of estimating the costs of the ORC module at the hand of
correlations that link the costs of a component with its size or capacity. For instance, Lee, Kuo,
Chien, and Shih (1988) used component cost correlations to estimate the purchased equipment
costs of a 1108 kWgr ORC module. The module includes preheater, evaporator, regenerator,
condenser, turbine, generator, cooling tower and ORC pump and is valued at 1.3 M$ or 1163
$/kW (Lee et al., 1988).3 More recently, F. Yang et al. (2015) investigated the thermodynamic
and economic performance of an ORC system for heat recovery from a diesel engine, using
3 The value reported in Lee et al. (1988) is $ 1,090,676. However, the component costs sum to $ 1,288,722.
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
39
different working fluids. The ORC modules have designed power outputs around 10 kWnet and
the costs, estimated using component correlations, range between 502 k$ and 537 k$. The best
performance was measured for an ORC module with r245fa as working fluid, a 11.55 kWnet
output and a capital cost of 533 k$ (F. Yang et al., 2015). In the same way, M.-H. Yang (2015)
analysed a compact transcritical ORC (TCORC) for heat recovery in a marine diesel engine. The
costs of the major components are accounted for in the total cost estimation. The results range
between 3582 k$ and 4678 k$ for power outputs between 2132 kWnet and 3049 kWnet.
Component correlations: project
Many authors use a similar approach to estimate the costs of complete ORC projects. The total
costs of the project are then estimated based on the costs of the components. For instance,
Sylvain Quoilin, Declaye, Tchanche, and Lemort (2011) optimize an ORC for heat recovery
applications from a thermodynamic and an economic perspective, assessing the suitability of
multiple working fluids. The component costs are estimated with correlations from the
literature as well as correlations fitted by the authors for Belgian prices and the total project
costs are obtained by adding the labour costs. The project SIC are estimated between 2136
€/kW and 4260 €/kW, depending on the fluid operated, for small scale (< 5 Wnet) systems. An
important conclusion is that the operating point yielding maximum power does not coincide
with that of minimal SIC (Sylvain Quoilin et al., 2011). In the same way, Lecompte, Huisseune,
van den Broek, De Schampheleire, and De Paepe (2013) study an ORC coupled to a stationary
internal combustion engine for cogeneration: the excess heat is primarily used for heating; the
remainder serves as input for the ORC system. The thermo-economic optimization is performed
for different working fluids. Under optimal working conditions, the 182.3 kWnet ORC system
using R125a as working fluid has the lowest SIC (2862 €/kWe). A system operating on R123yf has
a lower power output (163.1 kWnet) and higher SIC (3072 €/kWe), whereas using R245fa gives a
higher output (206.7 kWnet) but also higher SIC (4028 €/kWe). However, the authors point out
that the actual operating conditions do not necessarily correspond to the optimized ones. For
instance, an optimized ORC with capacity 2078 kWnet had a SIC of 1870 €/kW. However, when
part-load conditions are taken into account the power output decreases and the actual SIC
increases to 2360 €/kW (Lecompte et al., 2013). In a subsequent work, Lecompte, Lazova, van
den Broek, and De Paepe (2014) compare a subcritical ORC (SCORC) and transcritical ORC
(TCORC) for heat recovery applications. The optimization is performed for different heat carrier
inlet temperatures and showed that the investment costs of the transcritical cycle may be as
much as 50% higher than those of the subcritical cycle for the same power output. For power
outputs between approximately 50 kWnet and 200 kWnet, the SIC for the SCORC module were
optimized between 1641 €/kW and 3745 €/kW for different heat inlet temperatures, whereas
those for the TCORC range from 2262 €/kW to 6045 €/kW. Finally, Lecompte, Lemmens,
Huisseune, van den Broek, and De Paepe (2015) compare a SCORC with a TCORC but support
CHAPTER TWO
40
the evaluations with a financial appraisal. Optimizing for minimum SIC gives 4114 €/kWe for a
681.8 kWe,net subcritical cycle and 5044 €/kWe for a 681.3 kWe,net transcritical cycle. Maximizing
the power output of the subcritical cycle is obtained at 791.5 kWe,net with a SIC of 4707 €/kWe
and an output of 1040 kWe,net for the transcritical cycle costing 8137 €/kWe. The maximum
power output of the transcritical cycle increases by 31.5% compared to the subcritical cycle, but
at the same time the SIC increases by 72.8% (Lecompte, Lemmens, et al., 2015).
Other authors estimating the costs of ORC projects using component cost correlations include
Pierobon, Nguyen, Larsen, Haglind, and Elmegaard (2013). They propose a methodology for ORC
design and apply it for a case of heat recuperation from a gas turbine of an offshore platform in
Kristiansund, The North Sea. The authors optimize for thermal efficiency and NPV, but also for
the total volume of the ORC. The purchased equipment costs of each piece of equipment are
estimated using correlations collected in the literature, the total investment costs are obtained
by applying a factor 3.7 to the sum of the purchased equipment costs. The results are displayed
for two specific cases. The first one entails a 30 m³ volume limit. The optimized system has a net
power output of 6.04 MWe,net and the total investment costs (13.1 M$) are for a large share
represented by the axial turbine (78.6%). The second case represents the situation where the
available volume is greater than 100m³ and the optimum is defined at the point of maximum
NPV. The net power output in this case amounts 6.43 MWe,net, the total investment costs sum to
15 M$. The NPV amounts 20.1 M$ and the payback period 5.2 years. Also Le, Kheiri, Feidt, and
Pelloux-Prayer (2014) apply component cost correlations. They optimize a subcritical ORC,
maximizing the exergy efficiency as well as minimizing the levelized cost of electricity (LCOE).
When optimizing for the lowest LCOE, the best performance was estimated for the system
operating on pentane, with a capacity of 1624 kWnet and a SIC of 3184 $/kW (Le et al., 2014).
Similarly, Kwak, Binns, and Kim (2014) compare four different options to recover excess heat
from an industrial plant. The authors analyze an ideal scenario with a fixed heat source
temperature and a more realistic case where the heat is available at varying temperatures,
considering different operating conditions for winter and summer. The investment costs of the
ORC are estimated at 27.43 M$ for the fixed-temperature scenario, while the total plant output
increases from 39.6 MW to 65.12 MW. In the varying-temperature winter scenario the output
increases with 6.69 MW at a cost of 10.72 M$ and in the varying-temperature summer scenario
the output augments by 8.64 MW for and 15.96 M$ (Kwak et al., 2014). Amini et al. (2015) study
a transcritical CO2 ORC cycle for heat recovery from a combined-cycle power plant. The thermo-
economic analysis involved an optimization of the benefit-cost ratio. The optimized plant has a
turbine power output of 8220 kW, a net power output of 4040 kWe,net and the specific
investment costs amount 2625 $/kW (Amini et al., 2015). M.-H. Yang and Yeh (2015) optimize
an ORC for heat recovery from large marine diesel engines, using different working fluids.
However, the cost optimization does not involve SIC minimization, but a maximization of the net
power output index (NPI), defined as the ratio of the net power output over the total cost of the
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
41
system. The maximum NPI was obtained for the working fluid R1234yf, with a net power output
of 320.2 kWe,net and a total cost estimated at 1.2 M$. Di Maria and Micale (2015) investigate the
use of ORC technology to recovery exhaust air from the aerobic treatment of organic waste. A
design with a net power output of 19.4 kWnet had an estimated total project cost of 55.7 k€, or
2873 €/kW (Di Maria & Micale, 2015). Finally, Heberle and Brüggemann (2016) perform a
thermo-economic analysis for a heat recovery ORC system. Using component correlations, the
costs of the major components are estimated between 1162 €/kWnet and 1336 €/kWnet for
power outputs between 331 kWgr and 388 kWgr. The total project costs are estimated using
several techniques and multiplication factors, which leads to diverging results.
Estimation method not clarified
Some authors estimate the costs of the ORC system under investigation, but give no insight in
the estimation method used. For instance, Najjar and Radhwan (1988) investigated the use of
an ORC for cogeneration together with a gas turbine. The authors estimated the additional costs
of the 74 kW ORC system at 43,800 £ including installation, but the estimation method was not
clarified (Najjar & Radhwan, 1988). In the same way, Kalina (2011) investigates the performance
of an ORC system for heat recovery in a biomass gasification internal combustion engine,
considering different system setups. One configuration involves a thermal oil circuit to recover
the heat, which would imply an ORC power output of 53.21 kWnet, with the total investment
costs estimated at 1149 k$. An alternative is a double cascade system configuration, where the
power output is gauged at 40 kWnet from the upper cycle and 31 kWnet from the lower cycle,
with its costs totalling 1220 k$ (Kalina, 2011). Finally, Law, Harvey, and Reay (2013) compared a
high-temperature heat pump and an ORC system for waste heat recovery from the same heat
source. The ORC system has a net power output of 119 kWnet and the capital costs are
estimated between 107 and 250 k£. The ORC system is recommended over the heat pump
because of superior economic and environmental performance (Law et al., 2013).
Key characteristics missing
Furthermore, there are authors who estimate the costs of the discussed ORC system, but do not
disclose some key characteristics of their system. For instance, Lukawski (2009) estimates the
capital costs for two geothermal ORC projects and two operating on excess heat. The purchased
equipment costs of the ORC are calculated using component correlations, whereas the
remaining integration requirements are gauged using multipliers. The optimized results gave a
SIC of approximately 2100 $/kW for the heat recovery applications, calculated for the reference
year 2008, but the power output of the system is unclear (Lukawski, 2009). Likewise, Imran et
al. (2014) perform a thermo-economic optimization to compare different ORC cycle setups. The
economic modelling considers solely the costs of the components; other costs such as piping
working fluid and labour are not included. The obtained SICs in the 3274 to 4155 €/kW range for
CHAPTER TWO
42
the basic ORC, between 3453 and 4571 €/kW for the single stage regenerative ORC and
between 3739 and 4960 €/kW for the double stage regenerative ORC, depending on the
working fluid operated. However, no indication was given on the power range (Imran et al.,
2014).
Numerical results cost estimate not provided
Other authors estimate the costs of the analysed ORC system, but provide no numerical results
for the estimated (specific) investment costs. For instance, Papadopoulos, Stijepovic, and Linke
(2010) present a systematic design and selection of optimal working fluids for ORCs. The
authors assume that more than 90% of an ORC’s capital costs stem from the evaporator and the
condenser, but the origin of this statement is not clarified. The equipment costs of these two
components are used as proxy for the total and the goal is to define the heat exchanger area
that combines maximum energy recovery with minimum capital cost. The ASPEN software is
used for the estimation, but no numerical results are provided (Papadopoulos et al., 2010).
Similarly, Cayer, Galanis, and Nesreddine (2010) investigate a transcritical ORC for low
temperature heat recovery. The performance criteria for the system include thermal efficiency,
specific net output, exergetic efficiency, surface of the heat exchangers, relative cost of the
system and total UA, defined as the product of the heat transfer coefficient and the heat
transfer area. The results are displayed graphically (Cayer et al., 2010). Grabińsky (2011)
investigates the feasibility of an ORC system to recover the excess heat from a bioliquid plant.
The financial feasibility of the designs is discussed in detail, but the numerical values of the
investment costs are not provided (Grabińsky, 2011). Finally, Meinel, Wieland, and Spliethoff
(2014) investigate ORC systems for excess heat recovery from a biomass digestion plant internal
combustion engine. The authors compare a standard ORC cycle, a recuperator cycle and a two-
stage regenerative pre-heating cycle. The purchased equipment costs of the components are
estimated using component correlations and the results are presented in €/kWh. Compared to
the German electricity sales price, the authors conclude that the heat source must have a
thermal power of minimum 1 MWth (Meinel et al., 2014).
Economics assessed with non-economic measures
Lastly, some authors assess the economics of ORC technology via performance measures that
are in fact not economic in nature. For instance, Z. Q. Wang, Zhou, Guo, and Wang (2012)
assume that 80 to 90 % of the system’s capital costs stem from the heat exchangers (referring
to the statement made by Papadopoulos et al. (2010)), and therefore optimize their ORC system
by minimizing the ratio of the heat exchanger area and the net power output (Z. Q. Wang et al.,
2012). Y.-R. Li, Wang, and Du (2012) assess the cost-effectiveness of the ORC system by the
inverse of the ratio used by Z. Q. Wang et al. (2012): the net power output over the total heat
transfer area is taken as proxy for the investment costs. The reasoning behind this statement or
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
43
the practical (economic) implications of this metric is not clarified by the authors (Y.-R. Li et al.,
2012). Also D. Wang, Ling, and Peng (2013) use the ratio of the net power output over the heat
transfer area to measure the cost-effectiveness of using various working fluids. Only the
evaporator and condenser heat transfer areas are considered. The costs of other components
are not accounted for, based on the assumption that 80-90% of the system costs stem from the
heat exchangers (D. Wang et al., 2013). In the same way, J. Wang, Yan, Wang, Ma, and Dai
(2013) perform a thermodynamic optimization of an ORC system for low-grade waste heat
recovery, measuring the economic performance by the ratio of the net power output and the
total heat transfer area (J. Wang et al., 2013). Most recently, Wu, Zhu, and Yu (2016) assess
different working fluids for ORC systems, using the ratio of the net power output and the UA of
the evaporator and the UA (Wu et al., 2016).
Heat recovery ORC applications: assumed capital costs
Finally, several authors discuss the costs and financial feasibility of ORC systems using capital
costs that are neither real nor estimated, but merely assumed to perform the analysis. For
instance, Yari and Mahmoudi (2010) compare the simple ORC, the ORC with internal heat
exchanger and the regenerative ORC for heat recovery from a gas turbine-modular helium
reactor. The ORC SIC is assumed to be between 500 and 2500 $/kW and the total capital
investment is taken as 6.32 times the purchased equipment costs. Arvay et al. (2011) discuss the
application of ORC technology at two potential sites. An ORC for heat recovery from a
biopolymer plant would deliver an electric output of 250 kW, at an assumed cost of 1500 $/kW
for the ORC and another 1500 $/kW for the installation. Another case concerns a cogeneration
facility where the ORC could be installed as bottoming cycle. The output is gauged at 1.25 MW,
for a total installed cost of 3.45 M$ (Arvay et al., 2011). Schuster, Karellas, Kakaras, and
Spliethoff (2009) discuss potential applications for ORC technology. For the case of a biogas
digestion plant, an ORC system could be used to recover the excess heat at an assumed SIC of
3755 €/kWe for a 35 kW project (Schuster et al., 2009).
Heat recovery ORC applications: an overview of capital costs
Excess heat is generally understood as the residual thermal energy that is unavoidably
generated as by-product in many industrial processes. Depending on the type of the heat flow
and the local conditions, this energy could be recovered and put to use in the form of heat or,
after transformation, to meet cooling or power demands. Seeing the wide variation in industrial
processes, the list of potential excess heat sources is vast. This diversity is reflected in the
research on ORC technology applied in heat recovery settings. Common heat sources under
investigation include processes in energy intensive sectors such as the glass, steel or cement
industry, but also exhausts from internal combustion engines or gas turbines are among the
options. Figure 2.1 displays a summary of the investment costs for heat recovery ORC systems,
CHAPTER TWO
44
as collected with the literature review: the specific investment costs are presented as function
of the installed capacity of the system. To allow for comparison, all cost figures were converted
to Euros and actualized to 2015 values using the Chemical Engineering Plant Cost Index (CEPCI).
Before interpreting this graph, several remarks have to be taken into account. Firstly, several
papers provide the results in terms of net power output of the system, whereas others consider
the gross installed capacity. In this summary, the gross installed capacity is used where
available, and in the other cases the net power output is used as proxy. Secondly, the large
majority of the papers did not attach a date stamp to their cost figures. Economic values are
influenced by inflation and should therefore be accompanied by their time frame to allow
interpretation. For the literature cases without explicit date stamp, the publication date was
taken as proxy. Thirdly, the records included in Figure 2.1 stem from papers that apply different
system scopes, which means that the components included in one module’s costs may differ
from that of another, and the integration requirements depend strongly on the application at
hand. Finally, the estimated costs have been attained using a wide variety of estimation
methods. These remarks hinder valid comparison between the cost figures, so that the
summary graph should be interpreted in a rather indicative way, and not as a summary of true
system costs.
Nevertheless, Figure 2.1 shows a dramatically wide variation in the published costs and quotes
of heat recovery ORC systems. The real ORC projects have specific investment costs between
5775 €2015/kW for a 2 kW laboratory prototype and 1993 €2015/kW for a 5.5 MW ORC used for
heat recovery from a gas turbine. The price quotes refer to ORC systems from different
manufacturers and vary between 717 €2015/kW for a 6 MW and 5200 €2015/kW for a 500 kW
exhaust gas recovery project. Overall, the collection of costs and quotes suggests a slightly
inverse relation between the SIC and to the installed capacity of the system. However, the
majority of the costs figures for heat recovery ORCs have been estimated. The literature review
reveals there exist various approaches to estimate the costs of a system, from scaling methods
to bottom-up cost estimation with component-specific cost correlations. This variety of
approaches may explain the wide diversity in the obtained results. Figure 2.1 shows small scale
projects (< 10 kW) with a SIC around 3000 €2015/kW, but at the same time projects of 400 kW
and of 1 MW with costs around 8000 €2015/kW. The heat recovery modules likewise show very
diverging specific investment costs. Based on these estimated cost data, no conclusions can be
drawn about the economics of heat recovery ORCs. Finally, simply assuming ORC costs is
precarious because of all the factors that can influence the specific project under study, but the
costs assumed in the literature are in line with the rest of the figures.
The capital investment is at the core of the ORC economic evaluation. It is determined by the
technical setup of the system itself and has a large influence on the financial feasibility if the
investment. Nevertheless, when the goal is to compare electricity generation technologies
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
45
among each other, the criterion of relevance is the cost at which the system can generate
electricity. A commonly used metric to compare the competitiveness of electricity generation
technologies is the levelized cost of electricity (LCOE). The LCOE relates the expenses for an
energy project to its revenues, expressed in terms of electricity production. This yields a cost per
MWh, often interpreted as the cost at which the generated electricity must be sold for the
project to break even over its lifetime. Only few of the studied papers investigated the LCOE of
the system under consideration. The electricity costs of the heat recovery projects are
estimated between 51 and 134 €2015/kWh, but are mostly situated around 103 €2015/kWh. There
is only one study that calculated the LCOE for a heat recovery ORC module, between 22 and 24
€2015/kWh. Note that these LCOEs have been calculated using diverging project assumptions,
such as the annual load hours or the discount rate.
Figure 2.1. Investment costs for heat recovery ORC systems, as published in the literature.
Legend: Overview of the capital costs for heat recovery ORC projects (P) and modules (M), as retrieved by the literature review.
The overview distinguishes between the costs of real systems, representative quotes from ORC manufacturers, estimated costs
and assumed costs.
2.3.3. The costs of ORC technology: biomass applications
Biomass sources are particularly interesting for small-scale applications in which conventional
plants are too complex and expensive (Algieri & Morrone, 2012). Moreover, according to
Schuster et al. (2009) ORC plants are the only proven technology for power production from
biomass in decentralized applications smaller than 1 MWe. The suitability of biomass sources for
power generation with ORC technology is confirmed by the 282 operational and 53 additionally
sold biomass ORC systems (anno July 2016). This corresponds to 46% of all ORCs worldwide and
makes biomass the most-used heat source for ORC applications. The practical implementation
of biomass ORC systems has been discussed by e.g., Erhart, Strzalka, Eicker, and Infield (2011)
who discuss the monitoring results of a biomass ORC CHP plant coupled to the district heating
100
1.000
10.000
100.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Real
M Real
P Quote
M Quote
P Estimated
M Estimated
P Assumed
M Assumed
CHAPTER TWO
46
network in Ostfildern, Germany. Stoppato (2012) studies the optimal operation modus for a
1255 kWe,net cogeneration ORC plant installed in Assiago, Italy. The literature review discloses 28
publications that discuss the economics of biomass ORC systems (see Table 2.3 and Appendix
A).
Table 2.3. Amount of biomass ORC publications according to system and economic scope.
System scope Module Project
Economic scope
Real 1 7
Quote 2 3
Estimated - 6
Assumed 2 7 Legend: The table summarizes the literature review by providing the number of publications according to the scope of the ORC
system discussed in the publication and the scope of the economic data provided.
Biomass ORC applications: capital costs of real systems
The first biomass ORC system was installed in 1999 in Admont, Austria, with a capacity of 400
kWe,gr and a project investment cost of 3,2 M€, including monitoring and dissemination. It was
undertaken as a demonstration project and operates as a cogeneration system, where the heat
is delivered to on-site users and to a monastery via a district heating network. The majority of
the annual operational costs of 381 k€/y stems from the biomass fuel (70%) and the annual
income of 620 k€/y is generated from heat (75%) and electricity sales (25%) (STIA - Holzindustrie
Ges.m.b.H., 2001). A follow-up project was commissioned in 2001 in Lienz, Austria. The installed
capacity was more than doubled compared to the first demonstration project in Admont while
the specific investment costs (SIC), i.e. the costs per kW installed capacity, were reduced. The
1000 kWe,gr ORC system had an investment cost of 2765 k€, of which 1360 k€ was attributed to
the ORC module. The ORC system operates on biomass from the word industry and forestry in
the region and is used for electricity generation and heat supply to a district heating network. (I.
Obernberger, Carlsen, & Biedermann, 2003; Ingwald Obernberger, Thonhofer, & Reisenhofer,
2002)
The demonstration projects established the feasibility of ORC technology in combination with
biomass energy. More biomass ORCs are continuously installed since then and several authors
have reported on their practical and economic matters. For instance, Barz (2008) discusses a
550 kWe biomass cogeneration ORC installed in 2001 in Friedland, Germany. The investment
costs were relatively low (1.25 M€) because the system was connected to an existing district
heating plant and the operation follows the community’s heating demand. Bini et al. (2004)
discuss a 1.1 MWe,gr cogeneration ORC installed in Tirano, Italy, with an investment cost of 1.4
M€ after deduction of a 33% investment grant (hence an investment of 2.1 M€). In comparison
to a ‘heat only’ hot boiler district heating system, the addition of the ORC was found to have a
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
47
positive economic impact. When the ORC follows electricity demand and the ORC excess heat is
dissipated, the economic evaluation depends on the remuneration from green certificates (Bini
et al., 2004). Other biomass ORC systems of which the costs are reported are a 1 MW ORC
installed in Scharnhauser Park, Stuttgart-Ostfildern, Germany (6 M€ project costs) (Tańczuk &
Ulbrich, 2013) and a 1.13 MWgr ORC with a project SIC of 4630 €/kW of which the location is
unknown (Bolhar-Nordenkampf, Pröll, Aichernig, & Hofbauer, 2004). Finally, a 1.5 MW
cogeneration ORC project was commissioned in 2005 in Salzburg, Austria in a wood processing
plant for an investment cost of 3 M€ (Duvia, Bini, Spanring, & Portenkirchner, 2007).
Biomass ORC applications: representative manufacturer’s quotes
Other publications do not discuss the costs of commissioned ORCs, but give quotes for the
investment costs of potential applications. For instance, Peretti (2008) discusses the application
of ORC technology for cogeneration in sawmills. The total investment costs of three ORC
projects of different sizes are calculated by multiplying the known costs of the modules with a
factor 2.75 and adding 100 k€ for the grid connection. This gives an investment cost of 2.8 M€
for a 538 kWe,net ORC project, 3.3 M€ for one with a capacity of 1155 kWe,net and 5.4 M€ for a
2079 kWe,net project. Based on the assumptions made for the analysis, ORC systems with an
installed power larger than 1.5 MW are found to have a beneficial financial assessment (Peretti,
2008). A similar analysis is discussed by Duvia and Tavolo (2008) for applications in the pellet
production industry. The multiplication factor applied to estimate the total project costs from
the ORC module costs depends on the type of dryer used for drying in the pellet process. When
belt dryers are used, the project costs range between 982 k€ for a 617 kWe,gross system and 1930
k€ for one with 2282 kWe,gross installed capacity. The costs of systems with rotary dryers are only
slightly higher (Duvia & Tavolo, 2008). Similarly, Duvia, Guercio, and Rossi di Schio (2009) discuss
the investment costs for various biomass cogeneration plants for district heating. The
approximate prices of seven ORC modules are listed and range from 4500 €/kWe for a 1803
kWe,net system to 10.2 k€/kWe for a 345 kWe,net system. Although the numbers in these
publications do not refer to commissioned ORC system, they can be considered representative
since they are published by the ORC manufacturer itself.
Biomass ORC applications: estimated capital costs
Other studies presenting insights in the economics of ORC technology do not use the costs of
commissioned ORC systems as basis, but estimate the capital costs of the system under
consideration. For instance, Chinese, Meneghetti, and Nardin (2004) investigate the use of ORC
technology in the wood manufacturing industry and compare three scenarios (heat only, ORC in
heat-driven mode and ORC in grid-driven mode). The investment costs have been obtained by
interpolation of data obtained from literature and manufacturer data. A refurbishment of an
existing biomass boiler together with an ORC for cogeneration would cost 765 k€ for 4 MW
CHAPTER TWO
48
output. An ORC and a new biomass boiler is rated at 838 k€ for 3.6 MW in an existing building
and at 886 k€ for 3.4 MW output on an available land plot. A distributed heat generation system
is concluded to be the preferred option in case of sufficient heat demand and whereas ORC
systems could be unaffordable for single firms they can be interesting when connected to a
district heating network where firms cooperate (Chinese et al., 2004). Uris, Linares, and Arenas
(2014) investigate a biomass cogeneration ORC plant. The performance of a basic and a
recuperative cycle has been compared under both subcritical and supercritical conditions. The
plants have been studied for 1 MWe and 2 MWe sizes. The investment costs for the 1 MWe
system, including biomass boiler, ORC system and auxiliaries but excluding the district heating
network, have been adopted from the 1 MWe plant in Lienz (Ingwald Obernberger et al., 2002).
Updated to 2013 values, the total project costs 7.6 M€ for the subcritical and 7 M€ for the
supercritical ORC with 1 MWe output. The costs for the 2 MWe have been obtained by applying
a 0.6 scaling factor and amount 11.5 M€ for the subcritical and 10.6 M€ for the supercritical
system (Uris et al., 2014).
A more meticulous cost estimation is performed by Huang, Wang, et al. (2013), who investigate
biomass-driven trigeneration ORC systems for buildings applications. The performance is
assessed for three different types of biomass input (willow chips, straw and rice husk) and the
operation of the system is modelled following the electric load. The techno-economic
assessment is performed using cost-engineering software, for power-only cogeneration and
trigeneration modes. For all three biomass sources, the economic performance is the best in
cogeneration mode, slightly worse in trigeneration mode and significantly worse in power-only
mode. The total investment costs are estimated between 800 k£ and 900 k£ for power outputs
around 200 kWe,net (Huang, Wang, et al., 2013). A similar work is published by Huang, McIlveen-
Wright, et al. (2013). Comparing a biomass-driven ORC for cogeneration with a gasification
system, both modelled at 150 kWe,gr the authors conclude that both systems are not
economically feasible for power-only operation. The ORC system has higher investment costs
(570,020 £) than the gasification system and is most suitable in buildings with a high heat to
power ratio (Huang, McIlveen-Wright, et al., 2013).
Other publications estimate the costs of the system under consideration, but do not reveal the
capital costs estimated. For instance, Ahmadi, Dincer, and Rosen (2014) simulate a biomass-
based system to produce electricity, cooling, heating, hydrogen, domestic hot water and
desalinated fresh water. The multi-objective optimization minimizes the total cost rate, defined
as the sum of the purchase costs of the components, the cost of the environmental impact and
the fuel, while maximizing the cycle exergy efficiency (Ahmadi et al., 2014).
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
49
Biomass ORC applications: assumed capital costs
Finally, several publications investigate the economics of ORC technology and its applications,
but the capital costs used are not from real projects neither have these been estimated within
the frame of the publication. Moro, Pinamonti, and Reini (2008) investigate the potential for
combined heat and power production from biomass waste from the furniture industry in Italy.
An ORC system with a capacity of 1000 kWe and 5600 kWth was modelled and the financial
assessment was performed using a capital cost of 1.3 M€ and financial support from green
certificates. The system was assessed beneficially with an IRR larger than 22%, an NPV of € 1.8
M€ and a payback period below 5 years (Moro et al., 2008). Maraver, Uche, and Royo (2012)
investigate the potential of a biomass-driven polygeneration system including multi-effect
desalination. The ORC system is the prime mover, the heat from the system is recovered for
heat, cold and desalination purposes. The costs of the ORC system are assumed at 3500 €/kW,
based on literature findings. The economic feasibility is demonstrated for varying investment
costs and biomass and desalted water reference prices. Algieri and Morrone (2014) investigate
biomass CHP ORC systems for single-family applications in the Italian context. The best electric
power performance was obtained under transcritical conditions with internal heat exchange,
with a power output of 0.76 kWe,net. The best thermal performance is obtained under
superheated conditions without internal heat exchange, the power output in this case amounts
0.23 kWe,net. The costs of these ORC systems have been assumed in the 5000 to 10,000 €/kW
range ta account for future price changes. The scope of these costs is not specified, but they
presumably refer to the complete project. Two cycle setups are compared in terms of energetic
and economic performance (Algieri & Morrone, 2014).
Other authors use assumed costs to compare the economics of ORC technology with other
potential applications. For instance Gard (2008), in his master thesis, investigates the
technological options for biomass-driven CHP systems in the 1-5 MWth range. Cost quotations
from other publications are used to compare the options and assess the financial feasibility.
Analysing the feasibility of a 450 kWe ORC system in Harads, Sweden, the ORC electricity
production itself would be unprofitable but the high return from the heat sales makes the
overall plant economically feasible. Finally, the cost of the generated electricity is most
influenced by the annual load hours (Gard, 2008). Rentizelas, Karellas, Kakaras, and
Tatsiopoulos (2009) compare the merits of ORC technology and gasification for biomass
trigeneration. The SIC of a 1 MWe reference plant is set at 2760 €/kWe. Similarly, Wood and
Rowley (2011) compare six biomass-based CHP systems for community housing, including a 200
kWe ORC system. The capital costs are assumed at 1.25 M€ (5950 €/kWe), based on
communication with manufacturers. Gonçalves, Faias, and de Sousa (2012) compare a steam
Rankine cycle with an ORC system for electricity generation from biomass in a sawmill plant. The
total investment costs of both projects are based on assumed equipment costs and cost for
CHAPTER TWO
50
integration. The specific investment costs of the ORC system (2636 €/kW) are lower than that of
the steam system (3092.5 €/kW), but the steam cycle has a larger power output (1010 kW) than
the ORC (385 kW) (Gonçalves et al., 2012). Likewise, Kempegowda, Skreiberg, and Tran (2012)
estimate the economic production costs of several technologies for biomass CHP in the
Norwegian context. Based on an average capital cost of 10,000 NOK/kWe for a 0.2 to 3 MW ORC
district heating system, the project is evaluated negatively when no financial support is
provided. In case the project is supported with green certificates, grid fee reductions and
investment support, the evaluation turns positive. Maraver, Sin, Royo, and Sebastián (2013)
study the optimal configuration of small-scale biomass-fuelled combined cooling, heating and
power (CCHP) cycles comparing an ORC system and a Stirling engine. The costs of an ORC
system would be in the range 1000 – 6000 €/kWe for power outputs between 3 kWe and 2000
kWe whereas the Stirling engine would cost 5000 – 14,000 €/kWe for outputs in the 1 kWe to
150 kWe range. The analysis is mainly technical, but a brief economic analysis using average
investment costs reveals ORC systems could be more interesting than the more efficient Stirling
engines due to the lower investment costs (Maraver et al., 2013).
Biomass ORC applications: an overview of capital costs
Biomass is a convenient source of energy. Because of its diverse forms, applications and its
potential to be stored until required, it is a promising source for renewable electricity
generation. Nearly half of the existing ORC applications worldwide operate on biomass and the
investigated literature contains several insights in the costs of real biomass ORC systems. Many
of the studies that study the economics of biomass ORCs consider cogeneration, or even
polygeneration, setups. The fact that the heat source is not for free, as is the case in most heat
recovery applications, influences the economic evaluation and encourages a further-reaching
energy use efficiency. The specific investment costs of the studied biomass ORC systems are
summarized in Figure 2.2. The same remarks hold for this figure as discussed for Figure 2.1.
There is more insight in the costs of commissioned biomass systems than for the case of excess
heat ORCs. The costs for a complete biomass project range between 2378 €2015/kW for 1.5 MW
capacity and 11,400 €2015/kW for 400 kW, of which the highest value stems from the very first
biomass project undertaken in 2001. There is a fair number of biomass price quotes available,
suggesting project costs between 2300 and 8900 €2015/kW and module prices in the 800 to 1500
€2015/kW range, with lower prices corresponding to relatively higher system capacities. These
price quotes are all published by the experienced manufacturer Turboden and are therefore
considered representative. Subsequently, there are significantly less cost estimates for biomass
than for excess heat recovery ORC systems, but their values are very diverging. The costs of
multiple biomass projects with designed power output around 220 kW are estimated between
4300 and 4900 €2015/kW, while other studies estimate the costs of a 350 kW and a 1 MW system
at approximately 11,000 €2015/kW. The largest estimated systems have power outputs between
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
51
3 and 4 MW, but specific investment costs that are drastically lower than all other values.
Finally, the assumed biomass costs tend to be in line with the other reported values, albeit
somewhat on the lower edge of the range. Only few authors estimate the LCOEs of biomass
ORC projects. In trigeneration and cogeneration mode, the cost of electricity generation is
estimated mostly between 110 and 140 €2015/kWh, but in power only mode this value can be as
high as 300 €2015/kWh.
Figure 2.2. Investment costs for biomass ORC systems, as published in the literature.
Legend: Overview of the capital costs for biomass ORC projects (P) and modules (M), as retrieved by the literature review. The
overview distinguishes between the costs of real systems, representative quotes from ORC manufacturers, estimated costs and
assumed costs.
2.3.4. The costs of ORC technology: geothermal applications
The combination of geothermal wells and ORC technology has been proven. The first
geothermal ORC was commissioned in 1984 and today there are more than 100 systems
operational and another 20 under construction, representing only 17% of all ORC systems
worldwide but more than three quarters of total ORC installed capacity. Practical
implementations of geothermal plants using ORC technology are reported by e.g., Bronicki
(1988) and by Brasz et al. (2005), who discuss the planned installation of an ORC plant using
low-temperature geothermal heat from Chena Hot Springs in Alaska. The literature review
disclosed 27 publications that discuss the economics of geothermal ORC systems (see Table 2.4
and Appendix A).
100
1.000
10.000
100.000
100 1.000 10.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Real
M Real
P Quote
M Quote
P Estimated
M Estimated
P Assumed
M Assumed
CHAPTER TWO
52
Table 2.4. Amount of geothermal ORC publications according to system and economic scope.
System scope Module Project
Economic scope
Real 0 5
Quote 0 2
Estimated 6 10
Assumed 1 3 Legend: The table summarizes the literature review by providing the number of publications according to the scope of the ORC
system discussed in the publication and the scope of the economic data provided.
Geothermal ORC applications: capital costs of real systems
There are few authors that discuss the economics of commissioned geothermal ORC plants.
Entingh and F. (2003) collected cost data for geothermal power systems. The authors report
prices of 5100 $/kW for an ORC installed in 1986 in East Mesa, Arizona, USA and 3600 $/kW for
a 1993 plant in Heber, California, USA. A prospect from 1998 stated a SIC of 2,700 $/kW,
whereas a representative estimated the costs for a system at 1700 $/kW in 2000. Power
outputs for each of the systems were not disclosed (Entingh & F., 2003). Pernecker (1999)
discloses the details of a 1 MW geothermal ORC project commissioned in 2001 in Altheim,
Austria. The geothermal well was already in use to provide district heating and domestic hot
water and in 1999 the decision was made to expand the activities with electricity production.
The total costs of the system, including a reinjection well amounted 4.5 M€, of which 1.6 M€
was financed by the European Commission and 640 k€ by the Austrian government (Pernecker,
1999). In 2006, an ORC system was installed in Fairbanks, Alaska for the Chena Hot Springs
Resort (Holdmann, 2007a). Two ORC modules were installed, for a combined capacity of 400
kW, at a total cost of approximately 2 M$ (Holdmann, 2007b). Finally, Lazzaretto, Toffolo,
Manente, Rossi, and Paci (2011) and Toffolo, Lazzaretto, Manente, and Paci (2014) compare the
results of a cost estimate with the known costs of a commissioned system. The commissioned
geothermal ORC system has a capacity of 33.6 MW, with a module cost of 72.7 k€ and a total
investment cost of 133 k€. The authors disclosed no further details about the reference plant
(Lazzaretto et al., 2011; Toffolo et al., 2014).
Geothermal ORC applications: representative manufacturer’s quotes
Forsha and Nichols (1991) gave insight in the factors affecting the costs of binary power plants,
from a manufacturer’s perspective. The costs of two representative systems are disclosed in
detail. A system operating on isobutane costs 1775 $/kW, whereas an ORC using ammonia has a
projected cost of 1415 $/kW. These figures are representative for systems with a capacity
ranging from 5 MW to 20 MW (Forsha & Nichols, 1991).
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
53
Geothermal ORC applications: estimated capital costs
The costs of geothermal ORC systems have been estimated by e.g., I.K. Smith, Stosic, Kovacevic,
and Langson (2007). They investigated the use of screw expanders in geothermal binary power
plants and developed a lab-scale ORC unit using a twin screw expander. Detailed price quotes
for all components were obtained from vendors. The numbers or the distribution of these
quotes is not given, but they are used to estimate the costs of larger systems via established
scaling methods. Building and installation of units in the 25 – 500 kW output range could be
possible for about 1500 to 2000 $/kW if water cooled and 2500 $/kW if air cooled. Compared to
a 200 kW turbine ORC these equivalent screw expander units would be about 30% cheaper (I.K.
Smith et al., 2007). Also Campos Rodríguez et al. (2013) make cost estimates on the basis of
previous purchase orders and quotations from professional estimators. They compare an ORC
and a Kalina cycle system for power production from a 3 km deep geothermal well, using
various working fluids. The project using the ORC costs 17.9 M€, including the well costs, of
which 1.7 M€ stems from the 1.8 MWgr ORC plant. Similarly, the 1.8 MWgr Kalina module costs
1.7 M€ and has a total project cost of 17.8 M€ (Campos Rodríguez et al., 2013).
Other authors use correlations to estimate the required costs, including Astolfi, Romano,
Bombarda, and Macchi (2014a) and Astolfi, Romano, Bombarda, and Macchi (2014b), who
investigate binary ORC plants for low and medium geothermal sources. The techno-economic
optimization investigates the optimal cycle setup and suitable fluids for the geothermal sources,
with minimum SIC as objective function. The investment costs of the system are estimated with
new correlations fitted based on experience in cooperation with manufacturers. The
thermodynamic and economic performance is compared for four working fluids: R134a,
ammonia, water and octane. The SIC of the cycles using water and octane range around 7000
€/kW and higher, making these cycles economically unfeasible. The SIC for R134a is estimated
at 2510 €/kW and 2697 €/kW for capacities of 10.21 MW and 10.51 MW respectively. For
ammonia, the total project SIC amount 2763 €/kW for 8.3 MW and 2869 €/kW for 7.4 MW.
These results included the assumption of 12 M€ investment costs for the well. A sensitivity
analysis on the geothermal well costs suggests the costs of the geothermal field have a strong
influence on the total project costs (Astolfi et al., 2014a, 2014b).
A detailed bottom-up cost estimation was performed by Lazzaretto et al. (2011) and later
elaborated by Toffolo et al. (2014). The authors assess the accuracy of the applied cost
estimation technique by comparing the results of the estimate with the costs of a commissioned
system. The costs of the project were estimated at 135 k$, which is fairly close to the 133 k$
investment costs for the actual 33.6 MW system. Knowing that the accuracy is no problem, the
authors estimate the costs for two geothermal ORC systems. The first one operates on
isobutane, has a power output of 3.9 kWe,gr, a module cost of 2234 $/kW and a project cost of
4639 $/kW. The second case operates on R134a, which gives an output of 4.8 kWe,gr for a
CHAPTER TWO
54
module cost of 2348 $/kW and a project cost of 4764 $/kW (Lazzaretto et al., 2011; Toffolo et
al., 2014). In the same way, M. Li et al. (2014) use component cost correlations to compare the
costs of a transcritical CO2 cycle and an ORC for low temperature geothermal power production.
The optimal results give values for the cost per net power output ranging from 1250 $/W to
1870 $/W with net power output values between 173.57 kW and 204.86 kW, for different
working fluids and with or without an internal heat exchanger (M. Li et al., 2014). Also Heberle,
Bassermann, Preißinger, and Brüggemann (2012) perform an exergoeconomic optimization of
ORCs for low-temperature geothermal sources. The costs of the ORC module are composed as
the sum of the component costs, estimated using correlations. The obtained module costs
range between 980 k$ and 117 k$ for installed capacities between 1530 kWe,gr and 1639 kWe,gr.
In a later work, Heberle and Brüggemann (2015) estimate the costs of the components using
correlations, applying a multiplication factor of 6.32 to obtain the total investment costs. The
investigation concerns the use of zeotropic instead of pure working fluids. The obtained SIC
values depend on concentration and the type of the working fluids mixtures. In general, the
authors demonstrate how zeotropic mixtures can increase the SIC because of the required heat
exchange area, but at the same time decrease the electricity generation costs due to the
increased power output (Heberle & Brüggemann, 2015). Zare (2015) takes the same approach
to estimate the costs for their work, comparing three different cycle setups (the simple, the
regenerative and the ORC with an internal heat exchanger) from an exergoeconomic point of
view. The specific investment costs are presented graphically for the different cycle setups, as a
function of different geothermal brine temperatures. The ORC with internal heat exchanger is
superior in thermodynamic performance, whereas the simple ORC is the preferred option from
an economic point of view (Zare, 2015).
Furthermore, Walraven, Laenen, and D'Haeseleer (2015) optimize an air-cooled ORC system for
low-temperature geothermal applications. The costs of the system have been estimated using
component correlations and the costs for the geothermal well are established at 27.5 M€, based
on a proposed geothermal demonstration project in Belgium. The resulting SIC and NPV are
displayed graphically as a function of several design parameters, both technical and economic.
More than 80 % of the investment costs of the ORC system stem from the air-cooled condenser
(Walraven, Laenen, & D'Haeseleer, 2015). In a follow-up work, Walraven, Laenen, and
D’haeseleer (2015) compare the impact of a water-cooled or air-cooled condensation on the
economics of a low-temperature geothermal ORC system. The costs of the air-cooled ORC
system are significantly higher, with more than 80 % of the total system costs stemming from
the condenser costs, whereas the water-cooled condenser is responsible for about only a third
of its total system costs. Moreover, the system using a wet cooling tower has a slightly higher
power output (Walraven, Laenen, & D’haeseleer, 2015). Finally, M.-H. Yang and Yeh (2016)
investigate the performance of ORC systems for geothermal sources, assessing the impact of
several parameters on the net power output and the total investment costs. The costs are
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
55
estimated using component cost correlations and the economic performance is assessed with
the net power output index, defined as the ratio of the net power output over the total costs.
The best performance, i.e. the highest NPI, is obtained for the working fluid R600 with a net
power output of 333 kWnet and a total investment cost of 1.1 M$ (M.-H. Yang & Yeh, 2016).
Also in the geothermal ORC research field there are authors that assess the economics of their
investigated system at the hand of non-economic indicators. For instance, Madhawa
Hettiarachchi, Golubovic, Worek, and Ikegami (2007) investigate the best design criteria for an
ORC binary cycle used for low-temperature geothermal electricity production. Their objective
function is the ratio of total heat exchanger area over the net power output. No numerical
economic values are calculated, but the authors conclude that the choice of the working fluid
can have a strong impact on the costs of the power plant (Madhawa Hettiarachchi et al., 2007).
Geothermal ORC applications: assumed capital costs
Finally, several authors discuss the costs of geothermal ORCs based on assumptions for the
capital costs. For instance, Ian K. Smith, Stosic, and Kovacevic (2005) investigate the use of
screw expanders in geothermal binary ORC power plants. The costs were assumed for the
electric generator, but those of the condenser and cooling system were not considered. The
generator costs of a 879 kWe,gr ORC system would cost 63 k$, the heater would add a cost of 75
k$ (Ian K. Smith et al., 2005). Oguz Arslan and Yetik (2011) optimize a supercritical ORC-Binary
for the Simav geothermal field in Turkey. The costs of the geothermal plant include the
permitting cost, steam gathering cost, the power plant equipment-construction cost,
transmission cost, operation-maintenance cost and salvage cost. Each of these subcategories
was included based on assumptions. The most profitable design is a 64.2 MWgr ORC-binary plant
with a benefit of 124.88 M$ (Oguz Arslan & Yetik, 2011). In a follow-up work, O. Arslan, Ozgur,
and Kose (2012) investigate the possibilities to utilize the thermal energy of the Simav
geothermal field in Turkey. Because of the occurrence of two phases in the geothermal field,
both an ORC-binary and a steam-binary combined cycle have been considered. The costs for
drilling of the well are not included since they are ready to use, the costs of the ORC-binary cycle
(1300 USD/kW) and the combined cycle (1,650 USD/kW) have been assumed based on similar
values found in literature (O. Arslan et al., 2012). Finally, Preißinger, Heberle, and Bruggemann
(2013) investigate eight different working fluids for a geothermal ORC system, assessed for sub-
and transcritical operation conditions. Using the case of the working fluid RC318, the authors
assume SIC of 1250 €/kW for subcritical and transcritical conditions. Because the output of the
transcritical cycle (4040 kWgr) is higher than that of the subcritical one (3611 kWgr), the
transcritical is evaluated more beneficial. When the costs of the transcritical cycle are assumed
to be higher, at 1500 €/kW for 4040 kWgr output, the results are still in favour of the transcritical
cycle (Preißinger et al., 2013).
CHAPTER TWO
56
Geothermal ORC applications: an overview of capital costs
Geothermal energy refers to thermal energy stored within the earth. Although it is a well-known
source of energy, it is less commonly used for ORC applications. This is reflected in the available
information on its costs. Figure 2.3 shows an overview of the data retrieved in the literature.
The available information comprises a project investment of 4500 €2015/kW for a 400 kW project
and 3000 €2015/kW for a 33.6 MW system. But the majority of the published geothermal ORC
cost information has either been estimated, or assumed. The estimated costs are, again, very
diverse. The module specific investment costs range generally above 1000 €2015/kW, but differ
significantly, with values of 6000 €2015/kW for a 6.3 MW module and 1000 €2015/kW for one with
an installed capacity of 7.4 MW. The assumed costs are all situated around 1000 €2015/kW, for
systems with only a few MW installed capacity but similarly for very large systems of more than
60 MW. There are only two papers that discuss the LCOE of geothermal ORC projects, both of
which based on estimated system costs. The LCOEs range between 61 €2015/kWh for a 3.8 MW
system to 213 €2015/kWh for a project with 1.8 MW installed capacity.
Figure 2.3. Investment costs for geothermal ORC systems, as published in the literature.
Legend: Overview of the capital costs for geothermal ORC projects (P) and modules (M), as retrieved by the literature review. The
overview distinguishes between the costs of real systems, representative quotes from ORC manufacturers, estimated costs and
assumed costs.
2.3.5. The costs of ORC technology: solar applications
Solar energy combined with ORC systems is less conventional: only 10 solar-driven ORC plants
exist today and a few more are under construction. Practical implementations are reported by
e.g., Canada, Cohen, Cable, Brosseau, and Price (2004) for a 1 MWe parabolic trough plant in
Arizona, USA. In Lesotho there is a 75 m2 parabolic trough 3 kWe ORC project (S. Quoilin, Orosz,
Hemond, & Lemort, 2011). Besides electricity generation, solar ORC systems can be utilized for
e.g., water desalination. A prototype 2.5 kWe solar reverse osmosis desalination system in
100
1.000
10.000
100.000
100 1.000 10.000 100.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Real
M Real
P Quote
M Quote
P Estimated
M Estimated
P Assumed
M Assumed
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
57
Athens, Greece is discussed by Manolakos, Kosmadakis, Kyritsis, and Papadakis (2009). X. D.
Wang et al. (2010) analyse a low-temperature solar ORC experimental setup. There are 20
identified publications that discuss the costs of solar ORC applications (see Table 2.5 and
Appendix A).
Table 2.5. Amount of solar ORC publications according to system and economic scope.
System scope Module Project
Economic scope
Real 1 2
Quote 1 2
Estimated 5 4
Assumed 1 4 Legend: The table summarizes the literature review by providing the number of publications according to the scope of the ORC
system discussed in the publication and the scope of the economic data provided.
Solar ORC applications: capital costs of real systems
In Saguaro, Arizona, USA a 1 MW solar ORC plant was installed in 2006. The plant utilizes
parabolic trough collectors and n-pentane as a working fluid (Canada et al., 2004). The costs of
this system could not be obtained directly, but are referred to by Vélez et al. (2012) as 5730
€/kW. Alternatively, Manolakos, Mohamed, Karagiannis, and Papadakis (2008) compare a
photovoltaic reverse osmosis (RO) desalination system with a solar-ORC system. Both systems
exist as lab-scale prototypes in the Agricultural University of Athens. The PV-driven system has a
PV peak power of 846 W, a DC motor of 510 W and a desalination system with capacity 0.1
m³/h. The system had an investment cost of 18 k€, of which 61% stems from the desalination
and 39 % from the energy system. The ORC system has 90 m² of vacuum tube collectors, a
Rankine engine with a 2 kW expander and a 100 kW condenser and the reverse osmosis systems
has a capacity of 0.3 m³/h. The investment costs amount 80 k€, of which 72.6 % from the energy
system and the remainder from the desalination system (Manolakos et al., 2008). Finally, Baral
and Kim (2015) analyse a 1 kW solar driven ORC system with a scroll expander for four different
working fluids (R245fa, R123, R141b and Ethanol). The models are validated by the laboratory-
scale ORC available, using R245fa. The cost analysis was based on the available invoices from
the purchase of the components for the lab ORC. Depending on the costs of the solar collectors
and the working fluid, the total power system costs (solar system and ORC) are summed
between 19 k$ and 22.3 k$ (Baral & Kim, 2015).
Solar ORC applications: representative manufacturer’s quotes
Barber (1978) discussed the costs of solar ORC systems. The estimated costs of an ORC system
are graphically displayed and amount about 700 $/kW for 100-200 kW systems. The installation
costs are included in the estimate and are taken as 50% of the costs of the ORC system. The
costs of the total project depend strongly on the solar technology applied. Estimated for a 100
CHAPTER TWO
58
kW installed system, also including 50% installation costs, the Fresnel lens is estimated cheapest
(2200 $/kW), followed by the evacuated tube (2300 $/kW), tracking concentrators at 315°C
(2300-3400 $/kW), tracking concentrators at 200°C (2500-3800 $/kW) and flat plate collectors
(2500-5700 $/kW). All costs are expressed in 1976 US Dollars. The costs are estimated using
present available techniques at the time of publication, the author estimates lower costs in case
the items would be in full production (Barber, 1978). More recently, the feasibility of solar
through ORC systems was evaluated within the frame of the STORES project, (Prabhu, 2006).
The equipment costs of a 5 MWgr module, including working fluid, were projected at 5.9 k$, or
approximately 1100 $/kW. The total installed costs are calculated by multiplying with a factor
2.7, which gives 15 k$. Adding the projected 10 to 12.5 M$ for the solar through fields gives a
total of 25 to 26 M$. For a 1 MWgr plant, the costs would amount 2.1 M$ for the components or
5.6 M$ including installation. Adding the solar through costs of 2.5 M$, the total costs would
amount between 9 and 10 M$.
Solar ORC applications: estimated capital costs
Georges, Declaye, Dumont, Quoilin, and Lemort (2013) present the design of a prototype ORC
module, suitable for solar applications. The 3 kWnet ORC system using R245fa and two scroll
expanders had a total module cost of 14,321 €, which gives a specific investment cost of 4774
€/kW. The largest cost shares stem from the evaporator, the low pressure expander and the air
condenser. Together, they represent 61% of total costs (Georges et al., 2013). Oliveira et al.
(2002) analyse a hybrid solar-gas Rankine cycle used for combined heating, cooling and power
supply for buildings. The system is powered by solar collectors and a gas burner complements
for periods of low radiation. Two prototypes of the system were installed, with a turbo-
generator of 1.5 kW and nominal cooling capacities of 2 kW and 5 kW. The prototype in
Loughborough, UK receives input from a 20 m² solar collector array; the one in Porto, Portugal
uses a gas burner. The economic analysis is simulated for three different locations to account
for varying climatic conditions: the two locations of the prototypes and additionally Darwin
(Australia). The projected initial ORC cost amounts 8500 €, excluding solar collector costs
(Oliveira et al., 2002). More recently, Freeman, Hellgardt, and Markides (2015) investigated a
solar ORC system for domestic cogeneration applications in the UK, analysing three solar heat
source variants: a non-concentrating evacuated-tube collector (ETC) array facing south with a
fixed inclination angle, a fixed concentrating parabolic-through collector (PTC) array and a 2-axis
solar tracking PTC. The component costs are calculated using cost correlations for the heat
exchangers, costs for the expander were retrieved from compressor costs and the working fluid
pump costs were estimated from collected pump data. The costs for the electricity-generating
components range between 34 £/W and 72 £/W for power outputs between 54 W and 89 W.
The total project costs range from 54 £/W to 102 £/W (Freeman et al., 2015).
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
59
Many authors combine the need for sustainable energy provision with the demand for fresh
water in solar-abundant regions. For instance, Kosmadakis, Manolakos, Kyritsis, and Papadakis
(2009) investigate a two-stage solar organic Rankine cycle used for reverse osmosis
desalination. The addition of a second stage in the ORC increases the efficiency and the
desalinated water production. The system has a high-temperature expander (8.5 kW) and a low-
temperature expander (6.9 kW), produces 2684.8 m³/y of desalinated water and yields 5712
kWh/y for the desalination. The system has an estimated total cost of 186 k€, the ORC system
makes up 45 k€ or 25.36 % of the total. The solar collector system is responsible for the largest
share (40.83 %). The remainder of the capital costs is composed by the desalination unit, civil
engineering and other costs (Kosmadakis et al., 2009). In the same way, Nafey and Sharaf (2010)
analysed a solar ORC combined with a reverse osmosis installation for water desalination. Water
desalination driven by renewable energy sources would be especially interesting in remote
areas where there is no or limited traditional energy supply. The authors present various solar
collector techniques together with the suitable working fluids. The capital costs of the ORC are
estimated using component cost correlations, originating from (Voros, Kiranoudis, & Maroulis,
1998). The resulting SIC values are not presented, but the costs are presented in terms of costs
per m³ desalinated water. The conclusion is that the PTC with toluene under superheat
conditions is economically the best option to drive the RO desalination process (Nafey & Sharaf,
2010). In a follow-up work, Nafey, Sharaf, and García-Rodríguez (2010) expand the analysis to
three different configurations: besides the basic configuration, the Pelton Wheel Turbine (PWT)
and Pressure Exchanger (PEX) configurations are investigated. The authors perform a thermo-
economic analysis where the costs are estimated using correlations for the respective
components. The specific annual desalination costs are lowest for the PEX configuration (0.572
$/m³), followed by the PWT (0.683 $/m³) and the basic configuration (0.898 $/m³) (Nafey et al.,
2010). Finally, Sharaf, Nafey, and García-Rodríguez (2011) perform a similar analysis by
comparing two solar ORC configurations for multi effect distillation (MED). The first
configuration utilized the solar energy directly for the MED process; the second configuration
includes an ORC and produces both electricity and desalted water. The economics are discussed
in terms of costs per year (Sharaf et al., 2011).
Finally, Yamaguchi, Zhang, Fujima, Enomoto, and Sawada (2006) investigate a supercritical CO2
ORC powered by solar energy. The costs of the system are estimated and compared to those of
a petroleum-fired plant. The cost estimation method is unclear and the obtained investment
costs for the system are not disclosed (Yamaguchi et al., 2006).
Solar ORC applications: assumed capital costs
Water desalination using renewable energy sources can be interesting in case of remote areas
with limited energy supply or very high energy costs, but solar-ORC desalination is still in an
early stage of development with only a few pilot and demonstration plants (Bruno, López-
CHAPTER TWO
60
Villada, Letelier, Romera, & Coronas, 2008). Bruno et al. (2008) compare solar ORC systems for
reverse osmosis desalination using different types of solar collectors and with a photovoltaic
system used to drive the RO desalination system. The simulation is performed for two
representative cases, in Barcelona and Almería. The costs of the ORC system are assumed at
2000 €/kW, the costs of the solar collector fields and reverse osmosis system have been based
on other studies (Bruno et al., 2008).
Kosmadakis, Manolakos, and Papadakis (2011) consider a system combining a concentrated
photovoltaic system with an ORC. The excess heat from the CPV system is recovered by the ORC
to increase the total power output. The authors perform a detailed financial analysis of the total
system, taking a cost for the ORC of 2 €/We. The CPV-ORC system has a better economic
performance than the simple CPV system (Kosmadakis et al., 2011). Similarly, Karellas, Terzis,
and Manolakos (2011) study the financial feasibility of a solar thermal ORC-PV desalination
system for use on the Chalki Island in Greece. The ORC system has a nominal output power of
250 kWgr. The complete system has an investment cost of 2.3 M€, the annual costs are near
18,000 €/y and the Levelized Water Costs are estimated at 10.17 €/m³. The ORC system itself
has an estimated SIC of 2550 €/kW. The origin of these figures is unclear (Karellas et al., 2011).
Villarini et al. (2013) investigate a solar collector field coupled to a 3 kWgr ORC system, an
absorber unit and a thermal storage unit. A prototype is installed in the University of Tuscia in
Viterbo, Italy, and the thermodynamic and economic performances are simulated. The costs of
the ORC are assumed at 10 k€, the total system costs amount 60.2 k€ including the solar
thermal plant, the ORC, the absorption refrigerator and the costs for controls and installation.
Solar ORC applications: an overview of capital costs
The most abundant source of energy is the sun, yet only a minor share of the energy supply
worldwide stems from solar energy. There are only a few solar ORC applications operational,
but the subject is well-researched. Figure 2.4 shows an overview of the specific investment costs
of solar ORC systems, as collected in the literature review. The majority of these costs is
estimated or assumed. The real costs of solar ORC systems are published around 5000 €2015/kW
for a 1 MW system. All other systems are laboratory prototypes with module investment costs
between 17 and 20 k€2015/kW for 1 kW capacity and of approximately 28 k€2015/kW for a 2 kW
project. The values of the other references, quoted, estimated and assumed, vary considerably.
The overall trend is one of decreasing specific investment costs for increasing power output of
the ORC system. There is only one study that investigates the LCOE of the designed solar ORC
system. The obtained results are substantial, with electricity production costs between 592 and
1265 €2015/kWh, for solar systems with capacities between 54 and 80 W.
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
61
Figure 2.4. Investment costs for solar ORC systems, as published in the literature.
Legend: Overview of the capital costs for solar ORC projects (P) and modules (M), as retrieved by the literature review. The
overview distinguishes between the costs of real systems, representative quotes from ORC manufacturers, estimated costs and
assumed costs.
2.4. Discussion of the results
The aim of this chapter was to get a basic understanding of the technical potential of ORC
technology and to summarize the state-of-the art knowledge on its economics. This section
summarizes the most important findings. The primary applications for ORC technology are
renewable power generation from biomass, geothermal and solar sources and industrial energy
efficiency improvement by means of excess heat recovery. The most used applications today
involve biomass and excess heat as energy source, so that most economic insights are available
for these applications.
A thorough review of the literature lead to several insights. First of all, there is only a minor
share of the academic literature on ORC technology that studies the development of the
technology beyond the technical-engineering dimension. Although earlier efforts were carried
out to summarize insights in the economics of ORC technology (see Sylvain Quoilin et al. (2013);
Vélez et al. (2012)), these studies span only a limited volume of the available economic insights.
This chapter offers an updated, detailed exploration of the economics of ORC technology.
Secondly, although the basic working principle of ORC technology is straightforward, fine-tuning
its setup and optimizing its performance is a highly complex matter, and moreover, most
research does not consider the basic system setup but explores new designs. This obviously
complicates comparison. To organize the review to the extent possible, this chapter proposed to
structure the literature according to the application of the ORC system, its scope and the extent
to which it discusses the economics. But even after this classification, several cataloguing issues
remained. For instance, the scope of the ORC system was not always explicit and varied
100
1.000
10.000
100.000
0,01 0,10 1,00 10,00 100,00 1.000,00 10.000,00
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Real
M Real
P Quote
M Quote
P Estimated
M Estimated
P Assumed
M Assumed
CHAPTER TWO
62
noticeably among the different studies and the general lack of date stamps complicates
economic interpretation.
Most studies that investigate the economics of ORC technology focus on the capital investment.
In general, the published costs for the different heat sources vary considerably, but the limited
amount of data on real ORC systems suggests the specific investment costs are the highest for
ORC systems that operate on solar energy. Note that there is very little experience with solar
ORC systems so there have not been many opportunities to benefit from experience. The costs
of the first biomass ORC systems were substantial, but today it is possible to invest in a biomass
project from approximately 2300 €2015/kW onwards. One explanation is that accumulating
application of a particular system setup increases the manufacturer’s experience, which is
likewise reflected in the charged system costs (see chapter 5). A significant share of the
literature does not discuss the economics of real, commissioned ORC systems, but rather
estimate the costs of newly proposed system designs. There are various approaches to estimate
the costs of industrial processes, from scaling methods to bottom-up cost estimation with
component-specific cost correlations, and the resulting specific investment costs are very
diverse. The question here is to what extent these estimated costs can be considered
representative for the actual system costs. Finally, the focus on the investment costs of the
technology is understandable because of its direct relation with both the technical system setup
and the financial feasibility of the system investment. But the criterion generally used to assess
the competitiveness of an electricity generation technology is the LCOE. This metric reflects the
costs at which electricity can be produced with the system under consideration. Nonetheless,
there are only very few studies that assess the LCOE of ORC applications. The LCOE ranges
between 51 and 134 €2015/kWh for heat recovery projects, between 110 and 140 €2015/kWh for
biomass ORCs in trigeneration and cogeneration mode and up to 300 €2015/kWh when in power
only mode, between 61 and 213 €2015/kWh for geothermal ORC projects and as high as 592 to
1265 €2015/kWh for the investigated solar ORC systems. Note that the majority of these LCOEs
have been calculated for ORC systems for which the investment costs have been estimated, so
that they are based on a variety of assumptions regarding the ORC project, such as its capacity
factor or the valid discount rate, as well as concerning the cost estimation itself.
2.5. Chapter conclusions
The ORC is a heat-to-power conversion cycle with the same working principle as the
conventional steam Rankine cycle. Because it operates using different working fluids, typically
with a lower evaporation temperature, the cycle is very suitable to process lower-grade thermal
energy sources including industrial excess heat flows and several renewable sources. Hence, the
merit of the technology is the potential to contribute to address current challenges in energy
sustainability. But the diffusion of a promising technology depends on more than its technical
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
63
appeal only. This chapter focusses on one of the factors of influence, the costs of the technology
itself. Although ORC technology constitutes a very active domain of research, the large majority
of efforts are focussed on technical optimization. A relatively small share of the literature
engages in discussing the economics of ORC technology and the majority of these entail
engineering studies that provide bottom-up cost projections of the technology design under
consideration. By knowledge of the author, no comprehensive review of the economic ORC
literature has been made until today. Hence, the contribution of this chapter consists of an
extensive insight in the current knowledge about the costs of ORC systems in different
applications. By knowledge of the author, such an exhaustive review of the economics of ORC
technology has not been published before.
An important conclusion from this chapter is the overall lack of data on the economics of
commissioned ORC systems. A large share of the literature discusses the economics based on
costs that have been estimated, but nearly none of these critically assess to what extent these
estimates are representative. Therefore, chapter 3 investigates multiple cost estimation
techniques and critically assesses their usefulness for ORC technology. Moreover, there is only a
limited amount of data on real ORC systems and the financial feasibility of ORC investments
remains opaque. Chapter 4 addresses these topics by presenting an in-depth study of a heat
recovery ORC application, with detailed financial analysis and parameter sensitivity analysis.
Finally, ORC technology is increasingly applied in practice. Most experience today exists with
biomass applications and the cost data published for the first biomass ORC systems suggests
that acquired insights may have decreased system costs. To investigate this hypothesis in detail,
chapter 5 investigates the diffusion of ORC technology worldwide and discusses the influence of
scale and experience on its costs.
CHAPTER TWO
64
Bibliography
Ahmadi, P., Dincer, I., & Rosen, M. A. (2014). Thermoeconomic multi-objective optimization of a novel biomass-based integrated energy system. Energy, 68, 958-970. doi:10.1016/j.energy.2014.01.085
Algieri, A., & Morrone, P. (2012). Comparative energetic analysis of high-temperature subcritical and transcritical Organic Rankine Cycle (ORC). A biomass application in the Sibari district. Applied Thermal Engineering, 36, 236-244. doi:10.1016/j.applthermaleng.2011.12.021
Algieri, A., & Morrone, P. (2014). Techno-economic Analysis of Biomass-fired ORC Systems for Single-family Combined Heat and Power (CHP) Applications. Energy Procedia, 45, 1285-1294. doi:10.1016/j.egypro.2014.01.134
Amini, A., Mirkhani, N., Pakjesm Pourfard, P., Ashjaee, M., & Khodkar, M. A. (2015). Thermo-economic optimization of low-grade waste heat recovery in Yazd combined-cycle power plant (Iran) by a CO2 transcritical Rankine cycle. Energy, 86, 74-84. doi:10.1016/j.energy.2015.03.113
Angelino, G., Gaia, M., & Macchi, E. (1984, 13-15 February). Medium temperature 100kW ORC engine for total energy systems - Experimental results. Paper presented at the International Seminal of the Commission of the European Communities on Energy Conservation in Industry, Düsseldorf.
Arslan, O., Ozgur, M. A., & Kose, R. (2012). Electricity Generation Ability of the Simav Geothermal Field: A Technoeconomic Approach. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 34(12), 1130-1144. doi:10.1080/15567031003773254
Arslan, O., & Yetik, O. (2011). ANN based optimization of supercritical ORC-Binary geothermal power plant: Simav case study. Applied Thermal Engineering, 31(17-18), 3922-3928. doi:10.1016/j.applthermaleng.2011.07.041
Arvay, P., Muller, M. R., Ramdeen, V., & Cunningham, G. (2011, 26/07 29/07 2011). Economic Implementation of the Organic Rankine Cycle in Industry. Paper presented at the 2011 ACEEE Summer Study on Energy Efficiency in Industry, Niagara Falls, NY.
Astolfi, M., Romano, M. C., Bombarda, P., & Macchi, E. (2014a). Binary ORC (organic Rankine cycles) power plants for the exploitation of medium–low temperature geothermal sources – Part A: Thermodynamic optimization. Energy, 66, 423-434. doi:10.1016/j.energy.2013.11.056
Astolfi, M., Romano, M. C., Bombarda, P., & Macchi, E. (2014b). Binary ORC (Organic Rankine Cycles) power plants for the exploitation of medium–low temperature geothermal sources – Part B: Techno-economic optimization. Energy, 66, 435-446. doi:10.1016/j.energy.2013.11.057
Baral, S., & Kim, K. C. (2015). Simulation, Validation and Economic Analysis of Solar Powered Organic Rankine Cycle for Electricity Generation. Journal of Clean Energy Technologies, 3(1), 62-67. doi:10.7763/jocet.2015.v3.170
Barber, R. E. (1978). Current costs of solar powered organic Rankine cycle engines. Solar Energy, 20(1), 1-6.
Barbier, E. (2002). Geothermal energy technology and current status: an overview. Renewable and Sustainable Energy Reviews, 6, 3-65.
Barz, M. (2008). Biomass Technology for Electricity Generation in Community. International Journal of Renewable Energy, 3(1), 1-10.
Bejan, A., Tsatsaronis, G., & Moran, M. (1996). Thermal design and optimization New York: Wiley & Sons. Bendig, M., Maréchal, F., & Favrat, D. (2013). Defining “Waste Heat” for industrial processes. Applied
Thermal Engineering, 61(1), 134-142. doi:10.1016/j.applthermaleng.2013.03.020 Bini, R., Duvia, A., Schwarz, A., Gaia, M., Bertuzzi, P., & Righini, W. (2004). Operational results of the first
biomass CHP plant in Italy based on Organic Rankine Cycle turbogenerator and overview of a number of plants in operation in Europe since 1998.
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
65
Bolhar-Nordenkampf, M., Pröll, T., Aichernig, C., & Hofbauer, H. (2004). Techno Economical Assessment of Combined Heat and Power Production from Biomass. Paper presented at the 2nd World Conference and Technology Exhibition on Biomass for Energy, Industry and Climate Protection, Rome, Italy.
Brasz, J. J., Biederman, B. P., & Holdmann, G. (2005). Power Production from a Moderate -Temperature Geothermal Resource. Paper presented at the GRC Annual Meeting, Reno, NV, USA.
Broberg Viklund, S., & Johansson, M. T. (2014). Technologies for utilization of industrial excess heat: Potentials for energy recovery and CO2 emission reduction. Energy Conversion and Management, 77, 369-379. doi:10.1016/j.enconman.2013.09.052
Bronicki, L. Y. (1988). Electrical power from moderated temperature geothermal sources with modular mini-power plants. 17(1), 83-92.
Bronicki, L. Y. (Sine Dato). Organic Rankine cycle power plant for waste heat recovery (pp. 1-6). Nevada, USA: ORMAT International Inc. .
Bruno, J. C., López-Villada, J., Letelier, E., Romera, S., & Coronas, A. (2008). Modelling and optimisation of solar organic rankine cycle engines for reverse osmosis desalination. Applied Thermal Engineering, 28(17-18), 2212-2226. doi:10.1016/j.applthermaleng.2007.12.022
Campos Rodríguez, C. E., Escobar Palacio, J. C., Venturini, O. J., Silva Lora, E. E., Cobas, V. M., Marques dos Santos, D., . . . Gialluca, V. (2013). Exergetic and economic comparison of ORC and Kalina cycle for low temperature enhanced geothermal system in Brazil. Applied Thermal Engineering, 52(1), 109-119. doi:10.1016/j.applthermaleng.2012.11.012
Canada, S., Cohen, G., Cable, R., Brosseau, D., & Price, H. (2004). Parabolic Trough Organic Rankine Cycle Solar Power Plant. Paper presented at the DOE Solar Energy Technologies Program Review Meeting, Denver, Colorado.
Cayer, E., Galanis, N., & Nesreddine, H. (2010). Parametric study and optimization of a transcritical power cycle using a low temperature source. Applied Energy, 87(4), 1349-1357. doi:10.1016/j.apenergy.2009.08.031
Chen, H., Goswami, D. Y., & Stefanakos, E. K. (2010). A review of thermodynamic cycles and working fluids for the conversion of low-grade heat. Renewable and Sustainable Energy Reviews, 14(9), 3059-3067. doi:10.1016/j.rser.2010.07.006
Chinese, D., Meneghetti, A., & Nardin, G. (2004). Diffused introduction of Organic Rankine Cycle for biomass-based power generation in an industrial district: a systems analysis. International Journal of Energy Research, 28(11), 1003-1021. doi:10.1002/er.1012
David, G., Michel, F., & Sanchez, L. (2011). Waste heat recovery projects using Organic Rankine Cycle technology - Examples of biogas engines and steel mills applications. Paper presented at the World Engineers' Convention, Geneva, Switzerland.
Declaye, S., Quoilin, S., Guillaume, L., & Lemort, V. (2013). Experimental study on an open-drive scroll expander integrated into an ORC (Organic Rankine Cycle) system with R245fa as working fluid. Energy, 55, 173-183. doi:10.1016/j.energy.2013.04.003
Di Maria, F., & Micale, C. (2015). Exergetic and Economic Analysis of Energy Recovery from the Exhaust Air of Organic Waste Aerobic Bioconversion by Organic Rankine Cycle. Energy Procedia, 81, 272-281. doi:10.1016/j.egypro.2015.12.097
DiPippo, R. (1999). Small geothermal power plants: design, performance and economics. GHC Bulletin, 1-8.
Duvia, A., Bini, R., Spanring, H., & Portenkirchner, K. (2007). Application of ORC units in the MDF and particleboard sector - general considerations and overview of the experiences of the first ORC plant in this industry installed at MDF Hallein.
Duvia, A., Guercio, A., & Rossi di Schio, C. (2009). Technical and economic aspects of Biomass fuelled CHP plants based on ORC turbogenerators feeding existing district heating networks.
CHAPTER TWO
66
Duvia, A., & Tavolo, S. (2008). Application of ORC units in the pellet production field: technical-economic considerations and overview of the operational results of an ORC plant in the industry installed in Mudau (Germany).
Element Energy Ltd. (2014). The potential for recovering and using surplus heat from industry. London, Final report for DECC: 50 pp.
Entingh, D. J., & F., M. J. (2003). Historical Improvements in Geothermal Power System Costs. Geothermal Resources Council Transactions, 27, 533-537.
Erhart, T., Strzalka, R., Eicker, U., & Infield, D. (2011). Performance Analysis of a Biomass ORC Poly-generation System. Paper presented at the 2nd European Conference on Polygeneration, Tarragona, Spain
EU. (2009). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. Official Journal of the European Union.
Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC, (2012).
Eurostat. (2016). Glossary: European currency unit (ECU). Retrieved from http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:European_currency_unit_(ECU)
Forni, D., Vaccari, V., Di Santo, D., Rossetti, N., & Baresi, M. (2012, 11–14 September 2012). Heat recovery for electricity generation in industry. Paper presented at the ECEEE 2012 Summer Study on energy efficiency in industry, the Netherlands.
Forsha, M. D., & Nichols, K. E. (1991). Factors affecting the capital cost of binary power plants. Geothermal Resources Council Transactions, 15, 99-105.
Freeman, J., Hellgardt, K., & Markides, C. N. (2015). An assessment of solar-powered organic Rankine cycle systems for combined heating and power in UK domestic applications. Applied Energy, 138, 605-620. doi:10.1016/j.apenergy.2014.10.035
Gaia, M., & Macchi, E. (1981). Recovery of heat from exhaust gases at 250°C from a ceramic tunnel oven with an organic Rankine cycle engine. The Community's Energy R & D Programme Energy Conservation. Survey of Results (1975-1979) and Compilation of Selected New Projects (1979-1983), Commission of the European Communities: 91-92.
Gard, K. O. (2008). Biomass Based Small Scale Combined Heat and Power Technologies. (Master Thesis), Luleå University of Technology, Luleå, Sweden.
Georges, E., Declaye, S., Dumont, O., Quoilin, S., & Lemort, V. (2013). Design of a small-scale organic Rankine cycle engine used in a solar power plant. International Journal of Low-Carbon Technologies, 8(suppl 1), i34-i41. doi:10.1093/ijlct/ctt030
Ghirardo, F., Santin, M., Traverso, A., & Massardo, A. (2011). Heat recovery options for onboard fuel cell systems. International Journal of Hydrogen Energy, 36(13), 8134-8142. doi:10.1016/j.ijhydene.2011.01.111
Gonçalves, N., Faias, S., & de Sousa, J. (2012). Biomass CHP Technical and Economic Assessment applied to a Sawmill Plant. Paper presented at the International Conference on Renewable Energies and Power Quality (ICREPQ '12), Santiago de Compostela, Spain.
Grabińsky, P. (2011). Optimization of the performance of an organic Rankine cycle used as waste heat recovery system in a bio-liquid diesel engine power plant. (Master Thesis), University of Iceland & University of Akureyri.
Hammond, G. P., & Norman, J. B. (2014). Heat recovery opportunities in UK industry. Applied Energy, 116, 387-397. doi:10.1016/j.apenergy.2013.11.008
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
67
Heberle, F., Bassermann, P., Preißinger, M., & Brüggemann, D. (2012). Exergoeconomic optimization of an Organic Rankine Cycle for low-temperature geothermal heat sources. International Journal of Thermodynamics, 15(2). doi:10.5541/ijot.374
Heberle, F., & Brüggemann, D. (2015). Thermo-Economic Evaluation of Organic Rankine Cycles for Geothermal Power Generation Using Zeotropic Mixtures. Energies, 8(3), 2097-2124. doi:10.3390/en8032097
Heberle, F., & Brüggemann, D. (2016). Thermo-Economic Analysis of Zeotropic Mixtures and Pure Working Fluids in Organic Rankine Cycles for Waste Heat Recovery. Energies, 9(4), 226. doi:10.3390/en9040226
Holdmann, G. (2007a). 400kW Geothermal Power Plant at Chena Hot Springs, Alaska - Final Report Prepared for: Alaska Energy Authority. Fairbanks, Alaska, USA, Chena Hot Springs Resort: 38 pp.
Holdmann, G. (2007b). The Chena Hot Springs 400kW Geothermal Power Plant: Experience Gained During the First Year of Operation. Fairbanks, Alaska, USA, Chena Hot Springs / Chena Power: 9 pp.
HREII. (2013). EU Paper: ORC waste heat recovery in European energy intensive industries (LIFE Project Number LIFE10 ENV/IT/000397). Heat Recovery in Energy Intensive Industries 10 pp.
Huang, Y., McIlveen-Wright, D. R., Rezvani, S., Huang, M. J., Wang, Y. D., Roskilly, A. P., & Hewitt, N. J. (2013). Comparative techno-economic analysis of biomass fuelled combined heat and power for commercial buildings. Applied Energy, 112, 518-525. doi:10.1016/j.apenergy.2013.03.078
Huang, Y., Wang, Y. D., Rezvani, S., McIlveen-Wright, D. R., Anderson, M., Mondol, J., . . . Hewitt, N. J. (2013). A techno-economic assessment of biomass fuelled trigeneration system integrated with organic Rankine cycle. Applied Thermal Engineering, 53(2), 325-331. doi:10.1016/j.applthermaleng.2012.03.041
Huttrer, G. W. (2000, May 28 - June 10, 2000). The status of world geothermal power generation 1995-2000. Paper presented at the World Geothermal Congress, Kyushu - Tohoku, Japan.
Imran, M., Park, B. S., Kim, H. J., Lee, D. H., Usman, M., & Heo, M. (2014). Thermo-economic optimization of Regenerative Organic Rankine Cycle for waste heat recovery applications. Energy Conversion and Management, 87, 107-118. doi:10.1016/j.enconman.2014.06.091
IPCC. (2014). Summary for Policymakers. Climate Change 2014, Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, [Edenhofer, O., R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von Stechow, T. Zwickel and J.C. Minx (eds.)]. Cambridge, United Kingdom and New York, NY, USA., Intergovernmental Panel on Climate Change: 33 pp.
Kalina, J. (2011). Integrated biomass gasification combined cycle distributed generation plant with reciprocating gas engine and ORC. Applied Thermal Engineering, 31(14-15), 2829-2840. doi:10.1016/j.applthermaleng.2011.05.008
Karellas, S., Terzis, K., & Manolakos, D. (2011). Investigation of an autonomous hybrid solar thermal ORC–PV RO desalination system. The Chalki island case. Renewable Energy, 36(2), 583-590. doi:10.1016/j.renene.2010.07.012
Kempegowda, R. S., Skreiberg, O., & Tran, K.-Q. (2012). Techno-economic evaluations of various biomass CHP technologies and policy measures under Norwegian conditions. In G. Tranell (Ed.), Technoport 2012 - Sharing Possibilities and 2nd Renewable Energy Research Conference (Vol. 20, pp. 1-10).
Kosmadakis, G., Manolakos, D., Kyritsis, S., & Papadakis, G. (2009). Economic assessment of a two-stage solar organic Rankine cycle for reverse osmosis desalination. Renewable Energy, 34(6), 1579-1586. doi:10.1016/j.renene.2008.11.007
CHAPTER TWO
68
Kosmadakis, G., Manolakos, D., & Papadakis, G. (2011). Simulation and economic analysis of a CPV/thermal system coupled with an organic Rankine cycle for increased power generation. Solar Energy, 85(2), 308-324. doi:10.1016/j.solener.2010.11.019
Kwak, D.-H., Binns, M., & Kim, J.-K. (2014). Integrated design and optimization of technologies for utilizing low grade heat in process industries. Applied Energy, 131, 307-322. doi:10.1016/j.apenergy.2014.06.037
Law, R., Harvey, A., & Reay, D. (2013). Techno-economic comparison of a high-temperature heat pump and an organic Rankine cycle machine for low-grade waste heat recovery in UK industry. International Journal of Low-Carbon Technologies, 8(suppl 1), i47-i54. doi:10.1093/ijlct/ctt029
Lazzaretto, A., Toffolo, A., Manente, G., Rossi, N., & Paci, M. (2011, July 4-7, 2011). Cost evaluation of organic Rankine cycles for low temperature geothermal sources. Paper presented at the 24th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, Novi Sad, Serbia.
Le, V. L., Kheiri, A., Feidt, M., & Pelloux-Prayer, S. (2014). Thermodynamic and economic optimizations of a waste heat to power plant driven by a subcritical ORC (Organic Rankine Cycle) using pure or zeotropic working fluid. Energy, 78, 622-638. doi:10.1016/j.energy.2014.10.051
Lecompte, S., Huisseune, H., van den Broek, M., De Schampheleire, S., & De Paepe, M. (2013). Part load based thermo-economic optimization of the Organic Rankine Cycle (ORC) applied to a combined heat and power (CHP) system. Applied Energy, 111, 871-881. doi:10.1016/j.apenergy.2013.06.043
Lecompte, S., Huisseune, H., van den Broek, M., Vanslambrouck, B., & De Paepe, M. (2015). Review of organic Rankine cycle (ORC) architectures for waste heat recovery. Renewable and Sustainable Energy Reviews, 47, 448-461. doi:10.1016/j.rser.2015.03.089
Lecompte, S., Lazova, M., van den Broek, M., & De Paepe, M. (2014). Multi-objective optimization of a low-temperature transcritical organic Rankine cycle for waste heat recovery. Paper presented at the 10th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics (HEFAT2014), Orlando, Florida.
Lecompte, S., Lemmens, S., Huisseune, H., van den Broek, M., & De Paepe, M. (2015). Multi-Objective Thermo-Economic Optimization Strategy for ORCs Applied to Subcritical and Transcritical Cycles for Waste Heat Recovery. Energies, 8(4), 2714-2741. doi:10.3390/en8042714
Lee, K. M., Kuo, S. F., Chien, M. L., & Shih, Y. S. (1988). Parameters analysis on organic rankine energy recovery system. Energy Conversion and Management, 28(2), 129-136.
Leslie, N. P., Sweetser, R. S., Zimron, O., & Stovall, T. K. (2009). Recovered Energy Generation Using an Organic Rankine Cycle System. ASHRAE Transactions, 115(Part 1).
Li, M., Wang, J., Li, S., Wang, X., He, W., & Dai, Y. (2014). Thermo-economic analysis and comparison of a CO2 transcritical power cycle and an organic Rankine cycle. Geothermics, 50, 101-111. doi:10.1016/j.geothermics.2013.09.005
Li, Y.-R., Wang, J.-N., & Du, M.-T. (2012). Influence of coupled pinch point temperature difference and evaporation temperature on performance of organic Rankine cycle. Energy, 42(1), 503-509. doi:10.1016/j.energy.2012.03.018
Lukawski, M. (2009). Design and optimization of standardized organic Rankine cycle power plant for European conditions.
Madhawa Hettiarachchi, H. D., Golubovic, M., Worek, W. M., & Ikegami, Y. (2007). Optimum design criteria for an Organic Rankine cycle using low-temperature geothermal heat sources. Energy, 32(9), 1698-1706. doi:10.1016/j.energy.2007.01.005
Manolakos, D., Kosmadakis, G., Kyritsis, S., & Papadakis, G. (2009). On site experimental evaluation of a low-temperature solar organic Rankine cycle system for RO desalination. Solar Energy, 83(5), 646-656. doi:10.1016/j.solener.2008.10.014
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
69
Manolakos, D., Mohamed, E. S., Karagiannis, I., & Papadakis, G. (2008). Technical and economic comparison between PV-RO system and RO-Solar Rankine system. Case study: Thirasia island. Desalination, 221(1-3), 37-46. doi:10.1016/j.desal.2007.01.066
Maraver, D., Sin, A., Royo, J., & Sebastián, F. (2013). Assessment of CCHP systems based on biomass combustion for small-scale applications through a review of the technology and analysis of energy efficiency parameters. Applied Energy, 102, 1303-1313. doi:10.1016/j.apenergy.2012.07.012
Maraver, D., Uche, J., & Royo, J. (2012). Assessment of high temperature organic Rankine cycle engine for polygeneration with MED desalination: A preliminary approach. Energy Conversion and Management, 53(1), 108-117. doi:10.1016/j.enconman.2011.08.013
Meinel, D., Wieland, C., & Spliethoff, H. (2014). Economic comparison of ORC (Organic Rankine cycle) processes at different scales. Energy, 74, 694-706. doi:10.1016/j.energy.2014.07.036
Mills, D. (2004). Advances in solar thermal electricity technology. Solar Energy, 76(1-3), 19-31. doi:10.1016/s0038-092x(03)00102-6
Moro, R., Pinamonti, P., & Reini, M. (2008). ORC technology for waste-wood to energy in the furniture manufacturing industry. Thermal Science, 12(4), 61-73. doi:10.2289/TSCI0804061M
Muffler, P., & Cataldi, R. (1978). Methods for regional assessment of geothermal resources. Geothermics, 7, 53-89.
Nafey, A. S., & Sharaf, M. A. (2010). Combined solar organic Rankine cycle with reverse osmosis desalination process: Energy, exergy, and cost evaluations. Renewable Energy, 35(11), 2571-2580. doi:10.1016/j.renene.2010.03.034
Nafey, A. S., Sharaf, M. A., & García-Rodríguez, L. (2010). Thermo-economic analysis of a combined solar organic Rankine cycle-reverse osmosis desalination process with different energy recovery configurations. Desalination, 261(1–2), 138-147. doi:10.1016/j.desal.2010.05.017
Najjar, Y. S., & Radhwan, A. M. (1988). Cogenerating by combining gas turbine engine with organic Rankine cycle. Heat Recovery Systems and CHP, 8(3), 211-219.
Obernberger, I., Carlsen, H., & Biedermann, F. (2003). State-of-the-art and future developments regarding small-scale biomass CHP systems with a special focus on ORC and Stirling engine technologies. Paper presented at the International Nordic Bioenergy Conference and Exhibition, Jyväskylä, Finland.
Obernberger, I., Thonhofer, P., & Reisenhofer, E. (2002). Description and evalutation of the new 1,000 kWel Organic Rankine Cycle process integrated in the biomass CHP plant in Lienz, Austria. Euroheat & Power, 10/2002, 18-25.
OECD/IEA. (2011). Technology Roadmap: Geothermal Heat and Power. Retrieved from Paris, France: https://www.iea.org/publications/freepublications/publication/technology-roadmap-geothermal-heat-and-power.html
OECD/IEA. (2012). Technology Roadmap: Bioenergy for Heat and Power. Retrieved from Paris, France: http://www.iea.org/publications/freepublications/publication/technology-roadmap-bioenergy-for-heat-and-power-.html
OECD/IEA. (2013). Renewables Information 2013. IEA Statistics, International Energy Agency: 496 pp. OECD/IEA. (2014). Technology Roadmap: Solar Thermal Electricity. 2014 edition. IEA Statistics,
International Energy Agency: 496 pp. OECD/IEA. (2015). Key Renewables Trends. Exerpt from: Renewables Information 2015. IEA Statistics,
International Energy Agency: 8 pp. Oliveira, A. C., Afonso, C., Matos, J., Riffat, S., Nguyen, M., & Doherty, P. (2002). A combined heat and
power system for buildings driven by solar energy and gas. Applied Thermal Engineering, 22(6), 587-593.
CHAPTER TWO
70
Oluleye, G., Jobson, M., & Smith, R. (2014). A Hierarchical Approach for Evaluation of Waste Heat Utilization Opportunities. Chemical Engineering Transactions, 39, 1093-1098. doi:10.3303/CET1439183
ORCNext. (2011). IWT Vlaanderen SBO call 2010-2011. Project proposal strategic basic research (primary economical relevance): The next generation Oganic Rankine cycles (ORCNext). Flanders, Belgium, Universiteit Gent, Universiteit Antwerpen, Uliversité de Liège, HoWest, Atlas Copco, SET and Power-Link: 175 pp.
Papadopoulos, A. I., Stijepovic, M., & Linke, P. (2010). On the systematic design and selection of optimal working fluids for Organic Rankine Cycles. Applied Thermal Engineering, 30(6-7), 760-769. doi:10.1016/j.applthermaleng.2009.12.006
Papes, I., Degroote, J., & Vierendeels, J. (2015). New insights in twin screw expander performance for small scale ORC systems from 3D CFD analysis. Applied Thermal Engineering, 91, 535-546. doi:10.1016/j.applthermaleng.2015.08.034
Pellegrino, J. L., Margolis, N., Justiniano, M., Miller, M., & Thedki, A. (2004). Energy Use, Loss and Opportunities Analysis: U.S. Manufacturing & Mining. Prepared by Energetics, Incorporated and E3M, Incorporated for the U.S. Department of Energy Energy Efficiency and Renewable Energy Industrial Technologies Program: 169 pp.
Peretti, I. (2008). Application of ORC units in sawmills. Technical-economic considerations. Turboden: 12 pp.
Pernecker, G. (1999). Altheim geothermal plant for electricity production by ORC-turbogenerator. Bulletin d'hydrologie, 17, 225-232.
Pierobon, L., Nguyen, T.-V., Larsen, U., Haglind, F., & Elmegaard, B. (2013). Multi-objective optimization of organic Rankine cycles for waste heat recovery: Application in an offshore platform. Energy, 58, 538-549. doi:10.1016/j.energy.2013.05.039
Prabhu, E. (2006). Solar Through Organic Rankine Electricity System (STORES). Stage 1 - Power Plant Optimization and Economics. NREL National Renewable Energy Laboratory: 67 pp.
Preißinger, M., Heberle, F., & Bruggemann, D. (2013). Advanced Organic Rankine Cycle for geothermal application. International Journal of Low-Carbon Technologies, 8(suppl 1), i62-i68. doi:10.1093/ijlct/ctt021
Quoilin, S., Declaye, S., Tchanche, B. F., & Lemort, V. (2011). Thermo-economic optimization of waste heat recovery Organic Rankine Cycles. Applied Thermal Engineering, 31(14-15), 2885-2893. doi:10.1016/j.applthermaleng.2011.05.014
Quoilin, S., Orosz, M., Hemond, H., & Lemort, V. (2011). Performance and design optimization of a low-cost solar organic Rankine cycle for remote power generation. Solar Energy, 85(5), 955-966. doi:10.1016/j.solener.2011.02.010
Quoilin, S., van den Broek, M., Declaye, S., Dewallef, P., & Lemort, V. (2013). Techno-economic survey of Organic Rankine Cycle (ORC) systems. Renewable and Sustainable Energy Reviews, 22, 168-186. doi:10.1016/j.rser.2013.01.028
Rentizelas, A., Karellas, S., Kakaras, E., & Tatsiopoulos, I. (2009). Comparative techno-economic analysis of ORC and gasification for bioenergy applications. Energy Conversion and Management, 50(3), 674-681. doi:10.1016/j.enconman.2008.10.008
Schepers, B., & van Lieshout, M. (2011). IPO Nationale Routekaart Restwarmte: Een quickscan van de mogelijkheden. Delft, Interprovinciaal Overleg (IPO): 1-148.
Schuster, A., Karellas, S., Kakaras, E., & Spliethoff, H. (2009). Energetic and economic investigation of Organic Rankine Cycle applications. Applied Thermal Engineering, 29(8), 1809-1817.
Sharaf, M. A., Nafey, A. S., & García-Rodríguez, L. (2011). Exergy and thermo-economic analyses of a combined solar organic cycle with multi effect distillation (MED) desalination process. Desalination, 272(1-3), 135-147. doi:10.1016/j.desal.2011.01.006
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
71
Smith, I. K., Stosic, N., & Kovacevic, A. (2005). Screw Expanders Increase Output and Decrease the Cost of Geothermal Binary Power Plant Systems. GRC Transactions, 29, 787-794.
Smith, I. K., Stosic, N., Kovacevic, A., & Langson, R. (2007). Cost Effective Small Scale ORC Systems for Power Recovery from Low Enthalpy Geothermal Resources. Paper presented at the GRC Transactions.
STIA - Holzindustrie Ges.m.b.H. (2001). Biomass fired CHP plant based on an ORC cycle - Project: ORC-STIA-Admont. STIA - Holzindustria Ges.m.b.H. in cooperation with BIOS-Bioenergy Systems: 12 pp.
Stoppato, A. (2012). Energetic and economic investigation of the operation management of an Organic Rankine Cycle cogeneration plant. Energy, 41(1), 3-9. doi:10.1016/j.energy.2011.09.033
Tańczuk, M., & Ulbrich, R. (2013). Implementation of a biomass-fired co-generation plant supplied with an ORC (Organic Rankine Cycle) as a heat source for small scale heat distribution system – A comparative analysis under Polish and German conditions. Energy, 62, 132-141. doi:10.1016/j.energy.2013.09.044
Tchanche, B. F., Declaye, S., Quoilin, S., Papadakis, G., & Lemort, V. (2010, July 12-14, 2010). Economic feasibility study of a small scale organic rankine cycle in waste heat recovery application. Paper presented at the ASME 2010 10th Biennial Conference on Engineering Systems Design and Analysis (ESDA 2010), Istanbul, Turkey.
Tchanche, B. F., Lambrinos, G., Frangoudakis, A., & Papadakis, G. (2011). Low-grade heat conversion into power using organic Rankine cycles – A review of various applications. Renewable and Sustainable Energy Reviews, 15(8), 3963-3979. doi:10.1016/j.rser.2011.07.024
Toffolo, A., Lazzaretto, A., Manente, G., & Paci, M. (2014). A multi-criteria approach for the optimal selection of working fluid and design parameters in Organic Rankine Cycle systems. Applied Energy, 121, 219-232. doi:10.1016/j.apenergy.2014.01.089
Tumen Ozdil, N. F., & Segmen, M. R. (2016). Investigation of the effect of the water phase in the evaporator inlet on economic performance for an Organic Rankine Cycle (ORC) based on industrial data. Applied Thermal Engineering, 100, 1042-1051. doi:10.1016/j.applthermaleng.2016.02.117
U.S. DoE. (2008). Waste heat recovery: Technology and Opportunities in the US Industry. U.S. Department of Energy: 112 pp.
Uris, M., Linares, J. I., & Arenas, E. (2014). Techno-economic feasibility assessment of a biomass cogeneration plant based on an Organic Rankine Cycle. Renewable Energy, 66, 707-713. doi:10.1016/j.renene.2014.01.022
Vanslambrouck, B., Vankeirsbilck, I., Gusev, S., & De Paepe, M. (2011). Turn waste heat into electricity by using an Organic Rankine Cycle. Paper presented at the 2nd European Conference on Polygeneration, Tarragona, Spain.
Vélez, F., Segovia, J. J., Martín, M. C., Antolín, G., Chejne, F., & Quijano, A. (2012). A technical, economical and market review of organic Rankine cycles for the conversion of low-grade heat for power generation. Renewable and Sustainable Energy Reviews, 16(6), 4175-4189. doi:10.1016/j.rser.2012.03.022
Vescovo, R. (2009). ORC recovering industrial heat. Cogeneration and On-Site Power Production. Villarini, M., Bocci, E., Di Carlo, A., Sbordone, D., Falvo, M. C., & Martirano, L. (2013). Technical-Economic
Analysis of an Innovative Small Scale Solar Thermal - ORC Cogenerative System. Lecture Notes in Computer Science, 7972, 271-287.
Viswanathan, V. V., Davies, R. W., & Holbery, J. (2006). Opportunity Analysis for Recovering Energy from Industrial Waste Heat and Emissions. Retrieved from Richland, Washington:
Voros, N. G., Kiranoudis, C. T., & Maroulis, Z. B. (1998). Solar energy exploitation for reverse osmosis desalination plants. Desalination, 115(1), 83-101. doi:10.1016/S0011-9164(98)00029-0
CHAPTER TWO
72
Walraven, D. (2014). Optimization of the Energy Conversion Starting from Low-Temperature Heat: Application to Geothermal Binary Cycles. (PhD), KU Leuven
Walraven, D., Laenen, B., & D'Haeseleer, W. (2015). Economic system optimization of air-cooled organic Rankine cycles powered by low-temperature geothermal heat sources. Energy, 80, 104-113. doi:10.1016/j.energy.2014.11.048
Walraven, D., Laenen, B., & D’haeseleer, W. (2015). Minimizing the levelized cost of electricity production from low-temperature geothermal heat sources with ORCs: Water or air cooled? Applied Energy, 142, 144-153. doi:10.1016/j.apenergy.2014.12.078
Walsh, C., & Thornley, P. (2013). A comparison of two low grade heat recovery options. Applied Thermal Engineering, 53(2), 210-216. doi:10.1016/j.applthermaleng.2012.04.035
Wang, D., Ling, X., & Peng, H. (2013). Cost-effectiveness performance analysis of organic Rankine cycle for low grade heat utilization coupling with operation condition. Applied Thermal Engineering, 58(1-2), 571-584. doi:10.1016/j.applthermaleng.2013.04.057
Wang, J., Yan, Z., Wang, M., Ma, S., & Dai, Y. (2013). Thermodynamic analysis and optimization of an (organic Rankine cycle) ORC using low grade heat source. Energy, 49, 356-365. doi:10.1016/j.energy.2012.11.009
Wang, X. D., Zhao, L., Wang, J. L., Zhang, W. Z., Zhao, X. Z., & Wu, W. (2010). Performance evaluation of a low-temperature solar Rankine cycle system utilizing R245fa. Solar Energy, 84(3), 353-364. doi:10.1016/j.solener.2009.11.004
Wang, Z. Q., Zhou, N. J., Guo, J., & Wang, X. Y. (2012). Fluid selection and parametric optimization of organic Rankine cycle using low temperature waste heat. Energy, 40(1), 107-115. doi:10.1016/j.energy.2012.02.022
Wood, S. R., & Rowley, P. N. (2011). A techno-economic analysis of small-scale, biomass-fuelled combined heat and power for community housing. Biomass and Bioenergy, 35(9), 3849-3858. doi:10.1016/j.biombioe.2011.04.040
Wu, Y., Zhu, Y., & Yu, L. (2016). Thermal and economic performance analysis of zeotropic mixtures for Organic Rankine Cycles. Applied Thermal Engineering, 96, 57-63. doi:10.1016/j.applthermaleng.2015.11.083
Yamaguchi, H., Zhang, X. R., Fujima, K., Enomoto, M., & Sawada, N. (2006). Solar energy powered Rankine cycle using supercritical CO2. Applied Thermal Engineering, 26(17-18), 2345-2354. doi:10.1016/j.applthermaleng.2006.02.029
Yang, F., Zhang, H., Song, S., Bei, C., Wang, H., & Wang, E. (2015). Thermoeconomic multi-objective optimization of an organic Rankine cycle for exhaust waste heat recovery of a diesel engine. Energy, 93, 2208-2228. doi:10.1016/j.energy.2015.10.117
Yang, M.-H. (2015). Thermal and economic analyses of a compact waste heat recovering system for the marine diesel engine using transcritical Rankine cycle. Energy Conversion and Management, 106, 1082-1096. doi:10.1016/j.enconman.2015.10.050
Yang, M.-H., & Yeh, R.-H. (2015). Thermo-economic optimization of an organic Rankine cycle system for large marine diesel engine waste heat recovery. Energy, 82, 256-268. doi:10.1016/j.energy.2015.01.036
Yang, M.-H., & Yeh, R.-H. (2016). Economic performances optimization of an organic Rankine cycle system with lower global warming potential working fluids in geothermal application. Renewable Energy, 85, 1201-1213. doi:10.1016/j.renene.2015.07.067
Yari, M., & Mahmoudi, S. M. S. (2010). A thermodynamic study of waste heat recovery from GT-MHR using organic Rankine cycles. Heat and Mass Transfer, 47(2), 181-196. doi:10.1007/s00231-010-0698-z
THE ORGANIC RANKINE CYCLE: TECHNICAL CHARACTERISTICS, APPLICATIONS AND ECONOMICS
73
Zare, V. (2015). A comparative exergoeconomic analysis of different ORC configurations for binary geothermal power plants. Energy Conversion and Management, 105, 127-138. doi:10.1016/j.enconman.2015.07.073
This chapter is based on the publication: Lemmens, S., 2016, Cost Engineering Techniques and Their
Applicability for Cost Estimation of Organic Rankine Cycle Systems, Energies 9 (7): 485-503.
3. Cost engineering techniques and their
application for ORC technology
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
77
3.1. Introduction
One of the central quests in this thesis is the collection of insight in the economics of ORC
technology. The diffusion of an emerging technology is influenced by many aspects, one of
which is its economic appeal. ORC investments are typically characterized by substantial initial
investments and, in comparison, relatively low annual operation and maintenance costs. Hence,
the capital costs have an important influence on the financial assessment of an ORC project. At
the same time, the capital investment is to a large extent determined by the technical layout of
the system. Therefore, and because there is little information available about the costs of ORC
systems (cfr. chapter 1), several researchers attempt to get more insight in the economics of
ORC technology by estimating the capital costs of their designed system based on its technical
setup. This practice of bottom-up cost estimation is commonly used in process engineering
industries and now increasingly applied in ORC research. Most of these studies have an
engineering perspective and suggest innovations with respect to ORC system setup or aim for
improvements in the technology’s performance. Recognizing the influence of the system design
on its costs and the importance of the economic evaluation on its adoption and diffusion, these
studies expand their technical insights with a perspective on the costs of the designed system.
But the large majority of these ORC cost engineering studies do not interpret the economic
results in terms of their accuracy or with respect to their financial feasibility. Nevertheless, each
of the cost estimation methods is characterized by inherent limits to its accuracy, i.e. the extent
to which the estimated value is close to the true value. Therefore, the goal of this chapter is to
provide a critical perspective on the estimation of ORC investment costs and its interpretation.
The aim is to complement the existing literature on ORC cost estimation in order to encourage
more nuanced interpretations in future ORC cost engineering studies.
This chapter is structured as follows. Section 3.2 discusses established cost engineering
practices. The goal of this section is to introduce multiple, exiting methods developed to
estimate the investment costs of industrial plants, to discuss their approach and their
corresponding expected accuracy. In order to evaluate each of these methods in practice,
section 3.3 introduces the economics of an existing heat recovery project. Then, multiple cost
estimation approaches are applied to estimate the costs of this case study. Section 3.4 discusses
the most important findings from this chapter and a final section draws the concluding remarks.
3.2. Cost estimation for industrial plants
The up-front estimation of the investment costs of a new plant is a challenging task, iterating as
the design evolves to increased detail. Plant estimates are classified according to their level of
detail and thus their accuracy (Table 3.1). The accuracy of an estimate is understood as the
extent to which it is close to the actual value, in this caser the real system costs. The accuracy
ranges indicate variations regarding technological complexity of the project, suitable reference
CHAPTER THREE
78
information, and an appropriate determination of project contingencies (AACE International,
2005). The ranges represented in Table 3.1 are applicable for process industry projects. They
represent the typical percentage variation, with a 90 % confidence, of the actual costs from the
estimated costs, after the application of a contingency. (AACE International, 2005)
Underestimation of capital costs occurs mainly due to incomplete listing of all the equipment
needed in the process (Turton, Bailie, Whiting, Shaeiwitz, & Bhattacharyya, 2013). An increasing
level of detail implies a smaller accuracy range, but similarly an increasing amount of effort and
labour hours to make the estimate. Estimates performed in research are generally order-of-
magnitude, study and preliminary design estimates.
The equipment needed for construction of the plant is at the core of most cost estimates. The
best approach for the purchase cost of a piece of equipment is a current vendor’s price quote.
Data from previously bought but similar equipment are the next best alternative (Turton et al.,
2013). When the costs of a component are known but its capacity differs from that of the to-be-
estimated component, the costs can be roughly estimated using the correlation
ca
cb= (
Aa
Ab)
n
(3.1)
with Aa the equipment cost attribute of the required component; Ab the equipment cost attribute of the known component; ca the purchase costs of the required component; cb the purchase costs of the known component; n the exponent used to correlate the costs.
The exponent n differs per type of equipment, but it is often close to 0.6 for the chemical
industry. Therefore this extrapolation method is sometimes referred to as the six-tenths rule. It
provides only rough approximations of the actual costs. In case no purchased equipment costs
are known, but technical details are available, the costs can be estimated using equipment cost
correlations. Guidance, exponents and correlations for various types of process equipment are
provided by e.g., Bejan, Tsatsaronis, and Moran (1996), Couper, Penney, Fair, and Walas (2012),
Smith (2005), Towler and Sinnot (2008) and Turton et al. (2013).
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
79
Table 3.1. Classification of capital cost estimates for process industry projects.
Class Type of Estimate Description Accuracy Ranges
5 Order-of-magnitude estimate (also Ratio/Feasibility)
Based on limited information. Concept screening.
Low: −20% to −50%
High: +30% to +100%
4 Study estimate (also Major Equipment/Factored)
List of major equipment. Project screening, feasibility assessment, concept evaluation, and preliminary budget approval.
Low: −15% to −30%
High: +20% to +50%
3 Preliminary design estimate (also Scope)
More detailed sizing of equipment.
Low: −10% to −20%
Budget authorization, appropriation, and/or funding.
High: +10% to +30%
2 Definitive estimate (also Project Control)
Preliminary specification of all the equipment, utilities, instrumentation, electrical and off-sites.
Low: −5% to −15% High: +5% to +20%
Control or Bid/Tender.
1 Detailed estimate (also Firm/Contractor’s)
Complete engineering of process and related off-sites and utilities required.
Low: −3% to −10%
High: +3% to +15%
Check Estimate or Bid/Tender. Legend: The table gives an overview of the classification of cost estimation techniques. The cost estimation techniques are
typically classified into five categories, each with a corresponding accuracy. The better the accuracy, the more resources to be
committed to execute the estimate. Sources: AACE International (2005); Turton et al. (2013).
The total capital investment of a project can be estimated using various techniques. A simple
method is to use a capacity exponent ratio, similarly as previously described for equipment cost
estimates. The costs of a planned plant are estimated using the known costs of a similar
previously constructed plant. The accuracy of this method is rather low. It should be used for
order-of-magnitude or study estimates only (Peters, Timmerhaus, & West, 2004). Step count
methods take a different approach and utilize the number of functional units or plant sections
as a basis to estimate total investment costs. This method is designed for use in the chemical
process industry and not so suitable for usage in other manufacturing fields. The accuracy would
be in the range of order-of-magnitude estimates (Towler & Sinnot, 2008). Thirdly, factorial
estimation techniques are based on the costs of the major purchased equipment items and
apply multiplication factors to obtain the total capital investment. The Lang Factor method is
probably the first factorial method. Lang suggested multiplying the total delivered costs of the
major equipment parts with a factor that differs according to the type of process. The factors
are available for solid, fluid and mixed fluid-solid processing chemical plants (Towler & Sinnot,
2008). The Lang Factor technique utilizes only one multiplication factor and is therefore
expected to yield lower accuracies, it is suggested to use for order-of-magnitude estimates
(Peters et al., 2004). The Lang Factor method has been adapted numerous times since then. For
instance, Hand suggested utilizing multiplication factors for the equipment types instead of the
CHAPTER THREE
80
plant type (Towler & Sinnot, 2008). The utilization of multiple factors implies more detail, but
this method would probably still not provide very good accuracies. The detail of the estimate
can be improved further using cost factors for different items related to direct costs (erection of
equipment, piping, electrical, instrumentation and control, buildings and structures, ancillary
buildings, storage, utilities, site preparation). Dividing the process into subunits and applying
factors per subunit function improves the estimate’s accuracy and reliability (Towler & Sinnot,
2008). An even more detailed estimate is suggested by Guthrie and accounts for the installation,
piping and instrumentation costs of each equipment item individually. Inclusion of a factor for
the equipment materials used is said to improve accuracy even more (Towler & Sinnot, 2008).
Still, these estimates would have the accuracy of preliminary estimates. Another, somewhat
different, factorial method calculates the direct fixed costs and total investment costs as
percentages of the delivered-equipment costs. The factors used depend on, among other
things, the process type, design complexity, location and experience. This percentage of
delivered-equipment method is suitable for study and preliminary estimates (Peters et al.,
2004). When the goal is to achieve more detailed estimates than the ones formerly described,
this requires more detailed information and engineering effort. For instance, the unit cost
method is used for preliminary and definitive estimates. The method requires accurate
information on costs from previous projects, detailed estimates of equipment prices,
installation labour, instrumentation, electrical and other miscellaneous items. Also engineering
hours, drawing efforts, construction, contractor’s fee and contingencies are included. This can
yield relatively accurate results but requires sufficiently detailed information and engineering
time (Peters et al., 2004). Detailed item estimates, with high accuracies, generally concern
advanced project plans. At this stage, most details of the project are known, the drawings are
finished and the estimates are based preferably on delivered quotations. For most research and
development projects, both definitive and detailed cost estimates would range beyond the
scope of the project and the information available. Preliminary estimates are feasible, but the
accuracy of the results relies strongly on the quality of the information (i.e. factors) used.
Finally, the costs of materials and labour are subject to inflation which implies cost figures from
different years are not directly comparable. The most straightforward manner to update
historical data is by means of composite cost indices, using the equation
cj = ci × (
Ij
Ii)
(3.2)
with ci the costs in year i; cj the costs in year j;
Ii the cost index for year i; Ij the cost index for year j.
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
81
The composite cost index I is a weighted average index of various components costs commonly
used in a particular industry. Updating cost data using cost indices is acceptable for only shorter
periods of time; some say four to five years (Jelen & Black, 1983). The accuracy of the results
decreases when longer time periods are used.
3.3. Comparing the estimated and actual costs of a heat recovery ORC
system
This section presents the results of various cost estimation approaches to give a perspective on
their methodology and accuracy. First of all, the investment costs of an actual heat recovery
case are discussed. Then, the costs of this case study are estimated using both the costs of other
ORC systems as well as the technical parameters of the case study itself.
3.3.1. Case study: ORC for industrial heat recovery
The ORC system was installed in 2013 in Flanders, Belgium to recover excess heat from an
industrial plant. The excess heat is a low-medium temperature (range 150–250 °C) flue gas
stream from an industrial kiln. The ORC module was integrated into the plant using an
intermediate thermal oil circuit including a flue gas heat exchanger. The ORC unit itself has a
gross power output of 375 kWgr and is composed of a centrifugal pump, a one-step radial
expander and a generator. The evaporator is a plate heat exchanger and condensation occurs
air-cooled. The project has a SIC of 4216 €2013/kWgr, including integration (oil circuit, piping …)
and installation (delivery, construction and project management). Figure 3.1 displays the
partitioning of the capital costs. A major share of total costs stems from the ORC module itself,
which includes the pump, expander and generator. The intermediate thermal oil circuit
represents about 11% of total investment costs.
Figure 3.1. Diagram of the investment costs of the ORC case study.
Legend: The figure displays the distribution of the investment costs of an ORC installed in 2013 in Flanders, Belgium, and used for
excess heat recovery from an industrial kiln. Approximately half of the capital costs are for the account of the ORC unit, which
includes the expander, the generator and the pump.
CHAPTER THREE
82
3.3.2. Using cost data of other systems: the capacity exponent ratio method
Using the capacity exponent ratio method, the costs of a piece of equipment or a complete
project are estimated based on knowledge about the costs of a similar system from a different
capacity or scale. For instance, Ghirardo, Santin, Traverso, and Massardo (2011) investigate the
options to recover heat from on-board fuel cell systems, including an ORC system. The
reference ORC module costs 1.7 M€ for 1115 kW output, when an intermediate thermal oil
system is included the total investment costs of the reference plant are calculated at 2.3 k€.
There is no information on the time frame in which this reference price is collected. Using a
scaling exponent of 0.867, the authors estimated the capital investment of the 35 kW system as
follows: ca = cb (Aa
Ab)
n
= 2,345,000 (Aa
1115)
0.876
= 117 k€ (Ghirardo et al., 2011).4
Several questions arise when performing a cost estimate with the Capacity Exponent Ratio. First
of all, which ORC system is best to use as reference case? The most appropriate to use is a
system with similar characteristics, but the available information on real ORC costs is scarce. A
possible reference case for the study in this chapter is given by for instance Forni, Vaccari, Di
Santo, Rossetti, and Baresi (2012), who give insight into the feasibility of potential example
cases based on cost data from an experienced manufacturer. Alternatively, David, Michel, and
Sanchez (2011) discuss two potential heat recovery case studies. The costs of the system are
roughly gauged using prices of existing ORC modules and estimated additional costs for
integration and installation. For the steel mill project, two ORC units would be needed to obtain
250 kWgr output, at a cost of 500 €/kWe for the units and 1080 €/kW for the complete project.
A third option is a budget offer made to the City of Unalaska in 2012, which states a price of k$
185 for a 50 kW unit, to be used for heat recovery (Whealy, Taylor, & George, 2012). The
complete project, including labour and other project contingencies, is quoted at k$ 1889
(Whealy et al., 2012). Finally, a 5.5 MW heat recovery demonstration project installed in 2006 in
the USA had a cost of 2500 $/kW (Leslie, Sweetser, Zimron, & Stovall, 2009; Sweetser & Leslie,
2007). Besides a reference case, the Capacity Exponent Ratio method requires a suitable
exponent for the scaling. The commonly derived exponent for the chemical engineering industry
is 0.6; Ghirardo et al. (2011) use 0.867. Due to lack of sufficient data to calculate an exponent
for ORC heat recovery systems specifically, the standard exponent of 0.6 is used for this
analysis. The results of the cost extrapolation are displayed in Table 3.2, using the investment
costs available in each of the discussed potential references (David et al., 2011; Forni et al.,
2012; Leslie et al., 2009; Sweetser & Leslie, 2007; Whealy et al., 2012).
The project costs are calculated for information purposes (see Table 3.2), but because the
integration requirements are so different per case these cannot be compared to the costs of the
4 The scaling exponent 0.867 originates from Moynihan (1983). Note that such a four-digit scaling exponent
assumes a very accurate knowledge on the economies of scale of ORC systems.
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
83
case study. The module costs of the 375 kW case study are estimated between 1653 and 3337
€2013/kW, with an average of 2288 €2013/kW. This variation can have various origins, such as
manufacturer experience or margins, inclusion of freight or engineering costs or not. Comparing
these results to the costs of the case study, we would have expected case no. 6 to yield good
results since the heat recovery setup and the suggested ORC module are similar to that of the
case study. Similarly, case no. 7 concerns an actual budget quote and was expected to be
representative. Although case no. 6 and 7 lead to mutually similar results, there is a large
deviation with the 3280 €2013/kW of the real case. This, again, can be due to the specific sales
circumstances, due to the fact that the estimated costs do not take account of some essential
costs or perhaps due to overpricing of the real case. Because the detailed cost decomposition of
the reference cases is not sufficiently known, it is not possible to obtain very accurate estimates
or comparisons. Thus, the capacity exponent ratio method can be used only for order-of-
magnitude or study estimates (Peters et al., 2004). The accuracy of this estimation method for
ORC investments may increase as more and sufficiently detailed knowledge about actual ORC
systems is published.
Table 3.2. Cost estimation using the capacity exponent ratio method.
No. Reference Reference
Gross
Power (kW)
Reference
Module SIC
Reference
Project SIC
Estimated
Module SIC
(€2013/kW)
Estimated
Project SIC
(€2013/kW)
1 Forni et al. (2012) 1100 1818 €2012/kW 2818 €2012/kW 2796 4334
2 Forni et al. (2012) 1300 1154 €2012/kW 2923 €2012/kW 1897 4806
3 Forni et al. (2012) 5300 943 €2012/kW 3321 €2012/kW 2721 9579
4 Forni et al. (2012) 5400 1148 €2012/kW 2593 €2012/kW 3337 7535
5 Forni et al. (2012) 160 2594 €2011/kW 3375 €2011/kW 1845 2401
6 Forni et al. (2012) 250 2080 €2011/kW 4320 €2011/kW 1769 3673
7 Whealy et al.
(2012)
50 3700 USD2012/kW - 1653 -
8 Whealy et al.
(2012)
150 - 12,596 USD2012/kW - 8731
9 Leslie et al.
(2009); Sweetser
and Leslie (2007)
5500 - 2500 USD2006/kW - 7319
Legend: The table gives a summary of the cost estimation using the capacity exponent ratio method. The costs of a 375 kW case
study are estimated using costs data from other publications and a scaling exponent of 0.6. The module costs show very
diverging results. The project costs are calculated but cannot be considered because of the diverging integration requirements of
the reference cases. Note that the results have been displayed as four-digit costs. This is done solely to show the numerical
variation when different cases are used and when these numbers are interpreted as valid. Because the expected accuracy of the
capacity exponent ratio method is relatively low, the results of this method should not be interpreted in such detail.
CHAPTER THREE
84
3.3.3. Using technical parameters of the system: factorial estimation techniques
Factorial costs estimation techniques use information about the technical details of the
designed system itself, rather than cost information of other but similar systems. Again, the
purchased equipment costs form the basis of the total investment costs, but these are now
estimated using cost correlations for all major equipment items. The total investment costs are
then derived from these purchased equipment costs, using either multiplication factors or
percentages.
Estimating the purchased equipment costs
The costs of the major equipment items are commonly estimated from publicly available
equipment cost correlations, which have been established for a large number of commonly
used industrial equipment. For ORC systems, the major equipment items are the basic
components: the evaporator, expander and generator, condenser and pump. Because there is
little information about the component costs of actual ORC systems, most researchers use
publicly available correlations for standard equipment items. For instance, Lee, Kuo, Chien, and
Shih (1988) were early to use component correlations to estimate the costs of an ORC heat
recovery system. Quoilin, Declaye, Tchanche, and Lemort (2011) combine correlations from
literature (Bejan et al., 1996) with self-derived ones. Walraven, Laenen, and D’haeseleer (2015)
combine correlations from various sources (Kloppers, 2003; Smith, 2005; Towler & Sinnot, 2008)
for their analysis. Desai and Bandyopadhyay (2009) combine correlations from various ORC
research papers. An increasing number of researchers utilize the correlations from Turton et al.
(2013) for ORC cost estimation (e.g., (Lazzaretto, Toffolo, Manente, Rossi, & Paci, 2011; Toffolo,
Lazzaretto, Manente, & Paci, 2014; F. Yang et al., 2015; M.-H. Yang & Yeh, 2016; Yu, Shu, Tian,
Wei, & Liang, 2016)).
A similar procedure is performed to estimate the purchased equipment costs of the case study
discussed in this chapter. Knowing the technical parameters of the system, the component costs
are estimated from publicly available correlations. The evaporator, expander and pump costs
are calculated from the correlations of Turton et al. (2013), of the form
log10 Cp0 = K1 + K2 log10(A) + K3[log10(A)]2
(3.3)
with A the equipment attribute; Cp
0 the purchased equipment costs at ambient pressure and using carbon steel;
K1, K2 and K3 the coefficients determined for the type of equipment.
The condenser fan costs are estimated with a correlation from Smith (2005) and for the
generator the function available in Toffolo et al. (2014) is applied. The coefficients and
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
85
correlations are summarized in Table 3.3. All estimates are converted to 2013 Euros to allow for
comparison with the actual costs of the case study. The correlations from Turton et al. (2013)
yield values in USD2001. These are converted to Euros using a 1.1162 exchange rate (average
2001) and updated to 2013 using the Chemical Engineering Plant Cost Index (CEPCI), with
CEPCI2001 and CEPCI2013 values of 397 and 587.3 respectively. The results obtained from Smith
(2005) are converted from USDJan,2000 to €2013 using an exchange rate of 0.9857 (average Jan
2000) and CEPCI2000 equal to 394.1. The correlation from Toffolo et al. (2014) is in Euros and was
first published in 1993, so a CEPCI1993 of 359.2 was used.
Table 3.3. Coefficients and correlations for estimation of the purchased equipment costs.
Component Coefficients and Correlations Ref.
Evaporator K1 = 4.6656 K2 = −0.1557 K3 = 0.1547 Turton et al. (2013) Expander K1 = 2.2476 K2 = 1.4965 K3 = −0.1618 Turton et al. (2013) Pump K1 = 3.3892 K2 = 0.0536 K3 = 0.1538 Turton et al. (2013) Condenser
Cp0 = 12,300 × (
Q
50)
0.76with Q in kW
Smith (2005)
Generator Cp
0 = 1,850,000 × (P
11,800)
0.94with P in kW
Toffolo et al. (2014)
Legend: Overview of the correlations and the coefficients used to estimate the purchased equipment costs of the case study.
The results of the cost estimate are presented in Table 3.4. The module costs of the case study,
composed of the components but without integration costs, amount 3280 €2013/kW. The SIC are
estimated at 1843 €2013/kW when only the essential ORC components are considered. These
costs are often interpreted as the costs for the ORC module, but are significantly lower (−44%)
than the actual figure. However, this comparison may not be completely justified because part
of this difference could be due to the fact that the real module costs already include additional
costs such as freight, engineering or instrumentation. Estimating the purchased equipment
costs as the ORC module costs can give a first proxy but will likely underestimate the real
detailed composition of the ORC module.
Table 3.4. Factorial estimation of the purchased equipment costs.
Component 𝐂𝐩𝟎 (k€2013)
Evaporator 314 Expander 188 Pump 16 Condenser 47 Generator 126
Total 691
Specific investment costs (€2013/kWgr) 1,843 Legend: Results of the estimated purchased equipment costs of the case study,
calculated using component cost correlations.
CHAPTER THREE
86
A comparison of the real (a) and the estimated (b) component costs for the case study is
presented in Figure 3.2. The ORC module is composed of the evaporator, condenser and the
ORC unit which includes the expander, generator and pump. These are the purchased
equipment costs excluding integration and other additional costs. Figure 3.2 demonstrates not
only the difference between the real and the estimated costs, but also highlights the significant
divergence of the cost distributions. The expectation was that there could be a deviation in the
absolute values, but that the cost distribution would be more representative. Nevertheless,
there are large discrepancies. For instance, the evaporator constitutes only 10% of the real
purchased equipment costs but is estimated at a 45% share. The ORC unit, including expander,
generator and pump, represents 69% of actual purchased equipment costs but its share is
estimated at 47%.
Figure 3.2. (a) Real and (b) estimated purchased equipment costs of the case study.
Legend: The actual costs of the components of the case study are compared to the results of the estimated purchased equipment
costs, calculated using component cost correlations. The figure shows a wide diversion in both the absolute values and the
distribution of the components in the total costs.
Estimating the total investment costs: multiplication factors
The utilization of a multiplication factor to account for direct (purchased equipment, piping,
instrumentation and controls …) and indirect (engineering and supervision, construction,
contingencies) costs is attributed to Lang. The method has been adapted several times
throughout the years aiming for better, more accurate results. Also for ORC research cost
estimation this technique is more often used: the total investment costs are estimated using the
estimated purchased equipment costs. Some multiply the purchased equipment costs by a
factor of 6.32 to obtain the total investment costs (e.g., Heberle, Bassermann, Preißinger, and
Brüggemann (2012); Heberle and Brüggemann (2015, 2016)), as suggested by Bejan et al.
(1996). Note that this 6.32 multiplication factor is suggested for the erection of new systems; for
expansion of existing systems a factor of 4.16 is proposed (Bejan et al., 1996). Nevertheless,
such a general multiplication factor fails to take account of the project specific integration
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
87
requirements. The necessary piping, heat exchange or other intermediate equipment differs
strongly per case, but also per type of application. For new geothermal projects the integration
requirements are very likely much higher than for heat recovery projects where the base plant
is already operational. Suppose the 4.16 multiplication factor, for expansion of existing systems,
is used to estimate the case study’s costs. Using the estimated purchased equipment costs of
1843 €2013/kW as a basis, this would give a total investment cost of 7667 €2013/kW. This is a
deviation of 82% from the actual 4216 €2013/kW.
Other factorial approaches are applied by for instance Huang, McIlveen-Wright, et al. (2013);
Huang, Wang, et al. (2013), who use the ECLIPSE program, designed for technical,
environmental and economic process analysis, to estimate the costs of the proposed biomass
ORC systems. Finally, the module costing technique illustrated by Turton et al. (2013) is a
commonly used factorial cost estimation method based on the approach of Guthrie (Turton et
al., 2013). It is increasingly applied in ORC research (see e.g., Lazzaretto et al. (2011); S.
Lecompte, Lazova, van den Broek, and De Paepe (2014); Steven Lecompte, Lemmens,
Huisseune, van den Broek, and De Paepe (2015); Toffolo et al. (2014); M.-H. Yang and Yeh
(2015, 2016)). This technique is suitable for preliminary estimates in the −20% to +30% accuracy
range. The costs of the major purchased equipment parts are estimated based on technical
characteristics (Turton et al., 2013). These base costs Cp0 are multiplied with the bare module
cost factor FBM, which accounts for operating pressures and specific materials of construction,
as well as direct and indirect project expenses. This yields the bare module costs CBM (Turton et
al., 2013):
CBM = Cp0FBM
(3.4)
with CBM the bare module costs; Cp
0 the purchased equipment costs at ambient pressure and using carbon steel;
FBM the bare module cost factor.
The project costs CTM (referred to as the total module costs in (Turton et al., 2013)) include the
integration of the plant into an existing facility. They are obtained by adapting the bare module
costs with another multiplication factor (Turton et al., 2013):
CTM = 1.18 ∑ CBM,i
𝑛
𝑖=1
(3.5)
with CBM the bare module costs; CTM The project costs (referred to as the total module costs in Turton et al. (2013)); 𝑛 the total amount of equipment parts.
CHAPTER THREE
88
This approach is applied to estimate the total project costs of the case study, starting from the
purchased equipment costs estimate. Accounting for the used materials and pressures, the
direct (equipment, installation materials and labour) and indirect (freight, insurance, taxes,
overhead and engineering expenses) costs associated with the project yields a specific bare
module cost of 4390 €2013/kW. Finally, the total module costs include also contingencies and
contractor fees and auxiliary facilities. They are obtained by multiplying the bare module costs
with a factor 1.18 and are estimated at 5180 €2013/kW. The results are shown in Table 5. Based
on these estimates the ORC total module costs are almost three times as high as the costs of the
module itself, composed of the major equipment items.
The real project costs are composed of the equipment, but also the costs for integration and
installation and amount 4216 €2013/kW. The estimated total module costs amount 5180
€2013/kW, which is 23% higher. These total module costs are commonly interpreted as the total
costs associated with an ORC project, including the ORC module itself and its integration into an
existing plant or into the facilities for conversion of the energy source. However, these total
costs have been obtained using multiplication factors that have been established based on
other industrial applications. In practice, the actual requirements for integration differ per
application. For this heat recovery case study, the integration involved an intermediate thermal
oil circuit and associated piping. The integration costs and the costs charged by the ORC vendor
to install the system amount 22% of the total project costs. For the estimate the module costs
represent only 35.6% of the total, the remainder interpreted as the integration and other
project costs.
Table 3.5. Total investment costs estimation with the Module Costing Technique.
Component 𝐂𝐩𝟎 (k€2013) 𝐂𝐁𝐌 (k€2013) 𝐂𝐓𝐌 (k€2013)
Evaporator 314 680 Expander 188 657 Pump 16 53 Condenser 47 66 Generator 126 189
Total 691 1646 1943
Specific investment costs (€2013/kWgr) 1843 4390 5180
Legend: Results of the estimated purchased equipment costs of the case study, using the Module Costing Technique (Turton et
al., 2013). The purchased equipment costs are the basis to estimate the bare module costs and the total project costs.
Estimating the total investment costs: percentages of delivered equipment costs
The percentages of delivered equipment approach starts from the purchased equipment costs
and calculates the other cost factors as percentages of the former. The methodology and
applicable percentages are illustrated by Bejan et al. (1996) and Peters et al. (2004). The applied
percentages are based on the average percentages suggested by Bejan et al. (1996). Starting
from the estimated purchased equipment costs of 1843 €2013/kW, the total investment costs are
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
89
Table 3.6. Total investment costs estimation using percentages of delivered equipment costs.
Cost Breakdown Percentage Range Applied Percentage (Bejan et al., 1996)
Cost Estimate (€2013/kW)
1. Fixed-capital investment (FCI)
1.1. Direct fixed-capital investment (DFCI)
1.1.1. Onsite costs (ONSC) Purchased-equipment cost (PEC)
15%-40% of FCI (Bejan et al., 1996; Peters et al., 2004)
/ 1843
Purchased-equipment installation
6%–14% of FCI; 20%–90% of PEC (Bejan et al., 1996; Peters et al., 2004)
45% of PEC
826
Piping 4%–17% of FCI (Peters et al., 2004); 3%–20% of FCI (Bejan et al., 1996); 10%–70% of PEC (Bejan et al., 1996)
31% of PEC
571
Instrumentation and controls 2%–12% of FCI (Peters et al., 2004); 2%–8% of FCI (Bejan et al., 1996); 6%–40% of PEC (Bejan et al., 1996)
10% of PEC
184
Electrical equipment and materials
2%–10% of FCI (Bejan et al., 1996; Peters et al., 2004); 10%–15% of PEC (Bejan et al., 1996)
11% of PEC
203
1.1.2. Offsite costs (OFSC) Land 0%–2% of FCI (Bejan et al., 1996);
1%–2% of FCI (Peters et al., 2004); 0%–10% of PEC (Bejan et al., 1996)
/ 0
Civil, structural, and architectural work
5%–23% of FCI; 15%–90% of PEC (Bejan et al., 1996)
44% of PEC 811
Service facilities 8%–20% of FCI (Bejan et al., 1996); 8%–30% of FCI (Peters et al., 2004); 30%–100% of PEC (Bejan et al., 1996)
20% of PEC 369
Buildings 2%–18% of FCI (Peters et al., 2004) / 0
Yard improvements 2%–5% of FCI (Peters et al., 2004) / 0
Total DFCI 4811 1.2. Indirect Fixed-Capital Investments (IFCI)
Engineering and supervision 4%–21% of FCI (Peters et al., 2004); 4%–21% of FCI (Bejan et al., 1996); 6%–15% of DFCI (Bejan et al., 1996); 25%–75% of PEC (Bejan et al., 1996)
30% of PEC 553
Construction costs including contractor’s profit
4%–17% of FCI (Peters et al., 2004); 6%–22% of FCI (Bejan et al., 1996); 15% of DFCI (Bejan et al., 1996)
15% of DFCI 722
Contingencies 5%–15% of FCI (Peters et al., 2004); 5%–20% of FCI (Bejan et al., 1996); 8%–25% of all direct and indirect costs, without legal costs (Bejan et al., 1996);
10% of FCI 692
Legal costs 1%–3% of FCI (Peters et al., 2004) 2% of FCI 138
Total IFCI 2105
2. Other outlays
Start-up costs 5%–12% of FCI (Bejan et al., 1996) 10% of FCI 692
Working capital 10%–20% of TCI (Bejan et al., 1996) / 0
Costs of licensing, research, and development
/ / 0
Allowance for funds used during construction (AFUDC)
/ / 0
Total capital investment 7608
CHAPTER THREE
90
estimated as shown in Table 3.6. The resulting value of 7608 €2013/kW is much higher than the
actual costs, but close to the 7667 €2013/kW obtained with the simple 4.16 factor multiplication.
This is not surprising, since the general multiplication factors proposed by Bejan et al. (1996)
have been derived from this percentage approach.
3.4. Discussion of the results
The goal of this chapter was to identify established cost estimation methods and evaluate them
for use within the frame of ORC research. This section summarizes and discusses the most
important findings from this chapter. First of all, an exploration of established practices reveals
there are multiple approaches to estimate the costs of an industrial system design. One
approach is to use knowledge about the economics of an existing system and use a scaling
factor to estimate the costs of the newly anticipated system. Necessary conditions for this
method include, first of all, knowledge about the proper scaling factor for the type of
component or energy system under consideration, and secondly, that the known and the
anticipated system are sufficiently similar to justify a simple scale relation between them. A
second approach to estimate the costs of a designed system starts not from the economics of
similar systems, but from the technical characteristics of the system itself. In this case, the costs
of the essential components are typically estimated with component-specific costs correlations
and the costs of the complete system are estimated based on these essential component costs,
using either multiplication factors or percentages. To estimate the suitability of these
techniques to draw conclusions about the costs of newly designed ORC systems, this chapter
applied each of them to estimate the costs of a heat recovery ORC case study. The results of this
examination are summarized in Figure 3.3: the actual ORC module and entire project costs of
the case study are displayed, as well as the estimated values and their corresponding accuracy
ranges. Whereas the module costs of the actual system amount 3280 €2013/kW, they were
estimated between 1653 and 3337 €2013/kW using scaling methods at 1843 €2013/kW when
commonly used component cost correlations were applied. The results for the total project
investment cost display an even wider variation. The actual total investment costs amount 4216
€2013/kW, but were estimated at 7667 €2013/kW in case one single factor was used to relate the
project costs to the module costs. When the module costing technique was applied the project
costs were gauged at 5180 €2013/kW and using percentages of delivered equipment costs
yielded 7608 €2013/kW. Note that this analysis compares the estimated and real costs for only
one case study. For instance, Toffolo et al. (2014) perform a similar analysis and they report only
a very small difference between estimated and actual costs.
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
91
Figure 3.3. Summary of the real and the estimated investment costs for the case study.
Legend: The actual costs of the case study are compared to the results using various cost estimation methods. CER = Capacity
Exponent Ratio; Comp. Corr. = component correlations; Mult. Fact. = multiplication factor, MCT = Module Costing Technique;
PDEC = percentages of delivered equipment costs.
There are several potential reasons for these deviations between the actual and the estimated
investment costs. Firstly, the cost estimation techniques used within the frame of research
typically have accuracies in the range of order-of-magnitude, study or preliminary design
estimates. Capacity exponent ratio methods have expected accuracies suitable for order-of-
magnitude or study estimates, which means that deviations of −50% to +100% are not
uncommon. Factorial estimation methods such as the module costing technique are suitable for
preliminary estimates, with expected accuracies between −20% and +30%. The best results are,
as expected, obtained with the module costing technique. Second, the accuracy of factorial
estimation methods depends strongly on the quality of the information that was used to
establish the multiplication factors. Because there is little public information about the actual
costs of ORC systems and their components, approximations from other industrial fields are
used. The quality of the estimates can likely be improved by developing cost correlations and
multiplication factors for ORC technology in particular. Additional inaccuracies and uncertainties
stem from treatment of costs over time periods. Extrapolation of costs over large periods of
time decreases the accuracy of the results. Most of the open-source correlations available in
text books are at least nine years old and thus provide less accurate results. Moreover, some of
these references refer back to original factors and correlations published by Guthrie in 1969 or
1974 and updated with few recent data points or using cost indices. This makes these
correlations less reliable. Also, the choice of indicator for cost escalation and local conditions
may have an influence and create additional deviations. Furthermore, it is important to
acknowledge the difference between costs and prices. Costs reflect the amount that is required
to produce a certain item, whereas the price is the amount you pay to purchase it. The costs
0
2
4
6
8
10
12
14
16
18
20
Spe
cifi
c In
vest
me
nt
cost
s [k
€2
01
3/kW
]
Module
Project
CHAPTER THREE
92
associated with producing an ORC system will thus differ from the price paid to acquire that
system. Many correlations used to estimate costs are obtained using vendor prices. In cases
where the ORC developer would purchase most equipment instead of developing, it this is not a
problem. For innovative system designs (e.g., expanders) this method would be less suitable.
Finally, for the specific case of ORC projects the costs depend strongly on the type of
application. Whereas the estimated “total module costs” are often interpreted as the complete
ORC project costs, these are all using the same multiplication factors and fail to take account of
the diversity of integration requirements. ORC projects using biomass, geothermal, solar or
excess heat have different integration requirements and corresponding costs. These
multiplication factors have been established for applications in mostly the chemical engineering
industry, so their use for other industries may be less reliable.
3.5. Chapter conclusions
The adoption and diffusion of innovative technologies is strongly influenced by economic
aspects. For ORC systems, the capital costs have a major influence on the financial feasibility of
the investment project. The importance of the economic dimension is increasingly recognized
and an expanding number of researchers incorporate an economic section in their work.
Because there is still not much published information about the actual costs of ORC systems and
their components, the capital costs of the proposed designs are often estimated from publicly
available component correlations or aggregated cost data. The aim of this chapter is to provide
a perspective on these cost engineering techniques and their applicability for of ORC systems.
There are several methods available to estimate the costs of industrial projects. Simpler
approaches require less effort but yield lower accuracy. High accuracies are possible with
definitive and detailed estimates, but these require a far-reaching detail of plant design. Cost
estimates pursued in the frame of research generally achieve accuracies in the range of order-
of-magnitude, study or preliminary estimates. The results of this chapter suggest that the
estimated costs can differ considerably according to the applied approach and from the actual
costs. This is partly due to the inherent inaccuracies entailed in cost estimation, but likely also
due to the fact that the cost estimation methods have been established for process industries in
general. Most of the equipment in the ORC module is fairly standard, but for more specialized
components, such as the expander, the actual costs will differ from the costs estimated based
on correlations for conventional equipment. The same remark holds for the estimation of the
integration costs. These are very project-specific and can likely not be estimated very accurately
using simple multiplication factors from other industrial processes.
These findings have several implications. First and foremost, estimated cost figures should be
accompanied by accuracy ranges and a discussion on the reliability of the used methods. This to
avoid misinterpretation of the validity of the results. The correlations that are currently
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
93
available are useful to mutually compare various system designs, where the proportional
comparison is more important than the exact outcomes. The results are relative rather than
absolute. Secondly, proper delineation of the scope of the investigation is key to enhance
usability and comparability of the results. This relates to defining whether the module or the
project costs are considered, and which cost aspects are included or not, as well as providing a
date stamp for any cost figure. Finally, estimating the total investment costs of an ORC project
by extrapolation or multiplication does not take into account the case-specific integration
requirements. These are more difficult to estimate because of their diversity, but could be
approximated when there are sufficient real data to identify average multiplication factors per
application type. The accuracies of ORC cost estimates could increase as more technology-
specific cost information becomes available.
CHAPTER THREE
94
Bibliography
AACE International. (2005). Recommended Practice No. 18R-97. Cost estimate classification system - As applied in engineering, procurement, and construction for the process industries. TCM Framework: 7.3 – Cost Estimating and Budgeting. AACE International: 10 pp.
Bejan, A., Tsatsaronis, G., & Moran, M. (1996). Thermal design and optimization New York: Wiley & Sons, 542 pp.
Couper, J. R., Penney, R. W., Fair, J. R., & Walas, S. M. (2012). Chemical Process Equipment: Selection and Design (Third ed.): Butterworth-Heinemann, 838 pp.
David, G., Michel, F., & Sanchez, L. (2011). Waste heat recovery projects using Organic Rankine Cycle technology - Examples of biogas engines and steel mills applications. Paper presented at the World Engineers' Convention, Geneva, Switzerland.
Desai, N. B., & Bandyopadhyay, S. (2009). Process integration of organic Rankine cycle. Energy, 34(10), 1674-1686. doi:10.1016/j.energy.2009.04.037
Forni, D., Vaccari, V., Di Santo, D., Rossetti, N., & Baresi, M. (2012, 11–14 September 2012). Heat recovery for electricity generation in industry. Paper presented at the ECEEE 2012 Summer Study on energy efficiency in industry, the Netherlands.
Ghirardo, F., Santin, M., Traverso, A., & Massardo, A. (2011). Heat recovery options for onboard fuel cell systems. International Journal of Hydrogen Energy, 36(13), 8134-8142. doi:10.1016/j.ijhydene.2011.01.111
Heberle, F., Bassermann, P., Preißinger, M., & Brüggemann, D. (2012). Exergoeconomic optimization of an Organic Rankine Cycle for low-temperature geothermal heat sources. International Journal of Thermodynamics, 15(2). doi:10.5541/ijot.374
Heberle, F., & Brüggemann, D. (2015). Thermo-Economic Evaluation of Organic Rankine Cycles for Geothermal Power Generation Using Zeotropic Mixtures. Energies, 8(3), 2097-2124. doi:10.3390/en8032097
Heberle, F., & Brüggemann, D. (2016). Thermo-Economic Analysis of Zeotropic Mixtures and Pure Working Fluids in Organic Rankine Cycles for Waste Heat Recovery. Energies, 9(4), 226. doi:10.3390/en9040226
Huang, Y., McIlveen-Wright, D. R., Rezvani, S., Huang, M. J., Wang, Y. D., Roskilly, A. P., & Hewitt, N. J. (2013). Comparative techno-economic analysis of biomass fuelled combined heat and power for commercial buildings. Applied Energy, 112, 518-525. doi:10.1016/j.apenergy.2013.03.078
Huang, Y., Wang, Y. D., Rezvani, S., McIlveen-Wright, D. R., Anderson, M., Mondol, J., . . . Hewitt, N. J. (2013). A techno-economic assessment of biomass fuelled trigeneration system integrated with organic Rankine cycle. Applied Thermal Engineering, 53(2), 325-331. doi:10.1016/j.applthermaleng.2012.03.041
Jelen, F. C., & Black, J. H. (1983). Cost and optimization engineering (Second ed.): McGraw-Hill, Inc, 835 pp.
Kloppers, J. C. (2003). A critical evaluation and refinement of the performance prediction of wet-cooling towers. (PhD), University of Stellenbosch.
Lazzaretto, A., Toffolo, A., Manente, G., Rossi, N., & Paci, M. (2011, July 4-7, 2011). Cost evaluation of organic Rankine cycles for low temperature geothermal sources. Paper presented at the 24th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, Novi Sad, Serbia.
Lecompte, S., Lazova, M., van den Broek, M., & De Paepe, M. (2014). Multi-objective optimization of a low-temperature transcritical organic Rankine cycle for waste heat recovery. Paper presented at the 10th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics (HEFAT2014), Orlando, Florida.
COST ENGINEERING TECHNIQUES AND THEIR APPLICATION FOR ORC TECHNOLOGY
95
Lecompte, S., Lemmens, S., Huisseune, H., van den Broek, M., & De Paepe, M. (2015). Multi-Objective Thermo-Economic Optimization Strategy for ORCs Applied to Subcritical and Transcritical Cycles for Waste Heat Recovery. Energies, 8(4), 2714-2741. doi:10.3390/en8042714
Lee, K. M., Kuo, S. F., Chien, M. L., & Shih, Y. S. (1988). Parameters analysis on organic rankine energy recovery system. Energy Conversion and Management, 28(2), 129-136.
Leslie, N. P., Sweetser, R. S., Zimron, O., & Stovall, T. K. (2009). Recovered Energy Generation Using an Organic Rankine Cycle System. ASHRAE Transactions, 115(Part 1).
Moynihan, P. I. (1983). Application Guide for Waste Heat Recovery with Organic Rankine Cycle Equipment. Retrieved from Pasadena, California, USA:
Peters, M. S., Timmerhaus, K. D., & West, R. E. (2004). Plant Design and Economics for Chemical Engineers (Fifth ed.): McGraw-Hill
Quoilin, S., Declaye, S., Tchanche, B. F., & Lemort, V. (2011). Thermo-economic optimization of waste heat recovery Organic Rankine Cycles. Applied Thermal Engineering, 31(14-15), 2885-2893. doi:10.1016/j.applthermaleng.2011.05.014
Smith, R. (2005). Chemical Process Design and Integration. West Sussex, England: John Wiley & Sons, Ltd. Sweetser, R., & Leslie, N. (2007). Subcontractor Report: National Account Energy Alliance Final Report for
the Basin Electric Project at Northern Border Pipeline Company’s Compressor Station #7, North Dakota (ORNL/TM- 2007/158). Oak Ridge, Tennessee, USA, 45pp.
Toffolo, A., Lazzaretto, A., Manente, G., & Paci, M. (2014). A multi-criteria approach for the optimal selection of working fluid and design parameters in Organic Rankine Cycle systems. Applied Energy, 121, 219-232. doi:10.1016/j.apenergy.2014.01.089
Towler, G., & Sinnot, R. (2008). Chemical Egineering Design: Principles, Practice and Economics of Plant and Process Design: Butterworth-Heinemann, 1263 pp.
Turton, R., Bailie, R. C., Whiting, W. B., Shaeiwitz, J. A., & Bhattacharyya, D. (2013). Analysis, Synthesis, and Design of Chemical Processes (Fourth ed.). Upper Saddle River, New Jersey: Pearson Education International, 965 pp.
Walraven, D., Laenen, B., & D’haeseleer, W. (2015). Minimizing the levelized cost of electricity production from low-temperature geothermal heat sources with ORCs: Water or air cooled? Applied Energy, 142, 144-153. doi:10.1016/j.apenergy.2014.12.078
Whealy, R. W., Taylor, W., & George, E. (2012). City of Unalaska Powerhouse Exhaust Gas Waste Heat to Energy Project: Final Report (EPS Project No. 10-0159). Electric Power Systems, Inc.: 1-140.
Yang, F., Zhang, H., Song, S., Bei, C., Wang, H., & Wang, E. (2015). Thermoeconomic multi-objective optimization of an organic Rankine cycle for exhaust waste heat recovery of a diesel engine. Energy, 93, 2208-2228. doi:10.1016/j.energy.2015.10.117
Yang, M.-H., & Yeh, R.-H. (2015). Thermo-economic optimization of an organic Rankine cycle system for large marine diesel engine waste heat recovery. Energy, 82, 256-268. doi:10.1016/j.energy.2015.01.036
Yang, M.-H., & Yeh, R.-H. (2016). Economic performances optimization of an organic Rankine cycle system with lower global warming potential working fluids in geothermal application. Renewable Energy, 85, 1201-1213. doi:10.1016/j.renene.2015.07.067
Yu, G., Shu, G., Tian, H., Wei, H., & Liang, X. (2016). Multi-approach evaluations of a cascade-Organic Rankine Cycle (C-ORC) system driven by diesel engine waste heat: Part B-techno-economic evaluations. Energy Conversion and Management, 108, 596-608. doi:10.1016/j.enconman.2015.10.085
This chapter is based on the publication: Lemmens, S., & Lecompte, S., 2017, Case study of an organic
Rankine cycle applied for excess heat recovery: Technical, economic and policy matters, Energy
Conversion and Management 138: 670-685.
4. Case study of an organic Rankine cycle
applied for excess heat recovery
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
99
4.1. Introduction
Chapter 2 investigated the literature for insights in the economics of ORC systems. An important
finding from this chapter is the scarcity of information from practical ORC cases. The large
majority of the academic literature discusses the economics of ORC systems based on system
costs that have either been estimated or assumed. The most frequently used cost estimation
techniques have been examined in chapter 3. These methods are characterized by inherent
accuracy limits and should be interpreted as such. Hence, estimated ORC capital costs are
sufficient for preliminary budget approvals but are not detailed enough to give a decisive
answer regarding the value generated by an ORC project or the appeal of ORC technology
compared to other technologies. This fourth chapter contributes to the insight in the economics
of ORC technology by investigating the case of an ORC system installed in Flanders, Belgium,
used to recover industrial excess heat.
The starting point of this chapter is the observation that the published knowledge on the
economics of commissioned heat recovery ORC cases is limited to only a few publications.
Leslie, Sweetser, Zimron, and Stovall (2009) discussed the findings from a demonstration project
in the USA. The ORC system is used to recover thermal energy from a gas turbine driving a
natural gas pipeline compressor. The system was designed to operate continuously, except for
periodical maintenance pauses, and its performance was monitored for a period of one year.
The design power output amounts 5.5 MW, but the actual power output depended on, among
others, optimizing of controls and optimization and temporary shutdowns of the pipeline
compressor due to gas demand fluctuations. The capital costs of the complete ORC project
amounted circa 2500 $/kW and the annual operation and maintenance costs amount roughly
200 k$/y. The authors investigated the financial feasibility of the system by varying the
operation period and the cost of capital. The resulting NPV and IRR were beneficial, depending
on the scenario, but the financial analysis is not discussed more in detail (Leslie et al., 2009).
Tchanche, Declaye, Quoilin, Papadakis, and Lemort (2010) built a 2 kW ORC prototype for
laboratory experimentation. The system was built by converting an oil-free scroll compressor
into an expander. The costs of the system are sketched based on budget offers for the different
components and sum to 11.55 k€. The authors use this base cost to extrapolate the costs for
ORC systems with more capacity and investigate profitability of such systems for different
operational periods. The conclusion is that the 2 kW system is financially interesting in countries
with electricity prices above 15 c€/kWh. Assuming a 20 % subsidy, the proposed ORC system is
feasible with a capacity from 20 kW onwards (Tchanche et al., 2010). Finally, Tumen Ozdil and
Segmen (2016) investigate an ORC plant installed in Turkey to recover industrial excess heat.
The capital costs of the ORC module are provided in detail (500 k$ for 260.4 kWgr) and the
analysis focuses on the exergoeconomic performance of the system and the financial analysis is
based on a payback period assessment (Tumen Ozdil & Segmen, 2016).
CHAPTER FOUR
100
In other words, the few publications dealing with the economics of heat recovery ORCs provide
insight in the technical details and disclose their investment costs. And although each of them
performs a brief financial appraisal, the relevant input parameters, assumptions and results are
not discussed in detail. A better understanding of the parameters that influence the viability of
ORC projects is paramount to assess the technology’s economic potential. Therefore, this
chapter presents the study of a heat recovery ORC, with a brief description of the technical
setup but the primary focus on the economics. Besides detailed insight in the costs of a
commissioned system, the case study discusses the policy framework for ORC investments. The
application of ORC technology in industrial heat recovery applications is relatively innovative,
but its potential has been acknowledged at the level of public policy. The question at hand is the
extent to which these policies have an impact on the financial appeal, and thus the adoption, of
the technology. Moreover, the project assessment of the case study is based on the
assumptions that are relevant in these particular circumstances. Some conditions may change
over time or are specific to the firm’s situation. Therefore, this chapter investigates how the
project assessment is affected when the project parameters change. This gives insight in the
sensitivity of the results to changes in the project parameters, but also in the importance of
each of these factors.
This chapter is organized as follows. Section 4.2 draws the setup of the case study. The technical
setup of the ORC system is briefly outlined and the capital and relevant annual costs are
discussed in detail. Relevant policy goals and measures are identified, starting at the
overarching European level and subsequently descending to that of the Belgian federal and
Flemish regional policy makers (see also Lemmens (2015) and, for an elaborate clarification of
the setup of the policy measures, see Lemmens, Verbruggen, and Couder (2016)). Section 0
moves on to the financial appraisal of the project. After introducing the most common
assessment metrics and clarifying the relevant project parameters and assumptions, the results
of the financial assessment are presented. The influence of the public policy measures on the
results is assessed in section 4.4 and the sensitivity of the results for changes in the other
project parameters is investigated in section 4.5. Finally, section 4.6 discusses the findings from
this chapter and summarizes the most important implications.
4.2. An ORC excess heat recovery project in Belgium
This chapter discusses experience and insights gained from an operational ORC system installed
to recover excess heat from an industrial firm in the Flanders Region, Belgium. This section
discusses the technical setup and practicalities, gives insight in the project’s costs and provides a
brief outline of relevant public policies for excess heat electricity generation in Flanders.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
101
4.2.1. Technical setup
The ORC system is installed to recover flue gas from an industrial kiln. The flue gas has a
maximum temperature of 240-250 °C and should not be cooled beneath 160 °C. It amounts to
approximately 2.8 MW of thermal power. The ORC system is composed of three coupled units
with a total of 375 kWe gross installed capacity (292.4 kWe net) and operates on the working
fluid R245fa. Each unit is composed of plate heat exchangers, expander and generator and
constructed in a container. The system is coupled to the excess heat via an intermediate
thermal oil circuit and cooling occurs in an air cooled condenser.
The construction and on-site integration of the ORC system took about one year. The
integration of the flue gas heat exchanger in the main process initially caused a pressure drop
and a resulting decrease in primary production. After the production with the ORC started, it
was discovered that the original kiln fan was no longer sufficient to operate with both the
principal process and the ORC system. The fan was replaced and the system has now been
running for two years.
4.2.2. Investment and annual costs
An ORC employed for excess heat recovery has a rather straightforward economic picture. ORC
projects typically encounter large initial investments, but the annual costs are low since there
are limited operation and maintenance requirements and no fuel costs. The project discussed in
this chapter was initiated in 2013 in the Flanders Region, Belgium, and commissioned one year
later. The ORC system, consisting of three coupled ORC modules, and the integration
requirements were delivered and constructed not by the ORC manufacturer itself, but by a third
party. Figure 4.1 displays the structure of the total capital investment of the case study. The
ORC modules themselves, composed of a centrifugal pump, expander and generator, constitute
a bit more than half of the capital investment. About one fourth of the costs stem from the
evaporators and the air-cooled condensers. The intermediate thermal oil circuit is responsible
for 11 % of the costs and a similar share stems from the installation, including delivery,
construction and project management. The specific investment costs (SIC) for the complete
project amount 4217 €2013/kWgross. Considering only the price of the ORC modules, including
evaporators, expander, generator, pump and condensers, gives a SIC of 3280 €2013/kWgross. Note
that these investment costs are prices paid for the systems and not the costs to produce this
system. The price paid differs from the production costs because it is influenced by market
conditions of demand, supply and competition, and regulatory requirements and incentives
(Bejan, Tsatsaronis, & Moran, 1996; Verbruggen et al., 2010).
CHAPTER FOUR
102
Figure 4.1. Diagram of the case study’s total capital investment.
Legend: distribution of the investment costs of an ORC used for excess heat recovery from an industrial kiln, installed in 2013 in
Flanders, Belgium. The ORC unit itself (including the expander, the generator and the pump) is responsible for approximately half
of the capital costs.
In terms of annual costs the ORC system is rather advantageous. Fixed annual costs for an
industrial system include costs for safety necessities and health and environment requirements.
Both are zero for the ORC system: it constitutes no safety hazard due to low operating pressures
– in contrast with steam plants – and causes no direct risk to human or environmental health.
Variable costs are influenced by the rate of production and include fuels, raw materials, utilities,
waste treatment and disposal. This ORC system is operated by the excess heat from the main
industrial process and requires no additional inputs. Semi-variable costs including operation,
maintenance, supervision and replacement make up the total annual costs. An ORC system can
run independently after start-up. For an excess heat recovery ORC, integrated into an existing
plant, no specific additional operating and supervisory labour is required. It suffices for the plant
operator to check the ORC system during the routine plant tour. Maintenance for the system is
provided for in a maintenance contract with the supplier. The annual costs for the maintenance
contract amount 1.4 % of total investment costs. This is rather low compared to the commonly
expected 2-5 %, but the contract does not include replacement parts or equipment spares.
Insurance rates may vary according to the type of process that is being insured, the availability
of protection facilities (Peters, Timmerhaus, & West, 2004, p. 269), and on the company itself.
Annual insurance rates are usually estimated between 0.5 and 1.5 % of fixed capital investment
(Bejan et al., 1996, p. 369; Peters et al., 2004, p. 269). Since it is unknown how the ORC system
affects the plant’s insurance costs, these are not included in this analysis.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
103
4.2.3. Public policy
Because Belgium is a federal state and a member of the European Union (EU) there are three
relevant policy levels to take into account. The supreme level is the European, and in Belgium
the competences are distributed among the Federal State and the Regions. With the Europe
2020 strategy the European Commission aims for smart, sustainable and inclusive growth (EC,
2014). The objectives include a 20 % reduction of greenhouse gas emissions compared to 1990,
20 % of energy to be produced from renewables, and an increase of 20 % in energy efficiency by
2020 (EC, 2014). In 2014, the 20 % efficiency target has been extended and the goal for 2030
was set at 27 % (EUCO, 2014). These objectives are endorsed by, among others, the Energy
Efficiency Directive, the legislative framework on energy efficiency in the European Union (EU,
2012). The directive delineates the requirements for energy efficiency across the entire supply
chain, including the management of excess heat, and it requires all member states to set an
indicative national energy efficiency target to realize the common goal (EU, 2012).
The Belgian energy efficiency target for 2020 is set at an 18 % reduction of primary energy use
compared to projections (BE, 2014). In practice, the legislative competences in Belgium are
distributed among several administrative levels, namely the federal government and the
regions. The federal administration’s power concerning energy matters remains limited to
matters of national importance, such as nuclear energy, or with territorial interests, such as
offshore energy. The three regional governments, i.e. the Brussels Capital Region, Flanders and
Wallonia, are in charge of most energy matters, including the encouragement of energy
efficiency and renewable energy. In other words, the Belgian energy efficiency goal has to be
realized as a combined effort of the three regions. The availability of excess heat in Belgium is
not well documented, but it is acknowledged that the industry is the largest final energy
consumer. The total final energy use in Belgium amounted 34,047 Mtoe in 2014, of which 34.3
% (11,683 Mtoe) was accounted for by industry (Eurostat, 2016). The industrial energy use in
Flanders rises to 44.3 % of the 1458 PJ total, and an additional 19.9 % used by the energy sector
itself (MIRA, 2016). Recall that the federal government has no direct power to promote energy
efficiency or renewable energy, but it can indirectly do so through fiscal measures. For instance,
there exists a tax advantage for companies that invest in, among others, new energy equipment
for energy recuperation in industry. This ‘increased investment deduction’ allows to deduct part
of the investment from the company’s profits, so that less corporate income taxes have to be
paid. (FOD Financiën, 2014, Art. 68, §1, 1°; VEA, 2016)
The Flemish government can endorse its energy objectives directly. To address the industry’s
energy efficiency improvement potential the Flemish government issued two voluntary
agreements in 2002. The agreements were in force until the end of 2015 and respectively
addressed the more energy-intensive industry (benchmarking-covenant) (Vlaamse Regering,
2002) and the less energy-intensive industry (audit-covenant) (Vlaamse Regering, 2005). In 2015
CHAPTER FOUR
104
they have been succeeded by two – rather similar – new agreements, yet this time distinguished
according to whether the firm participates in the EU Emissions Trading Scheme (ETS) or not
(Vlaamse Regering, 2014a, 2014b). The covenants are voluntary in nature, but the advantages
are notable and participation is often required to attain other types of support. The firm is
required in return to perform an energy audit every four years and undertake profitable
measures that were identified with it. The cut-off point for profitability is set at an internal rate
of return (IRR) after taxes of minimum 14 % for ETS-companies and 12.5 % for non-ETS
companies (Vlaamse Regering, 2014a, 2014b). Measures with a lower IRR have to be
reinvestigated annually in case the IRR is above 10 %, or can be permanently discarded
otherwise (Vlaamse Regering, 2014a, 2014b).
In addition to these covenants, the Flemish government installed other, financially stimulating
arrangements. The most relevant measures are investigated for this study: the ecology
premium, tradable green certificates, combined heat and power (CHP) certificates and the
investment support for ‘green heat, waste heat and bio methane injection’. First of all, the
ecology premium (EP-PLUS) supports investment in particular technologies, identified as
ecologically beneficial. The premium reimburses part of the additional costs of the investment,
when compared to an investment in the standard technology without the beneficial ecological
effect. The subsidy percentage is a function of the company size and type of technology and
there is a bonus for firms with a valid first-line energy, environmental or eco-efficiency scan, a
valid environmental certification or a certified environmental management system.
Participation in the applicable voluntary agreement is required to be eligible for the EP-PLUS.
(Agentschap Ondernemen, 2014) Secondly, tradable green certificates (TGC) and CHP-
certificates support the exploitation, i.e. the production, of a system rather than the investment
in it. TGC stimulate electricity generated from renewable sources, whereas CHP-certificates
encourage the use of efficient cogeneration systems. For each MWh of renewable electricity
produced or, respectively, for each MWh heat-power savings achieved in comparison to
separate production, the operator obtains free certificates. The amount depends on the type of
technology applied. (VLAIO, 2016a, 2016b) Finally, the investment support for ‘green heat,
waste heat and bio methane injection’ is organized as a biannual call, and the recuperation of
excess heat is only supported for reuse in the form of heat (Vlaamse Overheid, 2016a).
Table 4.1 presents an overview of the policies discussed and whereas each of them has
relevance for heat recovery investments, they do not all apply for ORCs. The increased
investment deduction and the EP-PLUS unambiguously include ORC investments in their target
group. (Agentschap Ondernemen, 2016; VEA, 2016) TGCs can be requested in case (part of) the
energy source for the ORC system is renewable and CHP-certificates if the system meets
particular requirements. Both certificate schemes cannot be combined with the EP-PLUS.
(VLAIO, 2016a, 2016b; VREG, 2016) Because the investment support for green heat targets
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
105
investments that deliver output in the form of heat, it cannot be used to support an ORC
system. (Vlaamse Overheid, 2016b) For case study discussed in this chapter the financial
support from the ecology premium and the increased investment deduction are valid.
Table 4.1. Policies for heat recovery at the European level, from the Belgian federal government and the Flanders Region.
EU Belgium (federal) Flanders
Energy Efficiency Directive Europe 2020 strategy:
20 % increase in energy efficiency by 2020
Updated energy efficiency goal: 27 % by 2030
Increased investment deduction
Energy policy agreements Ecology premium (EP-PLUS) CHP-certificates Tradable green certificates Investment support for green
heat, waste heat and bio methane injection
Legend: Summary of policy goals and measures at the European level, from the Belgian federal government and from the
Flanders Region that are relevant for the recovery of excess heat. Not all policy measures are cumulative. The financial measures
relevant for the case study are the increased investment deduction and the ecology premium.
4.3. Financial appraisal of the excess heat recovery project
The ORC project is installed to recover and valorise the excess heat from an industrial kiln. Apart
from the technical requirements, the success of such projects is determined by the local
conditions and the firm’s own operational parameters. This section first discusses the most
important metrics for financial appraisal of investments, elucidates the assumptions for this
case study and discusses the results.
4.3.1. Metrics for financial appraisal
First of all, the net present value (NPV) reflects the value created by investing in the project. All
future cash flows are discounted back to time zero using a discount rate that reflects the
minimal required rate of return for the project. The NPV is calculated as:
NPV = ∑ − C0 +
ct
(1 + r)t
n
t=0
≥ 0
(4.1)
with C0 the capital investment; ct the net cash flows in year t; n the economic lifetime of the project; r the discount rate; t the point in time (year).
CHAPTER FOUR
106
The discount rate r is used to take the time value of money into account. It may also reflect the
risk from the investment. If the risk profile of the project is similar to that of the firm, the
weighted average cost of capital (WACC) is often used to calculate the discount rate. The project
creates value when the NPV is larger than zero. The NPV-method is the best for evaluating
project investments and for comparing mutually exclusive investments. (Jelen & Black, 1983)
Secondly, the internal rate of return (IRR) reflects the return of the project on the invested
amount of money. It is calculated as the interest rate that equals the present value of the
incoming and outgoing cash flows:
∑ − C0 +
ct
(1+IRR)t =0
n
t=0
(4.2)
with C0 the capital investment; ct the net cash flows in year t; IRR the internal rate of return; n the economic lifetime of the project; t the point in time (year).
In other words, the IRR is the rate of return required to set the NPV equal to zero. The project
creates value when the IRR is larger than an ex ante defined required rate of return. The IRR-
and NPV-method yield the same investment decision for a single project (in terms of accepting
or rejecting that particular project); or for multiple projects if those projects are independent
(i.e. they can be implemented simultaneously and one project will not affect the cash flow of
another) and they are not subject to capital rationing. For independent projects with capital
rationing or for mutually exclusive projects NPV and IRR may yield different investment
decisions, as a result of differences between the magnitude and timing of the cash flows. Due to
the re-investment assumption, IRR tends to prefer projects with similar-size investments that
have higher cash inflows in the early years. For the case of this ORC project, with one initial
negative cash flow and subsequent positive cash flows, the NPV and IRR method give the same
results.
Thirdly, the payback period (PP) indicates the period of time required to earn the initial
investment back. It is calculated based on the cumulative cash flows of the project and yields a
positive investment decision when it is smaller than an ex ante defined acceptable period.
However, the PP-method does not necessarily lead to a correct investment decision. This is
because it does not take account of the time value of money, nor of the cash flows that occur
after the period. The method favours “front-loaded” cash flows, i.e. an investment is preferred
the sooner its cash flows are received. It is still widely used disregarding its shortcomings.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
107
Lastly, the levelized cost of electricity (LCOE) is the break-even constant price (€/MWh) that
equalizes discounted expenses with discounted revenues for a project. It is popular when the
product of a project is considered to be homogeneous. It is used for comparing the expenses of
electricity generation projects. The LCOE is the ratio of the discounted expenses and the
discounted electricity production:
LCOE =
∑ C0 +ct
(1 + r)tnt=0
∑Et
(1 + r)tnt=0
(4.3)
with C0 the capital investment; ct the expenses in year t; Et the electricity production at time t; n the economic lifetime of the project; t the point in time (year).
It is intuitively bizarre to discount electricity production, but it makes sense when interpreted as
the earnings from this electricity, with lower values in the future than today (Kost et al., 2013).
The LCOE cannot be used to support investment decisions, but is useful to compare the
competitiveness of several electricity generation techniques. The LCOE should be interpreted
with caution, since it does not take account of the variability of electricity prices throughout the
day, month or year (electricity is not a truly homogeneous product in economic terms) and may
vary significantly according to the project’s location (Joskow, 2011).
4.3.2. Assumptions for the financial appraisal
Table 4.2 presents the parameters used in the financial assessment of the Flemish case study.
The attainable net power output from the ORC and the amount of load hours are fixed, as they
are determined by the technical characteristics of the ORC system and the plant’s primary
process. The capital investment, including the costs for the ORC module as well as its integration
costs, is set by the vendor of the system and the annual costs remain limited to the
maintenance of the system. There are no fuel costs because the system operates on excess
heat. The salvage value is assumed zero for this analysis, i.e. the costs for system
decommissioning are offset by its residual value. The electricity produced by the ORC system is
used in-house in the firm to offset part of the electricity bill, i.e. there are no electricity sales.
The change in price levels over time is reflected by inflation. However, the (long term) evolution
of prices is hard to predict and the Belgian inflation rates have been very low since 2013.
Consequently, the average Belgian inflation rate since the start of the millennium, calculated at
2 %, is applied for both electricity and general price trends. When inflation is explicitly taken
CHAPTER FOUR
108
into account, the real (constant) instead of the nominal (current) discount rate should be
applied.5
The appropriate value of the discount rate is subject of debate. Cost-benefit analyses are
performed to assess whether a project creates value, but the definition of value differs. Private
investors seek to maximize profits and invest in those projects where the financial gains are the
highest. The funds are invested in those projects with the highest returns and the required rate
of return is typically higher. Social cost-benefit analyses concern valuations that go beyond the
measurable money transactions, regardless whether performed for public investments or for
private investments with public interest. In this context, the creation of social value is prior to
financial gain. Therefore, social projects are argued to require lower discount rates than private
ones. The choice of the discount rate strongly influences the results of the project analysis. The
discount rate reflects the rate at which the value of money changes with time. Money in the
future is valued less than money today. Discount rates are used to take account of this
phenomenon and (exponentially) decrease future values to present values: higher discount
rates cause a faster decline. This implies that high discount rates reflect a lower valuation of
future cash flows, i.e. more ‘impatience’, and lower rates can be used if the future is valued
more. There is no scientific consensus regarding the correct value of the discount rate. For
investment projects with a long lead-time, such as investments in energy projects, the initial
costs tend to be very high and the benefits are acquired over a longer period of time. In this
case, applying high discount rates can undervalue the future benefits of e.g., energy efficiency
or renewable energy investments.
This analysis utilizes three different discount rates to represent different scenarios and
demonstrate the impact of this choice. The high scenario utilizes a real discount rate of 12 %,
representative for (private) investors with sufficient resources to spend on projects of their
choice. The low rate of 3 % is used to reflect a more social approach, where the long term
benefits of energy efficiency investment are valued to a greater extent. The intermediate
scenario is represented by a real discount rate of 6 %.
5 The real and the nominal rate are related as follows: (1 + 𝑛𝑜𝑚𝑖𝑛𝑎𝑙 𝑟𝑎𝑡𝑒) = (1 + 𝑟𝑒𝑎𝑙 𝑟𝑎𝑡𝑒) ∗ (1 + 𝑖𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛).
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
109
Table 4.2. Project parameters for assessment of the excess heat recovery ORC project.
Parameter Value Parameter Value
ORC net power 292.41 kWe Operating years 20 y Capital investment (𝐶0) 4,217 €2013/kWgross Integration costs 11 % of 𝐶0 Annual O&M contract (1.40 % of 𝐶0) €/ y Annual fuel costs 0 €/y Insurance costs / Salvage value 0 € Sales electricity prod. 0 % Purchase price electr. 90 €/MWh Sales price electr. / Load hours 5500 h/y Electricity price inflation 2 %/y General inflation 2 %/y Discount rate (real) High: 12 %
Medium: 6 % Low: 3 %
Legend: Summary of the relevant project parameters for the financial appraisal of the heat recovery case study in Flanders,
Belgium. Some of the parameters are project-specific, others have been assumed representatively.
Finally, Table 4.3 displays the public policy parameters relevant for this project analysis. The
standard corporate income tax rate in Belgium amounts 33.99 %. Depreciation refers to the
accounting practice of reallocating an asset’s costs over its lifespan, and the period depends on
the firm’s specific practices and tax situation. This analysis assumes linear depreciation over a
period of 5 years, analogous to the profitability calculations suggested in the Flemish energy
policy agreements. The increased investment deduction implies a one-time tax reduction of
14.5 % ∗ 33.99 % ∗ 𝐶0. A firm can apply for an ecology premium under the condition of
participation in the appropriate energy policy agreement, for a maximum of € 1 million per
three years. The actual percentage depends on the technology, the firm size, and whether or
not the firm has a particular environmental management system. The firm discussed in this
chapter is a large firm involved in the EU ETS, participates in the appropriate energy policy
agreement and has a valid ISO certification. This corresponds to a subsidy of 35.75 % of the
essential components of the ORC system.
Table 4.3. Government interventions of the excess heat recovery ORC project.
Parameter Value Parameter Value
Corporate income taxes 33.99 % Depreciation linear over 5 years Ecology premium 35.75 % of essential
component costs (excl. VAT)
Increased investment deduction
14.5 % for investments in 2013, tax year 2014
Legend: Summary of the policy measures that influence the financial appraisal of the heat recovery case study.
CHAPTER FOUR
110
4.3.3. Results of the financial appraisal6
The financial feasibility of the heat recovery ORC project is assessed based on the four criteria
NPV, IRR, PP and LCOE. Two versions of the LCOE are calculated: the simple LCOE and a policy-
inclusive one. The simple LCOE is conventionally calculated without any form of policy
interference, thus no taxes or subsidies are accounted for. The policy-inclusive LCOE includes
the effects of corporate income taxation and policy incentives. The ecology premium entails
investment support (CS) and is deducted from the investment costs, whereas the investment
deduction 𝐼𝑡 implies a reduction on the tax liability. It is accredited as a reduction of the total life
cycle costs and calculated as a fraction of the investment costs 𝐶0:
LCOEpolicy
=(C0 − CS) + ∑
O&Mt + CF,t
(1 + r)t (1 − 𝑇) − ∑𝐶0 ∙ 𝑇 ∙ 𝐼𝑡
(1 + 𝑟)𝑡𝑛𝑡=0 − ∑
𝐷𝑡
(1 + r)t ∙ 𝑇 −S
(1 + r)tnt=0
nt=0
∑Et
(1 + r)tnt=0
(4.4)
with C0 the capital investment; CF,t the fuel costs in year t; CS the investment support; 𝐷𝑡 the depreciation in year t; 𝐼𝑡 the investment deduction in year t; Et the electricity production at time t; n the economic lifetime of the project; O&Mt the operation and maintenance costs in year t; r the discount rate; S the salvage value at the end of the project lifetime; 𝑇 the applicable tax rate; t the point in time (year).
The cumulative cash flow trend of the ORC project is displayed in Figure 4.2. Note that this trend
follows the same course for all three discount rate scenarios (high, medium and low), because it
displays the cumulative non-discounted cash flows in each project year. Figure 4.3 shows the
6 The ecology premium is paid out in three stages: the first 30 % after the start of the investment, a second 30 %
after 60 % of the investment is completed and the final 40 % after full commissioning of the investment. Because the commissioning had a duration of one year, the analysis assumed that the ecology premium is retrieved completely in the first project year. Alternatively, assuming that the investment is commissioned in the year following the initiation of the project, and thus not in the same year, implies that 60 % of the premium is retrieved in the first year and 40 % in the second. This would change the project results only slightly. The NPV decreases to 892 k€, 493 k€ and 11 k€ for the low, medium and high discount rate scenario respectively. The IRR decreases from 12.6 % to 12.2 % and the PP remains at 6 years. The simple LCOE does not take account of the ecology premium and does not change. Finally, the policy LCOE increases, to 43 €/MWh, 54 €/MWh and 83 €/MWh for the low, medium and high discount rate scenario respectively.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
111
performance of the project for the three different discount rates; the numerical results are
presented in Appendix B. The NPV is the best criterion to evaluate project investments. It is
positive in all three cases, which means investing in the ORC yields a positive return for the
investor. The significant difference among the three cases demonstrates the importance of
establishing the correct discount rate. Both the IRR and the PP do not take account of the time
value of money, consequently they have the same value for all three scenarios. The IRR of 12.6
% implies that the investment has a return of 12.6 %. The link between the IRR and the NPV is
clearly demonstrated: the high case has an IRR of 12.6 %, which is slightly higher than the
discount rate of 12 % and hence yields a slightly positive NPV. A firm that requires an
investment return larger than 12.6 % finds a negative NPV and evaluates the project negatively.
The PP is estimated at around 6 years, but recall the shortcoming of this criterion in
incorporating necessary decision parameters. It is calculated in this analysis solely for
demonstration purposes. The simple LCOE does not take account of any government policies,
and ranges between 83 €/MWh for the low discount rate scenario and 148 €/MWh for the high
scenario. The higher the discount rate, the faster the project has to break even and thus the
higher the required price for the generated electricity. The policy LCOE includes the effect of
both taxes and subsidies and ranges between 35 and 66 €/MWh depending on the discount rate
scenario. Note that, when taxation is included in the LCOE analysis, the LCOE no longer simply
reflects the price to be obtained to break-even. The required revenues before taxation will be
higher in order to cover all costs and have the same return on investment (Short, Packey, &
Holt, 1995).
Figure 4.2. Cumulative non-discounted cash flow of the ORC project for all discount rate scenarios.
Legend: Cumulative non-discounted cash flows for the heat recovery case study in Flanders, Belgium. The course of the graph is
the same for all three discount rate scenarios since the figure concerns the non-discounted cash flows.
CHAPTER FOUR
112
Figure 4.3. NPV, IRR, PP, LCOE and policy LCOE of the ORC project for high, medium and low discount rates.
Legend: Project appraisal for the heat recovery case study, assessed by five criteria: NPV, IRR, PP, LCOE and the policy-inclusive
LCOE. The results are compared for the three discount rate scenarios.
4.4. The impact of public policy
The results of the project analysis in section 4.3.3 are based on the project parameters
presented in Table 4.2 and Table 4.3. The parameters were defined based on the actual firm’s
circumstances. This section takes a closer look at the impact of the different policy
interventions. Section 4.4.1 investigates the change in the results when one or more of the
policy measures are excluded. Section 4.4.2 looks at the alternative case where the firm
optimizes its tax liabilities.
4.4.1. The impact of public policy: changing the extent of government intervention
The ORC was purchased in 2013 in Flanders, Belgium. The investment was supported with a
subsidy from the Flemish government and a tax benefit from the federal government. The goal
of this section is to determine how the assessment of the case study changes when one or more
of these policy interventions are eliminated. Figure 4.4 shows the cumulative cash flow for the
case study, for several levels of policy intervention. Figure 4.5 displays the NPV for these
different policy scenarios, calculated for the three different discount rates. The numerical
results are displayed in Appendix B. The ‘No Policy’ scenario investigates the case without any
form of government intervention: no subsidies, but also no taxes are accounted for. Obviously,
a situation without corporate income taxation is not realistic in a for-profit setting, but it serves
the purpose of comparison. This scenario yields an IRR of 6.7 %, implying a positive assessment
when the required return amounts 3 % or 6 %, but a negative evaluation for the 12 % discount
rate scenario. Considering the ‘Only Taxes’ case in the Belgian context, a 33.99 % taxation is
levied on corporate income. With a linear depreciation schedule over a period of five years, the
project is financially interesting for the low discount rate scenario (r = 3 %). For the scenarios
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
113
with medium and high required return on investment, the project is no longer considered
interesting with NPV values below zero (Figure 4.5). Note that the inclusion of taxation leads to
less negative cumulative cash flows in the first years of the project compared the situation
without taxes, i.e. the slope of the curve is steeper for the first five project years (Figure 4.4).
This is due to the alleviating impact of depreciation, allowing companies to write the investment
costs of an asset off over several accounting years. In terms of accounting, this reduces the
earnings before taxes, thereby lowering the amount of taxes to pay and increasing the project’s
cash flow. After this period the slope of the curve is flatter than in the case without policy,
yielding an overall lower NPV for the case with taxes. This comparison illustrates why the PP is
not a suitable project appraisal metric. Looking at solely the break-even point bears the risk of
wrong project selection because cash flows that occur afterwards are ignored. The increased
investment deduction alleviates the tax burdens slightly, but is not sufficient to yield a positive
investment decision for the medium and high discount rates (cfr. the ‘Tax & Deduction’
scenario). The ecology premium subsidizes the ORC investment costs and has a truly beneficial
impact on the project assessment (‘Tax & Premium’ scenario). The premium offsets the tax
burdens and increases the NPV to levels higher than the ‘No Policy’ benchmark, for all discount
rate scenarios. The combined effect of taxation, the ecology premium and the increased
investment deduction sketches the real picture of the case study (section 4.3.3).
Figure 4.4. Cumulative cash flow for five different policy scenarios.
Legend: Cumulative non-discounted cash flows for the heat recovery case study, for five different policy scenarios. The course of
the graph is the same for all three discount rate scenarios since the non-discounted cash flows are considered.
CHAPTER FOUR
114
Figure 4.5. Comparison of the NPV for three different discount rates and five different policy scenarios.
Legend: Comparison of the NPV for the heat recovery case study in Flanders, Belgium, for five different policy scenarios and three
discount rate scenarios (r = 3 %; r = 6 % and r = 12 %). There is a strong influence of both policy measures and discount rates on
the results.
4.4.2. The impact of public policy: changing the impact of corporate income taxation
The analysis in section 4.4.1 considers the impact of the public policy measures, assuming that
the income generated by the ORC investment is entirely subject to corporate income taxation.
However, most companies will attempt to optimize their taxation duties as much as possible, for
instance by lowering their income via additional costs. Assuming that the company does
optimize its fiscal situation, the corporate income taxation is reduced to zero while the benefit
of the increased investment deduction is retained. The results of this situation are presented in
Figure 4.6 and Figure 4.7. There are four different policy scenarios, according to which of the
potential subsidies is included, but in each of them the corporate income taxation has no
impact. The ‘No Subsidy’ scenario is the same as the ‘No Policy’ scenario in section 4.4.1: there
are no subsidies involved and the taxes amount zero. Comparing the scenarios to their
equivalent scenario that does incorporate the full effect of taxation (section 4.4.1), the NPV for
the scenarios without taxation are higher. For the real case, i.e. including both the ecology
premium and the increased investment deduction, the NPV and the IRR increase from 504 k€
and 12.6 % to 674 k€ and 12.7 % respectively, for the intermediate discount rate scenario (r = 6
%).
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
115
Figure 4.6. Cumulative cash flow for four different policy scenarios.
Legend: Cumulative non-discounted cash flows for the heat recovery case study for four different policy scenarios, assuming that
the firm optimizes its tax duties. The course of the graph is the same for all three discount rate scenarios since the non-
discounted cash flows are considered.
Figure 4.7. Comparison of the NPV for three different discount rates and four different policy scenarios.
Legend: Comparison of the NPV for the heat recovery case study, for five different policy scenarios and three discount rate
scenarios (r = 3 %; r = 6 % and r = 12 %), assuming that the firm optimizes its tax duties.
4.5. Parameter sensitivity analysis
The public policy measures have proven to be influential for the financial assessment of the case
study. The question remains which of the other input parameters have the most important
impact on the case study’s evaluation. This section investigates the impact of the different input
parameters on the results. Two methods are applied: a ceteris paribus setup and a Monte Carlo
simulation. In the former the impact of changing the parameter is assessed while all others
remain constant (section 4.5.1). The latter considers changes in each of the input parameters
simultaneously (section 4.5.2).
CHAPTER FOUR
116
4.5.1. Sensitivity of the results: ceteris paribus analysis
The ceteris paribus sensitivity analysis investigates the sensitivity of the financial appraisal
results for changes in the input parameters. The sensitivity of a parameter is assessed by
modifying only that parameter and retaining all others constant. Some input factors are fixed,
such as the maximum net power output, others depend on external circumstances but can
change, such as electricity prices, and others depend on the firm’s own circumstances, such as
the annual load hours. This section assesses the impact of changes in eight key input
parameters: the ORC component costs, the annual O&M costs and its inflation rate, the annual
load hours, the electricity purchase price for the firm and its inflation rate, the lifetime of the
ORC system and the discount rate used for the financial appraisal.
Sensitivity analysis: ORC net power and capital investment
The net power output depends on technical characteristics and is considered fixed for this
analysis. The investment costs are externally determined, namely by the vendor, but could
change due to e.g., system innovations or series production of ORC modules, changing raw
material prices, or improved installation experience. Figure 4.8 displays the change in each of
the five assessment metrics when the ORC module costs vary with ± 10 %, holding the
integration and the O&M costs constant. Keeping all other parameters unchanged, a decrease in
the investment costs entails an improvement in the financial assessment. For the intermediate
case (r=6 %), a 10 % reduction in investment costs leads to an increase of 8.6 % in the NPV and 8
% in the IRR.
Figure 4.8. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the ORC module costs for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the ORC module costs, and compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %). Increasing the
investment costs has a negative impact on the financial feasibility of the project.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
117
Sensitivity analysis: annual O&M costs
A similar analysis is performed for the annual O&M costs. These amount about 1.4 % of the
capital investment for the base case. If the O&M costs increase by 10 %, the NPV and IRR will
grow by 3.8 % and 1.8 % respectively. The O&M costs thus represent only a minor factor in the
ORC’s costs and have little influence on the financial assessment, as shown in Figure 4.9.
Figure 4.9. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the annual O&M costs for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the annual O&M costs, and compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %). Increasing the annual
O&M costs has a slightly negative impact on the financial feasibility of the project.
Sensitivity analysis: load hours
The annual load hours are not free to choose. A heat recovery ORC system is coupled to another
process and depends on its exhaust. Nevertheless, the achievable load hours significantly
impact the financial feasibility of the project. Figure 4.10 displays the sensitivity of the five
assessment metrics to changes in the available load hours. All other parameters constant, both
the NPV and the IRR change substantially with varying load hours: more load hours implies
higher profitability. The lower the discount rate, the stronger the change in NPV. The low
scenario (r = 3 %) requires about 2640 operating hours per year for a positive profitability
assessment. A 10 % increase in load hours implies a 19.24 % growth in the NPV. When a higher
return of 12 % is requested, the system should operate for at least 5280 h/y. When the required
return amounts 6 % (the intermediate scenario), increasing the load hours by 10 % causes the
NPV to grow with nearly 26 %.
Sensitivity analysis: operating years
The attainable operating years of an ORC system are commonly assumed at 20 years. Figure
4.11 displays the changing of the results when the operating years are varied from 0 to 30. Since
the depreciation period cannot exceed the number of operating years, the depreciation period
CHAPTER FOUR
118
has been set equal to the number operating years for operating years less than 5; and equal to 5
for all other cases. The threshold is again strongly influenced by the discount rate. The ORC
system has to operate for nearly 8 years to be interesting in case the desired return amounts 3
%, but this figure mounts to about 18 years for the high scenario (r = 12 %). The NPV augments
by 14 % in case the number of operating years increases with 10 % (intermediate 6 % discount
rate).
Figure 4.10. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the load hours for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the annual load hours. The results are compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %). Increasing
the annual load hours has a strongly positive impact on the financial feasibility of the project.
Figure 4.11. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the operating hours for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the operating years. The results are compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %). Increasing the
operating years has a strongly positive impact on the financial feasibility of the project.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
119
Sensitivity analysis: electricity prices
The electricity price paid by a firm depends on numerous aspects. Firm-specific factors include
electricity demand or negotiating power. External circumstances entail inflation levels and
country regulations, but lie beyond the firm’s scope of control. With all other parameters as
assumed, the project is profitable when the electricity price is higher than about 60 €/MWh, for
the average scenario with a 6 % discount rate (Figure 4.12).
Figure 4.12. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the electricity price for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the electricity price. The results are compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %). Increasing the
electricity price has a strongly positive impact on the financial feasibility of the project.
Sensitivity analysis: inflation
The impact of varying inflation rates is shown in Figure 4.13 and Figure 4.14. Figure 4.13 shows
the results for changes in the general inflation rate, which affects the annual O&M costs. The
inflation rate is varied between zero and four percent, but the effect on the results of the
financial appraisal is minor. Increasing inflation has a negative influence on the financial
feasibility, but a 10 % inflation increase causes an NPV decrease of only 0.6 % (r = 6 %). The
effect of electricity price inflation is somewhat larger. An increase in the electricity price entails
an increasing value of the ORC production and thus an improved financial assessment. The
effect is slightly larger than for the O&M inflation, but still relatively low with an NPV
improvement of 4.8 % for an electricity price increase of 10 % (r = 6 %).
CHAPTER FOUR
120
Figure 4.13. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the general inflation rate for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the general inflation rate. The results are compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %).
Increasing the general inflation rate has a slightly negative impact on the financial feasibility of the project.
Figure 4.14. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the electricity price inflation rate for different discount rates.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed for changes in
the electricity price inflation rate. The results are compared for the three discount rate scenarios (r = 3 %; r = 6 % and r = 12 %).
Increasing the electricity price inflation rate has a positive impact on the financial feasibility of the project.
Sensitivity analysis: discount rate
Finally, the effect of changing the required return is clear from the previous analyses, but a
more extensive analysis of the impact is executed in this section. Figure 4.15 shows the course
of the five assessment metrics for a real discount rate changing between zero and 20. The IRR
and the PP are not affected by changes in the discount rate, but the NPV and LCOE change
dramatically. The NPV declines from about 1.500 k€ to nearly -300 k€ when the discount rate
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
121
increases from zero to 20 %. Similarly, the LCOE and policy-induced LCOE increase from about
60 €/MWh to roughly 230 €/MWh.
Figure 4.15. Sensitivity of the NPV, IRR, PP, LCOE and policy LCOE to changes in the real discount rate.
Legend: Sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium, assessed by five
criteria: NPV, IRR, PP, LCOE and the policy-inclusive LCOE. The sensitivity of the results is assessed for changes in the discount
rate. Increasing the discount rate has a strongly negative impact on the financial feasibility of the project.
Sensitivity analysis: summary
The sensitivity analysis shows the influence of each of the project parameters on the results. A
summary of the analysis is given in Figure 4.16 and Figure 4.17. Figure 4.16 shows the change in
NPV for varying the input parameters with ±20 %; Figure 4.17 shows the same analysis for the
IRR. The figures confirm the finding of the previous section that the results are most impacted
by changes in the electricity price and the annual load hours. In fact, both factors change the
results equally. Increasing either the load hours or the electricity price paid by the firm with 10
%, implies a 26 % increase in the NPV (r = 6 %). The second largest effect is obtained by changing
the total operating years of the ORC system: a 10 % extension of the project lifetime increases
the NPV by 14 %. The third largest influence stems from a change in the project’s investment
costs. However, a 10 % decrease in investment costs would change the NPV by 8.6 % so the
influence is proportionally less important. Altering all other parameters results in an NPV that
changes even less (<5 %).
CHAPTER FOUR
122
Figure 4.16. Sensitivity of the NPV to changes in various project parameters, for a real discount rate of 6 %.
Legend: Summary of the sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium,
assessed by the NPV criterion. The results are most sensitive for changes in the electricity price and the annual load hour,
followed by the operating years.
Figure 4.17. Sensitivity of the NPV to changes in various project parameters, for a real discount rate of 6 %.
Legend: Summary of the sensitivity analysis of the project appraisal for the heat recovery case study in Flanders, Belgium,
assessed by the IRR criterion. The results are most sensitive for changes in the electricity price and the annual load hour, followed
by the operating years.
4.5.2. Sensitivity of the results: Monte Carlo simulation
The ceteris paribus sensitivity analysis demonstrated the major importance of the electricity
price and the annual load hours. However, changing only one parameter at a time shows only
part of the picture. This section expands the analysis with a Monte Carlo simulation (performed
with the Crystal Ball software). For each of the input parameters, a representative probability
distribution is defined (see Table 4.4) and the financial appraisal is performed 10,000 times,
selecting the parameter values randomly from their distribution functions. Instead of one single
value, this Monte Carlo simulation thus gives a probability distribution of the results, giving
-
200,00
400,00
600,00
800,00
1.000,00
-30% -20% -10% 0% 10% 20% 30%
NP
V [
k€]
Percentage change
Investment costs change
Annual O&M costs change
Annual load hours change
Operating years change
Electricity prices change
O&M inflation change
Electricity price inflation change
0,00%
5,00%
10,00%
15,00%
20,00%
-30% -20% -10% 0% 10% 20% 30%
IRR
[%
]
Percentage change
Investment costs change
Annual O&M costs change
Annual load hours change
Operating years change
Electricity prices change
O&M inflation change
Electricity price inflation change
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
123
insight in the probability of a positive outcome. At the same time, the simulation incorporates
not only the uncertainty of each of the input parameters, but also their importance.
The distributions of the ORC capital investment and the annual O&M costs are determined
based on a collected database with ORC costs information. The minimum and maximum values
are based on those of ORC systems with a capacity between 375 ± 75 kW. Because there is not
sufficient data to fit the distribution, both the capital and the annual costs are assumed to be
uniformly distributed. The distribution of the inflation rates is fitted for the average Belgian
rates in the period 2000-2015 and that of the electricity purchase price is fitted based on prices
for Belgian industrial consumers (excluding VAT and other recoverable taxes and levies) for the
same period. Both the operating years and the annual load hours have a triangular distribution.
Table 4.4. Probability distribution assumptions for the input parameters for the Monte Carlo simulation.
Parameter Distribution Distribution characteristics
ORC net power / (constant) / ORC component costs [k€] Uniform Min: 563 Max: 1403 Integration costs / (constant) / Annual O&M contract [€/y] Uniform Min: 20,000 Max: 101,000 General inflation [%] Logistic Mean: 1.98 Scale: 0.65 Operating years [y] Triangular Min: 10 Likeliest: 20 Max: 30 Load hours [h/y] Triangular Min: 0 Likeliest: 7008 Max: 8760 Purchase price electr. [€/MWh]
Weibull Location: 23.14 Scale: 78.01 Shape: 5.46
Electricity price inflation [%] Logistic Mean: 1.98 Scale: 0.65 Discount rate (real) [%] 6 (constant)
The discount rate is assumed constant at 6 % for this analysis, to determine the influence of only
the other, non-financial parameters, on the results. If the discount rate were included in the
Monte Carlo simulation, assuming a uniform distribution between 0 and 20 %, it would be
responsible for 26.2 % of the variation in NPV (with a negative influence for increasing rates).
Keeping the discount rate constant at the intermediate rate of 6 %, the results of the Monte
Carlo simulation are displayed in Table 4.5 and
Figure 4.18 for the NPV. The largest contribution to the NPV stems from the annual load hours
and to a lesser extent from the electricity price and the annual O&M costs. There is a 69 %
probability for a positive NPV. The parameter influences are somewhat different than in the
ceteris paribus sensitivity analysis. This is because for the latter the changes are assessed for
deviations of only ± 20 % compared to the base case and because the parameter influences are
not considered simultaneously. The Monte Carlo simulation takes account of more realistic
scenarios because the parameters are evaluated based on a representative probability
distribution and for each of them simultaneously. The most remarkable differences are the
CHAPTER FOUR
124
importance of the annual O&M costs, the component costs and the electricity price. The Monte
Carlo simulation discloses a lesser importance of the electricity price and the investment costs
than in the ceteris paribus sensitivity analysis. The annual O&M costs appear to be of larger
importance. This confirms the finding that the annual O&M costs of the case study are relatively
low compared to expectations.
Table 4.5. Parameter contributions to the variance in NPV, established by the Monte Carlo simulation.
Parameter Contribution to NPV variance
ORC component costs 1.4 % (-) Annual O&M contract 10.6 % (-) General inflation 0.9 % (-) Operating years 3.3 % (+) Load hours 66.6 % (+) Purchase price electricity 12.7 % (+) Electricity price inflation 5.4 % (+)
Figure 4.18. Frequency distribution of the NPV, established by the Monte Carlo simulation.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
125
4.6. Discussion of the results
This chapter discussed the case study of an operational ORC system, primarily from an economic
point of view. This section recapitulates and discusses the most important results from this
chapter. Firstly, the project was evaluated favourably. The NPV is positive for all three
considered discount rate scenarios, which means the project creates value for the investor.
Obviously, this assessment depends strongly on the discount rate. The NPV is positive because
the IRR of the project, calculated at 12.6 %, exceeds the ex-ante defined hurdle discount rates
(r = 3 %; 6 % and 12 %). Recall the covenants from the Flemish government. When
participating in the agreement, the government requires the firm to perform regular energy
audits and undertake those measures that were identified and are evaluated as profitable. This
profitability cut-off point was set at an IRR after taxes of minimum 14 % for companies
participating in the EU ETS and at 12.5 % for non-ETS companies (Vlaamse Regering, 2014a,
2014b). When the measure has an IRR that is lower than this cut-off point but higher than 10 %,
it has to be reinvestigated annually. Measures with an IRR lower than 10 % is can be discarded
permanently (Vlaamse Regering, 2014a, 2014b). This implies that, according to the
requirements defined by the Flemish government, this investment would not have been
worthwhile. This assessment can change as one or more of the project’s parameters are
modified (cfr. the sensitivity analysis). If the goal of the government were to support the
investment to achieve the 14 % IRR threshold, the ecology premium would have to increase
even further, from 35.75 % to 40.7 % of the component costs. Alternatively, the IRR increases to
14 % if the ORC component costs would decrease from 3742 €/kW to 3234 €/kW.
A project appraisal takes account of project specific parameters but is also the result of the
assumptions made for the analysis. For instance, the inflation rate used in this analysis was
assumed at 2 % based on past inflation rates, but it remains a precarious exercise to extrapolate
this rate over 20 future years. An alternative approach therefore uses nominal instead of real
cash flows. The project assessment carried out in this chapter was evaluated along three
discount rate scenarios, where the low scenario was established to represent investments with
a more social approach and the highest discount represents firms with sufficient investment
choices. The discount rate represents the required return of the investment but also its risk,
which is the combination of a systematic risk, representing the risk in the market, and the
specific risk, which involves risk factors specific to the investment itself. Clearly, the specific risk
is higher for new technologies than for well-established ones, which implies a higher required
return for newer technologies. This means that instead of the approach set out in section 4.3.2,
an alternative is to work with nominal discount rates representing the technology-specific risk.
The discount rate scenarios could include a first one representing an investment in a known
technology in a social context (r = 3 %), a second scenario with investment in a known
technology in a regular context (r = 6 %), thirdly and investment in a new technology but with
CHAPTER FOUR
126
potential to replicate the investment within the firm (r = 9 %), and finally a scenario where the
firm invests in a new technology but there is no potential to repeat the investment within the
firm (r = 12 %). Retaking the financial assessment under these conditions leads to the results
presented in Figure 4.19. The results are slightly worse than those of the first approach (section
4.3.3). This can mainly be attributed to the elimination of inflation for the electricity price. Note
that this project concerned an investment in a relatively new technology, which will probably
not be replicated within the firm. This means a higher-risk investment and hence a discount rate
in the range of the highest scenario. With a 10.8 % IRR, the NPV turns out negative when the 12
% nominal discount rate scenario is considered representative.
Figure 4.19. NPV, IRR, PP, LCOE and policy LCOE of the ORC project for different nominal discount rates.
Legend: Project appraisal for the heat recovery case study, assessed by five criteria: NPV, IRR, PP, LCOE and the policy-inclusive
LCOE. The results are compared for four nominal discount rate scenarios.
Secondly, the LCOE allows comparison of the project’s economics to other energy generation
technologies, rather than considering the financial feasibility of the project itself. The LCOE
places the project’s expenses in perspective to its electricity generation and is often interpreted
as the minimum cost at which the generated electricity must be sold for the project to break
even over the lifetime of the project. The real LCOE of the ORC case study was calculated
between 83 €/MWh and 148 €/MWh, for the different discount rate scenarios, and between 37
€/MWh and 66 €/MWh for the same scenarios when policy intervention is included. The
nominal LCOE lies between 80 €/MWh and 145 €/MWh when excluding policy intervention and
between 33 €/MWh and 64 €/MWh when including it, for a nominal discount rate ranging
between 3 % and 12 %. Thus, both the real and nominal LCOE at 6 % are lower than the 2013
Belgian industrial electricity price of 110 €/MWh. With a higher hurdle rate (r = 12 %), both the
nominal and the real LCOE are too high compared to the average industrial end-user price.
Alternatively, the LCOE is often used to compare the costs of generating electricity among
different technologies. The International Renewable Energy Agency published insights in the
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
127
LCOE of renewable power generation technologies for the year 2014 (IRENA, 2015). The fossil
fuel LCOE benchmark is established between 45 and 140 USD/MWh, using a WACC of 10 %
(real). With an average 2014 USD-EUR exchange rate of 0.7539, this means a fossil fuel
benchmark in a range from 34 to 106 EUR/MWh. At a 10 % real discount rate, the LCOE of the
case study amounts 129 €/MWh (or 171 $/MWh) when no financial support is included and 56
€/MWh (or 74 $/MWh) when the ecology premium and the increased investment deduction are
accounted for. This means that electricity production with the ORC system is competitive only in
the case subsidies are provided. Figure 4.20 displays an overview of the LCOEs defined in the
study by IRENA (2015). Compared to other (renewable) power generation technologies, the ORC
case performs moderately. Without subsidies the ORC system’s LCOE lies above the high end of
the fossil fuel benchmark, but not higher than those of renewable energy technologies such as
solar photovoltaic, CSP or offshore wind. Including subsidies the ORC system is price-
competitive with technologies such as biomass, geothermal, hydro and onshore wind, but it is
not sufficiently low cost to outcompete the fossil alternative.
Figure 4.20. LCOE for utility-scale renewable power technologies.
Legend: The LCOE of multiple renewable power generation technologies is compared to the fossil fuel benchmark range of 45 to
140 USD/MWh (or 34 to 106 EUR/MWh)). Source: (IRENA, 2015).
CHAPTER FOUR
128
Thirdly, there is a clear influence of the policy measures on the financial assessment of the
project. In a situation without subsidies the project would have an IRR of only 5.4 %, or 6.7 % in
case the firm optimizes its corporate income taxation dues. This tax liability is strongly alleviated
when subsidies are included in the analysis. The Belgian federal tax deduction has a modestly
relieving effect, but the Flemish ecology premium can compensate for approximately one third
of capital investment, when certain conditions are met. Together, these two measures offset
the burden of taxation.
Fourthly, the sensitivity analyses pointed at the importance of the annual load hours of the
system and the electricity price. The ceteris paribus sensitivity analysis accredited equal weight
to the load hours and the electricity price in terms of influencing the project’s NPV, but the
Monte Carlo simulation showed a stronger effect from the load hours. The significance of the
electricity price is logical: electricity production is the main goal of the ORC and it is used to
offset part of the electricity bill. The strong influence of the load hours on the project results is
explained by the direct link between load hours and electricity production, and thus – again –
the income generated by the ORC system. Moreover, the investment costs come out with much
more importance than the annual O&M costs in the case study. The ceteris paribus sensitivity
analysis confirms that a change of 10 % in the investment costs brings about a stronger change
in the project’s NPV, while a change of 10 % in the annual costs has an almost negligible effect.
This finding holds for the case of heat recovery projects, where the annual costs are limited to
O&M efforts. In other settings the ORC may encounter fuel costs and the impact of these can be
somewhat different. Also, changing these parameters with only 10 % may fail to take account of
their complete range of realistic values. The Monte Carlo simulation varied the capital and the
O&M costs on probability ranges established from other cases and revealed the annual O&M
costs as the third most important contributor to changes in the project’s NPV. This confirms that
the annual O&M costs of the case study are relatively low and can have a greater impact than
suggested by the simple project assessment and ceteris paribus sensitivity analysis.
Finally, the results of a project assessment are based on ex-ante assumptions and projections,
but may turn out differently in practice. The plant manager responsible for the ORC system
reported ongoing issues to keep all ORC modules operational. Due to recurring defaults, one or
more of the three installed modules is regularly out of operation. The outages or reduced
production have several origins, mostly related to the integration of the ORC system into the
existing plant. In fact, many of the issues encountered can probably be attributed to the fact
that this ORC system was the first one installed by the vendor, a third-party supplier that did not
design the ORC module itself.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
129
4.7. Chapter conclusions
Improving the efficiency of energy use is increasingly acknowledged as an important means to
address resource- and climate-related challenges. The central topic in this chapter is the
improvement of energy use efficiency via electricity generation from industrially recovered
heat, using ORC technology.
It is commonly assumed that the potential of industrial excess heat recovery is substantial.
Although this is a reasonable hypothesis, a literature review reveals there is little empirical
evidence to support it. Nevertheless, ORC technology is increasingly used in industrial heat
recovery applications. Analysing real case studies is key to understand the practical difficulties
encountered and to assess the economic characteristics of ORC investments. The economics of
ORC technology are increasingly incorporated in research, yet most published ORC costs are
estimated rather than real. This chapter discusses the economics of a 375 kW heat recovery
ORC system installed in Flanders, Belgium.
The case study demonstrated how an investment in such a project is not unambiguously
successful. Without subsidies, the project has an IRR of approximately 5.4 % when the firm pays
33.99 % corporate income taxation. The return increases to 6.7 % when the taxation duties are
optimized, but this is likely not sufficient for firms with other investment opportunities and for
an investment in a relatively new technology. The financial appeal of the investment changes
drastically with the inclusion of subsidies. A significant investment subsidy from the Flemish
government and a fiscal advantage from the Belgian federal government lift the return up to
12.6 %. Excluding the inflation assumptions moderates this figure to 10.8 %. This is a
significantly better return than without the subsidies. Whether this is sufficient depends on
each firm’s individual risk and return assessment. The attainable load hours appear to be the
key driver to improve the results, but these depend on the operation of the principal process
and can often not be freely decided. The electricity price is second in importance, but lies
beyond control of the purchasing firm or the ORC vendor. The improvement in the technology’s
financial assessment thus most likely has to come from reductions in its capital and annual
costs. Compared to other renewable power generation technologies, the ORC system performs
reasonably well. Compared to the fossil fuel benchmark, the LCOE of the project remains too
high, even when substantial subsidies are involved.
Some remarks limit the interpretation of this study. The ex-ante assessment of a project’s
profitability is only an estimate and subject to several uncertainties and assumptions. The actual
profitability can only be evaluated after time has elapsed. Moreover, cost-benefit analyses
measure only one aspect of an investment: its cost efficiency. Other criteria such as
environmental performance are not considered. Although more empirical research is required
to draw general results, the study gives rise to the conclusions that heat recovery ORC
CHAPTER FOUR
130
investments are financially feasible when supported financially, and that excess heat recovery
gains appeal with rising energy prices.
This study discloses interesting tracks for future research. One issue to be addressed is a
quantification of industrial excess heat potentials. Although assumed large, the actual quantities
of excess heat are generally unknown. Insight in not only the availability of excess heat, but also
its technical and economic potential would allow a sound assessment of the energy efficiency
potential of heat recovery. Secondly, more empirical research is imperative to improve the
understanding of ORC economics. Insight in the economics allows an assessment of the actual,
practical potential of ORC technology for industrial heat recovery. The adoption of ORCs as
solution depends on its technical and economic parameters, but also on those of competing
technologies.
CASE STUDY OF AN ORGANIC RANKINE CYCLE APPLIED FOR EXCESS HEAT RECOVERY
131
Bibliography
Agentschap Ondernemen. (2014). EP-PLUS Algemene informatie ecologiepremie. Agentschap Ondernemen.
Agentschap Ondernemen. (2016). Limitatieve technologieënlijst. Technologieën alfabetisch gerangschikt per type. Geldig vanaf 21/04/2016.
BE. (2014). Belgian Energy Efficiency Action Plan According to the Directives 2006/32/EC and 2012/27/EU article 24.2 Annex XIV part 2. Retrieved from
Bejan, A., Tsatsaronis, G., & Moran, M. (1996). Thermal design and optimization New York: Wiley & Sons. EC. (2014). Europe 2020 - Europe 2020 in a nutshell - Targets. Retrieved from
http://ec.europa.eu/europe2020/europe-2020-in-a-nutshell/targets/index_en.htm Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy
efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC, (2012).
EUCO. (2014). European Council (23 and 24 October 2014) Conclusions. Brussels: European Council. Eurostat. (2016). Final energy consumption by sector. Retrieved from:
http://ec.europa.eu/eurostat/tgm/refreshTableAction.do?tab=table&plugin=1&pcode=tsdpc320&language=en.
Wetboek van de inkomstenbelastingen 1992 (versie aanslagjaar 2015), (2014). IRENA. (2015). Renewable Power Generation Costs in 2014. International Renewable Energy Agency, 159
pp. Jelen, F. C., & Black, J. H. (1983). Cost and optimization engineering (Second ed.): McGraw-Hill, Inc, 538
pp. Joskow, P. L. (2011, Februari 9, 2011). Comparing the costs of intermittent and dispatchable electricity
generation technologies. Kost, C., Mayer, J. N., Thomsen, J., Hartmann, N., Senkpiel, C., Philipps, S., . . . Schlegl, T. (2013). Levelized
Cost of Electricity Renewable Energy Technologies. Retrieved from Freiburg, Germany: Lemmens, S. (2015, June 30-July 3, 2015). The impact of energy policy on energy efficiency in Europe: a
case study of waste heat recovery with an organic Rankine cycle in Flanders. Paper presented at the The 28th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems (ECOS 2015), Pau, France.
Lemmens, S., Verbruggen, A., & Couder, J. (2016). Excess heat recovery with organic Rankine cycles (ORC) in Flanders: policy setting. Antwerp, Belgium, University of Antwerp: 38 pp.
Leslie, N. P., Sweetser, R. S., Zimron, O., & Stovall, T. K. (2009). Recovered Energy Generation Using an Organic Rankine Cycle System. ASHRAE Transactions, 115(Part 1).
MIRA. (2016). Energiegebruik per sector. Retrieved from: http://www.milieurapport.be/nl/feitencijfers/sectoren/energiesector/energiegebruik-in-vlaanderen/energiegebruik-per-sector/.
Peters, M. S., Timmerhaus, K. D., & West, R. E. (2004). Plant Design and Economics for Chemical Engineers (Fifth ed.): McGraw-Hill, 988 pp.
Short, W., Packey, D. J., & Holt, T. (1995). A Manual for the Economic Evaluation of Energy Efficiency and Renewable Energy Technologies. Retrieved from Colorado, USA NREL National Renewable Energy Laboratory (NREL/TP-462-5173), 120 pp.
Tchanche, B. F., Declaye, S., Quoilin, S., Papadakis, G., & Lemort, V. (2010, July 12-14, 2010). Economic feasibility study of a small scale organic rankine cycle in waste heat recovery application. Paper presented at the ASME 2010 10th Biennial Conference on Engineering Systems Design and Analysis (ESDA 2010), Istanbul, Turkey.
CHAPTER FOUR
132
Tumen Ozdil, N. F., & Segmen, M. R. (2016). Investigation of the effect of the water phase in the evaporator inlet on economic performance for an Organic Rankine Cycle (ORC) based on industrial data. Applied Thermal Engineering, 100, 1042-1051. doi:10.1016/j.applthermaleng.2016.02.117
VEA. (2016). Verhoogde investeringsaftrek voor energiebesparende investeringen: Aanslagjaar 2017. Vlaams Energieagentschap.
Verbruggen, A., Fischedick, M., Moomaw, W., Weir, T., Nadaï, A., Nilsson, L. J., . . . Sathaye, J. (2010). Renewable energy costs, potentials, barriers: Conceptual issues. Energy Policy, 38(2), 850-861. doi:10.1016/j.enpol.2009.10.036
Vlaamse Overheid. (2016a). Call groene warmte, restwarmte en biomethaan. Retrieved on 03/01/2016 from http://www.energiesparen.be/groene-energie-en-wkk/professionelen/steunregeling/call-groene-warmte-restwarmte-en-biomethaan.
Vlaamse Overheid. (2016b). Call groene warmte, restwarmte en biomethaan: Voorwaarden. Retrieved on 03/01/2016 from http://www.energiesparen.be/voorwaarden-call.
Vlaamse Regering. (2002). Benchmarking-Convenant over energie-efficiëntie in de industrie zoals goedgekeurd door de Vlaamse Regering op 29.11.2002. Vlaamse Regering.
Vlaamse Regering. (2005). Auditconvenant over energie-efficiëntie in de industrie zoals definitief goedgekeurd door de Vlaamse Regering op 10 juni 2005. Vlaamse Regering.
Vlaamse Regering. (2014a). Energiebeleidsovereenkomst voor de verankering van en voor blijvende energie-efficiëntie in de Vlaamse energie-intensieve industrie (niet VER-bedrijven) zoals goedgekeurd door de Vlaamse Regering op 04.04.2014. Vlaamse Regering.
Vlaamse Regering. (2014b). Energiebeleidsovereenkomst voor de verankering van en voor blijvende energie-efficiëntie in de Vlaamse energie-intensieve industrie (VER-bedrijven) zoals goedgekeurd door de Vlaamse Regering op 04.04.2014. Vlaamse Regering.
VLAIO. (2016a). Vlaamse Groenestroomcertificaten (GSC). Retrieved from http://www.vlaio.be/maatregel/vlaamse-groenestroomcertificaten-gsc
VLAIO. (2016b). Vlaamse warmtekrachtkoppelingscertificaten (WKC). Retrieved from http://www.vlaio.be/maatregel/vlaamse-warmtekrachtkoppelingscertificaten-wkc.
VREG. (2016). Warmte-krachtcertificaten. Retrieved on 13/02/2015 from http://www.vreg.be/nl/warmte-krachtcertificaten#sect1.
5. The dynamics of ORC technology:
technological innovation,
economies of scale and
learning by doing
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
135
5.1. Introduction
An important aspect in the study of innovations is the role of technology learning. It entails
research, development and demonstration (learning-by-searching), feedback from technology
users (learning-by-using), cost reductions as a result of accumulating experience (learning-by-
doing) or the way in which network interactions influence the diffusion of the technology
(learning-by-interacting). Other, related aspects include the cost dynamics from upscaling of the
technology (economies of scale) or from upscaling the production facilities (returns to scale).
The diffusion path of a particular technology is often visualized as a technology life-cycle, where
the deployment starts with its initial invention, followed by research, development and
demonstration and – in some cases – ends in market maturity and saturation.
This chapter investigates the innovation path of ORC technology along three key, quantifiable
dimensions: the diffusion of the technology, its economies of scale and the impact of
accumulating experience on market prices. To this end, two extensive databases have been
composed. The first comprises approximately 95 % of all ORC systems commissioned or sold
worldwide. The second contains valid cost data on more than 100 ORC systems, in different
applications and from various regions. By knowledge of the author, the diffusion and costs of
ORC technology have never been studied at such a scale before.
The chapter is organized as follows. Section 5.2 studies the innovation path of ORC technology.
After an exploration of the essential background on technology life-cycles, this section
investigates the past deployment of ORC technology. This section uses the first database and
provides insight in not only the historical diffusion, but also the current market distribution in
terms of applications, manufacturers and location. Section 5.3 investigates the cost
development of ORC technology. First of all, subsection 5.3.1 discusses the methodology
applied to collect the second database on ORC economics and explores the data. Then, the
database is used to assess the economies of scale for ORC technology. Finally, the two
databases are combined to investigate whether there is a correlation between the average
annual price of an ORC investment and the accumulating worldwide experience with the
technology. The evolution of the costs is analysed in terms of experience than merely in
function of time, since time trends simply reflect an evolution over time but give no insight in
the underlying drivers of change. Moreover, time in itself cannot be changed and gives the
entrepreneur or policy-maker no tools at hand. Section 5.4 analyses the results, with a focus on
the formation of the ORC market, the competition among the technology manufacturers, and
the evolution of the technology’s costs. A final section draws the conclusions.
CHAPTER FIVE
136
5.2. The innovation path of ORC technology
5.2.1. The technology life-cycle
Shifts in technological systems and techno-economic paradigms occur only when the combined
effect of multiple radical and incremental innovations has far-reaching impacts. Clearly, there
are significantly more innovations without such pervasive effects. The course of one single
innovation is commonly represented by a technology life cycle. The innovation-development
process goes from the recognition of a problem, need or opportunity over basic and applied
research to development, commercialization, diffusion and adoption. A final step is the
consideration of the impact of the technology’s adoption on its environment. (Rogers, 2003)
The adoption of a technology typically follows a bell-shaped pattern, so that the cumulative
adoption is represented by an S-shaped curve (see Figure 5.1). Adoption of the technology
occurs slowly at first, with only a few early adopters. As knowledge about the new technology
spreads and experience increases, the cumulative adoption accelerates to the maximum of the
bell-shaped curve. After this point, adoption continues but at a decreasing rate: fewer potential
adopters remain.
The development process is commonly subdivided into the stages invention, innovation,
diffusion, and ultimately maturity. Grübler, Nakicenovic, and Victor (1999) identify six, more
detailed stages: the diffusion stage is characterized by niche market commercialization and
pervasive diffusion; and the phase of senescence is added at the end of the life cycle. In the first
stage, invention, a new idea, process or material is coined (Jaffe, Newell, & Stavins, 2004; Jaffe
& Stavins, 1994; Rogers, 2003). An invention is often the result of knowledge creation through
fundamental or applied research, but it does not necessarily contain insight in its potential
application (Grübler, 1998; Rosenberg, 1996b). Subsequently, continued research discloses
specific characteristics or applications for the invention. Innovation occurs when the invention
sees its first practical application. The technology is developed and demonstrated in a
laboratory setting, ready for adoption. Diffusion occurs when the technology is adopted widely.
Rogers (2003) defines diffusion as “the process by which an innovation is communicated
through certain channels over time among the members of a social system”. Grübler et al.
(1999) differentiate diffusion into niche market commercialization, characterized by niche
applications and important learning effects for both producers and end-users, and pervasive
diffusion, when the technology finds application in different markets, and standardization, mass
production and economies of scale come about. In the early phase of diffusion, the technology
comes in various designs and is improved by learning from R&D and experimentation. The costs
are typically high but the technical promises of the new technology attract a lot of firms that try
to capture a share of the market. The volatility of the early diffusion phase moderates as the
technology has proven its merits and production is increasingly standardized. The design and
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
137
production process of the technology has improved through accumulated experience, and the
market share grows progressively. The market is characterized by higher concentration of
suppliers due to drop-outs, mergers and acquisitions. At the same time, there are fewer
remaining technology designs but these are incrementally improved to enhance performance
and address new applications. In contrast to the early phase of diffusion, the innovation is
incremental and the introduction of new designs is less frequent. (Grübler, 1998) At some point
the technology reaches maturity and there is a saturation of the market. There are no more
economies of scale and the technology is challenged by new, better competitors. The supplier
market is concentrated, the technology is mass-produced and competition focuses on cost
reductions rather than design improvements. But with the eminence of the technology comes
increased awareness of its drawbacks, and the technology becomes the subject of increasing
regulatory requirements. When the final stage occurs, senescence, the technology loses market
share due to competition from other (new) technologies. (Grübler, 1998; Grübler et al., 1999)
Figure 5.1. (a) Technology adoption and (b) cumulative technology adoption, over time.
Legend: The cumulative adoption of a technology is typically represented by an S-shaped curve. The early adopters launch the
diffusion process. The rate of adoption increases when experience with the new technology accumulates. After a certain point,
there are fewer potential adopters and the rate of adoption slows down.
CHAPTER FIVE
138
5.2.2. The technology life-cycle: use and interpretation
As discussed by Rosenberg (1996a), this textbook representation of a technology’s life-cycle is a
rather simplified sketch of reality. One concern is the treatment of continuity in stages of the
life-cycle. Sustaining the influential views of Schumpeter, many scholars treat the different
stages as distinct and separable. Schumpeter put emphasis on the pivotal role of the
entrepreneur and therefore made a strong distinction between the invention, the innovation
carried out by the entrepreneur, and the imitation stage, where other market players merely
imitate by doing similar things. This focus on the entrepreneur was the basis for the
interpretation of technological innovation as a discontinuous process. Besides, Schumpeter’s
analyses relate to major, radical innovations. His ideas have been truly influential and their
application has been extended to all sorts of innovation. However, the practical course of a
technology’s change process is far less partible and the characteristic nature of radical and
incremental innovations is not the same. (Rosenberg, 1996a)
Prior to analysing innovation, the context and drivers of the initial inventive activity have to be
understood. Defining an invention as the launch of a new idea, process or material leaves room
for interpretation. Does this point coincide with the earliest conception of the innovation, when
all the technical issues still have to be solved? Or does invention mark the moment when the
technical feasibility is demonstrated? In the former case, invention refers to the (scientific)
conceptualization, but there is mostly no new product or process readily available. The latter
interpretation takes the significance of engineering and technology experimentation and
knowledge into account. (Rosenberg, 1996a) In any case, appointing a particular date stamp to
an invention remains an arbitrary decision. Taking again the example of the invention of the
steam engine, which is commonly attributed to James Watt in 1781. Indeed, James Watt
patented the rotary-motion steam engine, but his invention builds on the work of preceding
scientists such as, among others, Jerónimo de Ayanz y Beamont (patented the first steam
engine in the early 17th century), Thomas Savery (steam pump, end 17th century), and Thomas
Newcomen, who invented the first commercial piston steam engine (early 18th century). Watt’s
engine was a modification of Newcomen’s one. The invention of lightweight steam engines,
which were small enough for widespread usage and for driving steam locomotives, is accredited
to Richard Trevithick (end 18th century). This example demonstrates that an invention is often
the result of multiple previous inventions and conceptions. Moreover, Rosenberg (1996a)
argues that assigning the date of invention to the first scientific conception neglects and
underrates the value of engineering knowledge that leads to the actual, technologically feasible
product or process. This ought not to be the case. Assigning a date stamp to an invention is
useful to situate the point in time, but the context should be made explicit, to understand the
anterior and contemporary dynamics in which the invention was made. The value of subsequent
research and development can be accredited for by elaborating on its course and merits.
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
139
A related question concerns the time interval between invention and innovation. As discussed
by Rosenberg (1996a), Enos (1962) investigated the timing between invention and innovation of
46 innovations. There appears to be a large variance in the time period between the market
invention and the innovation, from one year for Freon refrigerants to 79 years for the
fluorescent lamp. Such research exemplifies that the goal of identifying time ‘lags’ between
invention and innovation says little about the reasons for its existence. The goal should not be
the delineation of the time lag itself, but rather the disclosure of the technology’s development
dynamics between invention and actual innovation. Between these, there are steps from
concept to laboratory-scale application, to real-scale technical feasibility, and – last but not least
– to economic feasibility. The same remarks hold for the distinction between innovation and
diffusion. Innovation simply marks the first step in the diffusion process. Even after the first
commercial steps have been taken, a technology continues to change. Technical changes
advance matters such as functionality or efficiency, modify the technology for use on other
markets, but also change its economic layout. Again, the inventor and innovator may be
accredited rigorously, but also the subsequent alterations are of significant importance.
(Rosenberg, 1996a) In summary, the s-curve of technology diffusion is not explanatory: it gives
no insight in the underlying reasons for the observed development. (Grübler, 1998)
A second comment on the use of technology life-cycles concerns its interpretation and
extrapolation. The s-curve has been identified for numerous past innovations, but this gives no
guarantee that new innovations will follow the same path, neither that a particular invention
will be successful at all (Rogers, 2003). The creation of scientific knowledge is no warranty for its
transformation into applicable, economically valued knowledge. Only some, but not all, of the
created scientific knowledge will be transferred. (Rosenberg, 1996a) Even for technologies with
proven technical potential, it is hard to gauge the impact (Rosenberg, 1996b). There are ample
historical examples of inventions that were assessed to have minor importance, but which have
a significant impact on today’s life (Rosenberg, 1996b). For instance, in the early days of the
computer it was evaluated to be useful for faster only calculation in a limited number of
contexts and the worldwide market potential was estimated at just a few (Rosenberg, 1996b). In
the same way, there are many examples of technologies that did not win the battle, although
their performance is better than the currently dominant alternative. Think of the conventional
refrigerator today used in any household: it runs on a motor, driving a compressor and
condensing the working medium. An alternative design built in the early days of modern
refrigeration is a gas refrigerator, which is noiseless and less prone to break down because it
contains no moving parts. The first type is currently dominant because large companies involved
in the market assessed its profit potential higher. Hence, large R&D funds and marketing
campaigns were committed to the first type and the gas refrigerator was competed out of the
market, although technologically not inferior. (Rogers, 2003)
CHAPTER FIVE
140
The uncertainties of the innovation process prevail not only in the early days of the
development, but also after it has entered into commerce. A first uncertainty concerns the
inherent difficulty to identify potential applications, among other things because new
technologies are mostly seen in relation to established technologies and practices. Secondly, an
innovation often requires complementary technologies or inventions to enter into wider
diffusion. A related uncertainty is the difficulty to see the new technological system in which the
innovation will eventually reside. Major innovations often require a completely new
technological system, which is difficult to conceptualize in advance. Thus, the potential of the
technology will likely be misjudged because it is seen in light of the contemporary systems.
Moreover, new inventions may have very unanticipated applications. An invention often arises
as the result of research to solve one particular problem. It is not unusual that inventions
appear to have useful applications in very different fields. More extensively, this implies
unanticipated impacts of the technology on other industries and contexts. A final uncertainty
resides in the cost-effectiveness of the technology and hence its demand. Technical feasibility is
no guarantee for economic success. (Rosenberg, 1996b)
5.2.3. The innovation path of ORC technology: invention, research and development
The ORC is conceptually based on the conventional steam Rankine cycle. This cycle was invented
in 1859 by William Rankine, and since then technologically highly refined and widely applied in
steam cycle power plants. The idea of using organic fluids was coined not long after the
invention of the conventional steam cycle, but the idea remained unused, or at least subsurface,
until the second half of the 20th century. It was the research by physicist Harry Zvi Tabor and
engineer Lucien Bronicki that lead to the construction of the first working ORC installation in
1961. Driven by the quest to harness more solar energy, they developed a cycle suitable of
recovering heat at lower temperatures than conventionally. The technical feasibility of the
technology was demonstrated. Although ORC technology did not conquer the market, the first
steps in adoption were taken. In the subsequent decades, ORC research continued on the back
burner, as demonstrated by academic research output.
Academic publications
One measure to quantify R&D efforts is the publication of academic research. Mid-July 2016,
the Web of Science contained 1904 publications on ORC technology (Thomson Reuters, 2016). A
topic search for ‘organic Rankine cycle’ yields 1835 records, whereas querying the topic ‘ORC’
gives 2823 publications, but only 69 of these are relevant for this selection (e.g., ORC as
abbreviation for other research topics, unrelated to organic Rankine cycle technology). Figure
5.2 and Figure 5.3 display, respectively, the annual count of ORC-related publications and their
annual citation count. Figure 5.2 shows an ongoing but very moderate research output in the
period 1968-2006, but from 2007 onwards the number of publications increases sharply, with
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
141
only a minor reduction in 2008. Whereas in 2005 only 12 articles were published, the annual
output was rated at 423 in 2015. The increase from 2014 to 2015 is less sharp than in the
previous years, but the citation report (Figure 5.3) shows the academic community is still very
brisk.
Figure 5.2. Annual count of ORC-related publications in the Web of Science.
Legend: Presentation of the annual number of publications on ORC technology in the Web of Science (collected mid-July 2016).
The annual number of publications increased sharply from 2007 onwards. At the time of collection, the Web of Science contained
1904 publications on ORC technology. Source: Thomson Reuters (2016).
Figure 5.3. Annual citation count of ORC-related publications in the Web of Science.
Legend: Presentation of the annual number of citations of publications on ORC technology in the Web of Science (collected mid-
July 2016). The research community is very brisk, particularly from 2009 onwards. Source: Thomson Reuters (2016).
CHAPTER FIVE
142
5.2.4. The innovation path of ORC technology: technology diffusion and market formation
As the first leaps in the technological innovation process have been taken, the research
community is at cruising speed and ORC technology is increasingly applied in practice. The
diffusion path of the technology is investigated by means of a database, containing most of the
installed and contracted ORCs worldwide. The database was collected from reference lists
obtained from manufacturers and presented on their websites, from the literature, and from
online querying. A first version of the database was composed end 2015 (Lemmens, 2016), and
updated in 2016. The publication of the online ORC World Map (Tartière, 2016) early in 2016
allowed the addition of 91 supplementary records to the database. The total count in mid-July
2016 amounts to 736 ORC references worldwide, of which 113 under construction and 10 of
which neither the sales date nor the start year could be identified. The latter will be disregarded
for the analysis, which leaves 726 records under investigation. The database is not fully
exhaustive because some ORC manufacturers consider their reference list as confidential, but it
likely covers 95 % of the globally installed or under construction ORC systems. References of the
following manufacturers are included: ABB, Adoratec, BEP Europe (E-Rational), Calnetix, Dürr-
Cyplan, Electratherm, Enertime, ENEX, Enogia, Exergy, General Electric, GMK, gTET (gT Energy
Technologies), Johnson Control, Kaishan, Opcon, ORMAT, TAS, TMEIC, Triogen, Turboden and
UTC Power.
The database records hold information about the start year, heat source, manufacturer,
location, installed capacity, and, if available, application, end-user and the manufacturer’s ORC
product name. The heat sources are classified into four general categories: biomass,
geothermal, heat recovery, and solar. The biomass category includes solid biomass sources such
as pellets and waste from the wood industry, and other biofuels. The selection of ‘heat
recovery’ ORC systems refers to a broad range of excess heat sources, including internal
combustion engines, biogas engines, industrial plants, gas pipelines, landfills, and waste
incinerators. The categorization of geothermal and solar sources is straightforward. With this
information, the development of the worldwide ORC market is analysed from several
perspectives.
Worldwide installed ORC capacity: historical trend
The first ORC was constructed in 1961, and only a few systems were built in the following
decade (cfr. section 5.2.3). Because the details on the earliest references could not be traced
back, the first ORC recorded in the database stems from 1975. Figure 5.4 displays the
cumulative historical growth of ORC installations worldwide since 1975. The market growth in
the first decades of diffusion was very slow: by 1990 only little more than 20 references were
installed, by 2000 this number increased to 35. The diffusion speed increased after 2003 and the
growth rate continues to augment since then. Whereas about 80 ORC systems were sold in
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
143
2005, this number increased to over 250 in 2010 and more than 600 in 2015. Mid 2016, the
database holds 110 additional units under construction.
The early ORC systems were installed for heat recovery, but their number increased very slowly
at first. In 1984, the first geothermal ORC system was operational. As shown by Figure 5.5, the
introduction of typically larger-scale geothermal ORC systems implied a clear growth in total
installed capacity. Geothermal-driven ORCs have fewer units than biomass and heat recovery
systems, but they continue to represent the majority of worldwide installed capacity. The
worldwide installed capacity totalled over 2600 MW in July 2016, and another 750 MW is
currently under construction.
Figure 5.4. Cumulative annual growth of ORC references according to thermal source.
Scope: Global coverage, mid-July 2016.
Legend: Presentation of the cumulative annual growth of the number of ORC systems installed worldwide, classified according to
the heat source used: biomass, geothermal, heat recovery and solar.
Source: The data was collected via an extensive search and contains 726 valid records. The database is not fully exhaustive, but
covers likely 95 % of the globally installed or under construction ORC systems. References of the following manufacturers are
included: ABB, Adoratec, BEP Europe (E-Rational), Calnetix, Dürr-Cyplan, Electratherm, Enertime, ENEX, Enogia, Exergy, General
Electric, GMK, gTET (gT Energy Technologies), Johnson Control, Kaishan, Opcon, ORMAT, TAS, TMEIC, Triogen, Turboden and UTC
Power.
0
100
200
300
400
500
600
700
800
19
75
19
77
19
79
19
81
19
83
19
85
19
87
19
90
19
89
19
93
19
95
19
97
19
99
20
01
20
03
20
05
20
07
20
09
20
11
20
13
20
15
Co
nst
r.
Cu
mu
lati
ve n
um
be
r o
f O
RC
s
Start year
Solar
Heat Recovery
Geothermal
Biomass
CHAPTER FIVE
144
Figure 5.5. Cumulative annual growth of installed capacity of ORC references according to thermal source.
Scope: Global coverage, mid-July 2016.
Legend: The worldwide installed capacity, here classified according to the thermal source used (biomass, geothermal, recovered
heat and solar), totalled over 2600 MW in July 2016, and another 750 MW is currently under construction. Source: cfr. Figure 5.4.
Worldwide installed ORC capacity: distribution according to heat source
Figure 5.6 displays the distribution of the four different heat source categories for (a) the global
number of ORC systems installed or under construction by mid-July 2015 and (b) their installed
capacity. Nearly half of the worldwide installed ORC systems use biomass as input source, about
one third runs on recovered heat and another 17% subtracts their energy from geothermal
sources. Solar ORC systems represent only a minor fraction of the total. The distribution of the
heat sources in terms of worldwide installed capacity shows a somewhat different picture (see
Figure 5.6 (b)). ORC systems driven by biomass and recovered heat represent each 12% of
worldwide installed capacity. The geothermal-driven ORC systems are responsible for three
quarters of the total. They are smaller in number, but have larger capacities per unit. For
biomass and heat recovery ORCs this is the other way around. Figure 5.7 and Figure 5.8 confirm
this statement. Heat recovery ORCs are mostly smaller scaled, but also have application in the 5
to 10 MW range. Biomass systems exist mostly at sizes around 1 MW and show a Gaussian
pattern around this point. There are a few geothermal ORC systems in the smaller scales, but
the bulk of these systems have an installed capacity larger than 10 MW.
0
500
1000
1500
2000
2500
3000
3500
4000
19
75
19
77
19
79
19
81
19
83
19
85
19
87
19
89
19
91
19
93
19
95
19
97
19
99
20
01
20
03
20
05
20
07
20
09
20
11
20
13
20
15
Co
nst
r.
Cu
mu
lati
ve in
stal
led
cap
acit
y (M
W)
Start year
Solar
Heat Recovery
Geothermal
Biomass
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
145
Figure 5.6. (a) Number of ORC references per thermal source; (b) total installed capacity of ORC references per thermal source.
Scope: Global coverage, mid-July 2016.
Legend: Distribution of (a) the amount of ORC systems installed worldwide and (b) their total installed capacity, classified
according to the thermal source used. The majority of the installations used biomass as heat source whereas the majority of the
capacity is provided by geothermal ORCs. Source: cfr. Figure 5.4.
Figure 5.7. Number of ORC references for different size ranges, per thermal source.
Scope: Global coverage, mid-July 2016.
Legend: Amount of ORC systems installed worldwide, per thermal source, classified according to their installed capacity
[MWgross]. Source: cfr. Figure 5.4.
Biomass 46%
Geothermal 17%
Heat Recovery
35%
Solar 2%
(a)
Biomass 12%
Geothermal 76%
Heat Recovery
12%
Solar 0%
(b)
020406080
100120140160180200
Co
un
t
Size [MW]
Solar
Heat Recovery
Geothermal
Biomass
CHAPTER FIVE
146
Figure 5.8. Total installed capacity of ORC references for different size ranges.
Scope: Global coverage, mid-July 2016.
Legend: Illustration of the worldwide total installed capacity of ORC systems, classified according to their installed capacity
[MWgross] and according to the thermal energy source. Source: cfr. Figure 5.4.
Worldwide installed ORC capacity: regional distribution
ORC systems exist today on all continents (see Figure 5.9). About 70% of ORC references
worldwide are installed in Europe, followed by North America (9%), Eurasia and Asia (both 8%).
The predominance of Europe in these statistics is due to the dominant market position of the
Italian-based company Turboden, a pioneering ORC manufacturer with the majority of their
references installed in Europe. The distribution of the installed capacity over the continents is
displayed in Figure 5.9 (b) and shows, again, a different picture. Europe is responsible for only
15% of worldwide installed capacity, which is due to the fact that there are mostly biomass and
heat recovery systems operational (see Figure 5.10 and Figure 5.11). North America has the
largest share of total capacity (27%), followed by Eurasia (21%) and Asia (19%). In almost every
region in the world, the geothermal installations are responsible for the bulk of the installed
capacity. Only in Europe, the majority of installed capacity stems from other sources but
geothermal (besides MENA and South America, with a negligible share of worldwide capacity
and only solar systems operational).
0
500
1000
1500
2000
2500
Inst
alle
d c
apac
ity
[MW
]
Size [MW]
Solar
Heat Recovery
Geothermal
Biomass
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
147
Figure 5.9. (a) Number of ORC references per region; (b) total installed capacity of ORC references per region.
Scope: Global coverage, mid-July 2016.
Legend: Regional distribution of the ORC systems installed worldwide according to their amount and their installed capacity.
MENA = Middle East and Northern Africa; Eurasia = Turkey and the Russian Federation; NA = data not available.
Source: cfr. Figure 5.4.
Figure 5.10. Number of ORC references per region, according to thermal source.
Scope: Global coverage, mid-July 2016.
Legend: Regional distribution of the ORC systems installed worldwide according to their thermal energy source. MENA = Middle
East and Northern Africa; Eurasia = Turkey and the Russian Federation; NA = data not available. Source: cfr. Figure 5.4.
Africa 1%
Asia 8%
Central America
1%
Eurasia 8%
Europe 69%
MENA 1%
NA 1%
North America
9%
Oceania 2%
South America
0%
(a) Africa 4%
Asia 19%
Central America
3%
Eurasia 21% Europe
15% MENA 0%
NA 0%
North America
27%
Oceania 11%
South America
0%
(b)
0
100
200
300
400
500
600
Co
un
t Solar
Heat Recovery
Geothermal
Biomass
CHAPTER FIVE
148
Figure 5.11. Total installed capacity of ORC references per region, according to thermal source.
Scope: Global coverage, mid-July 2016.
Legend: Regional distribution of the ORC systems installed worldwide according to their thermal energy source and total installed
capacity [MW]. MENA = Middle East and Northern Africa; Eurasia = Turkey and the Russian Federation; NA = data not available.
Source: cfr. Figure 5.4.
Worldwide installed ORC capacity: manufacturers’ market share
The ORC market has been growing during the last decade by attracting new entrants. The
market share of the manufacturers, as available in the database, is presented in Figure 5.12.
Note that the database is not fully exhaustive, because it was not possible to obtain the
complete reference list from all manufacturers. Still, the references from the most important
market players are included so the shown market distribution is fairly representative. Figure
5.12 shows the dominance of the two pioneering market players: Ormat and Turboden. Ormat
covers about two thirds of the market in terms of capacity and Turboden is responsible for
nearly half of the number of ORC systems installed worldwide. Figure 5.13 shows for each
manufacturer the number of references and their heat source. It follows that the large majority
of biomass ORCs is commissioned by Turboden, whereas the geothermal market is dominated
by Ormat. New manufacturers are confronted with high R&D expenses and the costs of the
initially developed units is substantial, making market entry challenging. Nevertheless, some
new firms manage to start and they are gaining terrain since 2010 (see Figure 5.14).
0
100
200
300
400
500
600
700
800
900
1000
Inst
alle
d c
apac
ity
[MW
]
Solar
Heat Recovery
Geothermal
Biomass
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
149
Figure 5.12. (a) Number of ORC references per manufacturer; (b) total installed capacity of ORC references per manufacturer.
Scope: Global coverage, mid-July 2016.
Legend: Distribution of the ORC systems installed worldwide according to their manufacturer. Other = ABB, Calnetix, Dürr-Cyplan,
Enertime, ENEX, General Electric, gTET, Johnson Control, Opcon, TAS, TMEIC and UTC Power. Source: cfr. Figure 5.4.
Figure 5.13. Number of ORC references per manufacturer, according to thermal source.
Scope: Global coverage, mid-July 2016.
Legend: Count of the ORC systems installed worldwide according to their manufacturer. Other = ABB, Calnetix, Dürr-Cyplan,
Enertime, ENEX, General Electric, gTET, Johnson Control, Opcon, TAS, TMEIC and UTC Power. Source: cfr. Figure 5.4.
Adoratec 3%
BEP Europe
3%
Electratherm 9%
Exergy 6%
ORMAT 16%
Triogen 4%
Turboden 44%
Enogia 2%
GMK 3%
Kaishan 4%
Other 6%
(a) Adoratec 1%
BEP Europe
0% Electratherm 0% Exergy
10%
ORMAT 65%
Triogen 0%
Turboden 15%
Enogia 0%
GMK 0%
Kaishan 1%
Other 8%
(b)
0
50
100
150
200
250
300
350
Co
un
t
Solar
Heat Recovery
Geothermal
Biomass
CHAPTER FIVE
150
Figure 5.14. Cumulative annual growth of ORC references according to manufacturer.
Scope: Global coverage, mid-July 2016.
Legend: Cumulative count of the ORC systems installed worldwide according to their manufacturer. Other = ABB, Calnetix, Dürr-
Cyplan, Enertime, ENEX, General Electric, gTET, Johnson Control, Opcon, TAS, TMEIC and UTC Power. Source: cfr. Figure 5.4.
5.3. Cost development of ORC technology
The costs to produce a technology do typically not remain constant over time. Technological
progress in terms of acquired know-how, research and development or accumulating
experience will alter a firm’s production function. Alternatively, the production costs can be
influenced by economies of scale, i.e. a situation in which the average costs of the technology
reduces as the output increases. Confusing economies of experience with economies of scale
may lead to wrong interpretations on the drivers of the production cost behaviour. Economies
of scale involve the possibility to reduce the average costs of production by increasing the scale,
at a certain point in time, whereas economies of experience reflect unit cost reductions over
time as the result of experience. It is possible that both effects occur simultaneously, but this is
not necessarily the case. (Besanko & Braeutigam, 2005) To investigate the cost dynamics of ORC
technology, this section investigates both the economies of scale and the experience effects
based on an empirically composed database of ORC system prices. Section 5.3.1 discusses the
collection of valid data for the analysis. The economies of scale are analysed in section 5.3.2 and
the experience effects in section 5.3.3.
5.3.1. Data collection
Collecting representative cost figures was an important hurdle for our analysis. Literature
reviews from Quoilin, van den Broek, Declaye, Dewallef, and Lemort (2013) and Vélez et al.
(2012) discuss the technology, applications and costs of ORC systems. However, the number of
0
100
200
300
400
500
600
700
800
Cu
mu
lati
ve n
um
be
r o
f O
RC
s
Year
Adoratec
BEP Europe
Electratherm
Enogia
Exergy
GMK
Kaishan
ORMAT
Triogen
Turboden
Other
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
151
sources included in these reviews is limited and the origin of the data is not always clear. The
literature review in chapter 2 was more extensive, but revealed that the economics of ORC
systems are still relatively unknown: the majority of ORC costs published concern estimated
rather than actual expenses. The accuracy of such estimated cost data may not be sufficient for
detailed economic analyses (cfr. chapter 3). Therefore, the search for valid ORC cost data is
expanded in this chapter.
Methodology
The methodology for data collection involved multiple approaches to collect as many data as
possible. Data on ORC prices was collected from the academic literature, reports or press
releases published by end-users, presentations by manufacturers, budget offers, personal
communication with end-users and manufacturers, other researcher’s efforts, and via a survey
conducted to understand the experiences of ORC end-users (see Appendix C).
The survey involved an online questionnaire sent to ORC end-users worldwide. The
questionnaire aimed to collect information about the location of the ORC, its heat source,
application and manufacturer, but also the motivation for the end-user to install the ORC
system and the consideration of alternatives instead of an ORC system. Finally, the survey
queried details such as installation date, installed capacity, investment and annual costs and
policy interventions. The contact information of the end-users was collected manually from the
database with worldwide installed ORC systems, because neither of the manufacturers agreed
to participate by sharing customer contact details. Of the 736 ORC systems sold worldwide, 46%
was investigated for contact details. A few of the systems were owned by the manufacturer
itself and were disregarded, for about one third of the investigated sample it was impossible to
find contact details. After sending out the survey, there were 35 respondents in total, of which 9
blanks and 4 erroneous responses, in which an ORC manufacturer completed the survey instead
of an end-user. Another 10 completed the questions on motivation etc., but did not disclose
cost information. There are 12 respondents that give insight in the costs of their system.
The collection of economic data is not a straightforward task because the information is
generally considered confidential, yet these efforts allowed us to compose our second ORC
database with valid information on the costs of 127 ORC systems.
Data prospection
The extensive search for data on the costs of ORC technology lead to a database with
information on 127 systems. Each record of the database contains information about the
commissioning year, the location in case it concerns a real system, heat source, installed gross
capacity (kW), investment costs of the ORC module and/or the complete project, and, if
available, net capacity (kW), application, end-user, the manufacturer’s ORC product name, and
CHAPTER FIVE
152
the annual O&M costs. About half of the collected references provide prices of real ORC
systems, the other half are quotes from budget offers or orientation prices from manufacturers.
The quotes are considered representative of real system costs since they are communicated by
the manufacturers themselves. For 11 of the price quotes it was impossible to identify the
commissioning year. Such references without date stamp allow no meaningful economic
analysis so they are disregarded from the analysis. Additionally, there are three ORCs in the
database which are still under construction. These are also discarded because the final project
price may still differ from the anticipated one. This leaves the total amount of records under
investigation at 113 (60 real and 53 quotes). For 90 of these, there is information on the price of
the complete project and for 52 there is price data for the ORC module itself. There are 29
records with price data for both the module and the project. By knowledge of the author, it is
the largest amount of cost data collected and quality tested for ORC systems.
For this analysis, all prices have been transformed to 2015 Euros, using European Central Bank
exchange rates and the Chemical Engineering Plant Cost Index (CEPCI). The database records
have again been categorized into four categories according to the heat source used: biomass,
heat recovery, geothermal and solar. A closer prospection of the data reveals 2 outliers. One
record is eliminated because it is a 1 kW laboratory prototype. Its scale, setup and context are
different from that of commercial ORC systems in such a way that it would make comparison
invalid. Similarly, a second record was considered an outlier in this analysis because of scale
effects. There is one ORC module with an installed capacity of 33.6 MW, whereas all other
module references have capacities lower than 7 MW. The scale effects and technology of this
large-scale geothermal module are of a different order than those of the smaller-scaled
modules, none of which is used for geothermal conversion. For the ORC projects there are
multiple records with large scales, so that the project data for this record is retained for the
analysis. The elimination of these outliers brings the total of ORC records under investigation at
112, with price data for 50 modules and 90 projects and of which 28 records have data for both
modules and projects.
An overview of the characteristics of the records in the database is presented in Figure 5.15 to
Figure 5.18. The database contains records from each of the heat source categories in a
representative share (Figure 5.15). Secondly, the temporal distribution of the references,
measured by the commissioning year of the system, is presented in Figure 5.16. There are only
two references available from before 1999, but thereafter almost every year is represented.
Thirdly, ORC systems exist in a broad size range, from less than 100 kW to over 50 MW of
installed capacity (section 5.2.4), and this diversity is represented in our cost database (Figure
5.17). Finally, the records in the ORC costs database are categorized according to their specific
investment costs (SIC), in ranges (Figure 5.18). The SIC of an ORC module or project is calculated
as the ratio of the investment cost to the installed capacity. The ORC projects are represented in
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
153
all ranges, from lower than 1000 to more than 10,000 €2015/kW. Most ORC projects have a SIC
between 1000 and 5000 €2015/kW, with the majority of the observations in the 2000 to 3000
€2015/kW category. ORC module investments are mostly smaller than 2000 €2015/kW, but some
records show higher SIC values.
Figure 5.15. ORC cost database prospection: distribution of the references according to thermal source.
Legend: NTOT = 112; NP = 90; NM = 50.
Figure 5.16. ORC cost database prospection: references according to start year.
Legend: NTOT = 112; NP = 90; NM = 50.
Biomass 36%
Geothermal 13%
Heat recovery 47%
Solar 1%
Unknown 3%
0
5
10
15
20
25
19
84
19
85
19
86
19
87
19
88
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
Co
un
t
Start year
CHAPTER FIVE
154
Figure 5.17. ORC cost database prospection: references according to installed capacity, in ranges.
Legend: NTOT = 112; NP = 90; NM = 50.
Figure 5.18. ORC cost database prospection: references according to specific investment costs, in ranges.
Legend: NTOT = 112; NP = 90; NM = 50.
5.3.2. Cost development of ORC technology: economies of scale
Economies of scale occur when a scale or output increase corresponds with an average cost
reduction. Scale economies can originate from, for instance, the possibility to increase task
specialization of workers when the labour force is expanded. Alternatively, they may be the
result of returns to scale. This occurs in a situation where the production output increases
proportionally stronger than the proportionate increase in the production input factors. In such
a situation, increasing the plant production may lead to lower unit costs compared to a plant
which operates at a smaller scale. (Besanko & Braeutigam, 2005) For instance, electricity
generation was found to be subject to increasing returns to scale in the 1950s and 1960s
(Nerlove, 1963), but studies in the 1970s show that large-scale plants at that time were
probably marked by constant returns to scale (see Christensen and Greene (1976) and Cowing
and Smith (1978)). This corresponds to the observation that the scale of the power generation
05
10152025303540
Co
un
t
Installed capacity [MW]
Unknown
Solar
Heat recovery
Geothermal
Biomass
0
5
10
15
20
25
30
Co
un
t
Specific investment costs [€2015/MW]
Project
Module
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
155
units increased more strongly during the former period than the latter (Besanko & Braeutigam,
2005).
The scaling concept is often used in engineering studies to relate the costs of systems of
different scales (cfr. section 3.2). The relation is represented as:
c2
c1= (
P2
P1)
SF
(5.1)
with c1 the production costs of the first system; c2 the production costs of the up-scaled system; P1 the size of the first system; P2 the size of the up-scaled system; SF the scale factor.
The scale factor SF indicates whether the costs of a larger-scaled system increase more (SF > 1)
or less (SF < 1) than proportional compared to the size increase from the smaller to the larger
system. For energy systems, the size P refers to the capacity of the system.
The economies of scale for ORC technology
The economies of scale for ORC technology are assessed based on the collected sample of ORC
system prices (see section 5.3.2). The analysis is performed for different subsets of the dataset.
First, the scale effects are investigated for the complete dataset. In the dataset, the costs of the
projects are in general higher than the costs of the modules, a logical outcome (see Figure 5.19).
The study of the complete dataset suggests a positive scale effect for both the ORC projects and
the modules (see Table 5.1), i.e. upscaling the capacity leads to a less proportionate increase in
the price. However, considering scale-effects over such a long period of time likely includes not
only the effect of scale but also that of learning-by-doing (cfr. section 5.3.3), which makes it
difficult to discriminate between those effects. To exclude the influence of learning-by-doing
and focus on the scale effects solely, the analysis is performed for samples with data from one
year each (Koornneef, Junginger, & Faaij, 2007). Only samples with a minimum of four records
are considered. With a few exceptions, the statistical significance of the curve fittings is very
good and again, the data suggests strong economies of scale (Table 5.1). Figure 5.20 displays
the cumulative frequency function of the ORC modules and projects in the dataset. The graph of
the modules shows at least two noteworthy inclinations: one around a SIC of 1000 €/kW and a
second one around 1600 €/kW. A flatter course of the curve exists after the SIC exceeds 2200
€/kW. The curve for the ORC projects shows steps around a SIC of 1400 €/kW, 2400 €/kW, 2800
€/kW and 3200 €/kW. Hence, the costs of the modules and projects are typically situated
around these respective values.
CHAPTER FIVE
156
Figure 5.19. Specific investment costs of ORC systems: complete dataset.
Legend: Illustration of the specific investment costs of ORC projects and modules as a function of their installed capacity
(NTOT = 112; NP = 90; NM = 50). Project scale factor = 0.82 (R² = 0.77); module scale factor = 0.75 (R² = 0.9).
Figure 5.20. Cumulative frequency of ORC specific investment costs: complete dataset.
Legend: Illustration of the cumulative frequency of specific investment cost ranges for ORC projects and modules
(NTOT = 112; NP = 90; NM = 50).
-
5.000,00
10.000,00
15.000,00
20.000,00
25.000,00
- 10.000 20.000 30.000 40.000 50.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Complete dataset
M complete dataset
Power (P Complete dataset)
Power (M complete dataset)
0
20
40
60
80
100
Co
un
t
SIC range [€/kW]
Modules
Projects
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
157
Table 5.1. Economies of scale for ORC projects and modules: complete dataset.
Project Module Note
Data sample
Curve fitting costs vs. size
R² Sample size
Curve fitting costs vs. size
R² Sample size
All data 𝑦 = 11903𝑥0.8231 0.77 90 𝑦 = 7076.8𝑥0.7493 0.90 50 - 2002 𝑦 = 98859𝑥0.4798 0.99 4 - - - - 2006 𝑦 = 93994𝑥0.5498 0.90 6 - - - - 2007 - - - 𝑦 = 2818.7𝑥0.885 1 5 - 2008 𝑦 = 35562𝑥0.6427 0.59 13 𝑦 = 7764.9𝑥0.7296 0.74 10 - 2009 𝑦 = 40561𝑥0.6784 0.63 13 𝑦 = 63629𝑥0.4445 0.89 5 - 2010 𝑦 = 2099.8𝑥1.0073 0.96 6 - - - - 2011 𝑦 = 4006.2𝑥0.9415 0.94 6 𝑦 = 5744.9𝑥0.7903 0.95 7 - 2012 𝑦 = 7249.9𝑥0.8723 0.83 15 𝑦 = 7543.2𝑥0.7427 0.86 14 -
Legend: Overview of the economies of scale for the complete dataset, measured for the entire sample and for one-year
subsamples to minimize the impact of learning-by-doing. Note the poor goodness-of-fit for some project samples (2008, 2009).
However, the dataset contains ORC systems with different applications and constructed by
various manufacturers. These subcategories may involve different scale effects which are not
acknowledged when the dataset is investigated entirely at once. Therefore, the scale effects are
investigated for the different heat source categories, but also according to the manufacturer
that produced the ORC. The four heat source categories considered are recovered heat,
biomass, geothermal and solar sources. Figure 5.21 displays an overview of the specific
investment costs of the ORC projects and modules, as a function of their installed capacity. Each
of the heat source categories is present in the database, but there are not sufficient records to
make assertions about the economies of scale for solar ORC systems. The figure shows that
most of the large-scale systems are geothermal-driven. ORC systems driven by biomass sources
and using recovered heat are typically smaller. The heat recovery ORCs show a wider power
range, up to 6 MW, whereas all the biomass systems reported are smaller than 2.5 MW.
Because of the apparently large difference in size ranges, it makes sense to investigate each of
the different heat source categories separately.
CHAPTER FIVE
158
Figure 5.21. Specific investment costs as a function of installed capacity: complete dataset per heat source.
Legend: Illustration of the specific investment costs of ORC projects and modules as a function of their installed capacity,
categorized according to the thermal source used (NTOT = 112; NP = 90; NM = 50).
Heat recovery systems typically have a smaller scale than geothermal systems, but can still exist
up to several megawatts. The systems observed in our cost database range from 30 kW to 6
MW. In general, the SIC of heat recovery ORCs decreases as the installed capacity of the system
increases (Figure 5.22). The pattern is most distinct for the modules, whereas the heat recovery
projects show a more scattered course. This finding is not surprising, since the integration of the
ORC modules into the existing facilities varies strongly according to the case at hand. These
diverse integration requirements are reflected in the diverging heat recovery projects costs.
Also the heat recovery sample is divided into subsamples according to commissioning year in
order to measure potential scale effects. The heat recovery modules display, with very strong
statistical significance, a positive influence from upscaling with scale factors ranging from 0.46
to 0.89 depending on the analysis year (see Table 5.6). The quality of the heat recovery project
curve fitting is not as good as for the modules, but suggests scale factors in the range between
0.8 and 0.9. Finally, Figure 5.23 displays the cumulative frequency distribution of the specific
investment cost ranges of the heat recovery ORC systems in the database. The graph suggests
that most of the heat recovery modules have a SIC around 1200 €/kW, or in the range from
1600 to 1800 €/kW. The heat recovery projects have less outspoken scale categories, but the
most occurring prices are noted around 1400 €/kW, between 2200 €/kW and 2400 €/kW, and
nearby 3200 €/kW.
100,00
1.000,00
10.000,00
100.000,00
10 100 1.000 10.000 100.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Heat recovery
M Heat recovery
P Biomass
M Biomass
P Geothermal
P Solar
P NA
M NA
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
159
Figure 5.22. Specific investment costs as a function of installed capacity: heat recovery subset.
Legend: Illustration of the specific investment costs of ORC projects and modules, using recovered heat as thermal source, as a
function of their installed capacity (NTOT = 53; NP = 34; NM = 34). Project scale factor = 0.85 (R² = 0.74);
module scale factor = 0.78 (R² = 0.94).
Figure 5.23. Cumulative frequency of ORC specific investment costs: heat recovery subset.
Legend: Illustration of the cumulative frequency of specific investment cost ranges for ORC projects and modules using recovered
heat as thermal source (NTOT = 53; NP = 34; NM = 34).
-
2.000,00
4.000,00
6.000,00
8.000,00
10.000,00
12.000,00
- 1.000 2.000 3.000 4.000 5.000 6.000 7.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Heat recovery
M Heat recovery
Power (P Heat recovery)
Power (M Heat recovery)
05
10152025303540
Co
un
t
SIC range [k€/kW]
M heat recovery
P Heat recovery
CHAPTER FIVE
160
Table 5.2. Economies of scale for ORC projects and modules: heat recovery subset.
Project Module Note
Data sample
Curve fitting costs vs. size
R² Sample size
Curve fitting costs vs. size
R² Sample size
All data 𝑦 = 7995.4𝑥0.8515 0.74 34 𝑦 = 6007.9𝑥0.7782 0.94 34 - 2007 - - - 𝑦 = 2818.7𝑥0.885 1 5 (1)
2009 𝑦 = 322502𝑥0.3212 0.20 5 𝑦 = 53319𝑥0.4618 0.92 4 - 2011 - - - 𝑦 = 5376.3𝑥0.8049 0.95 6 - 2012 𝑦 = 10517𝑥0.8313 0.78 9 𝑦 = 6838.9𝑥0.7641 0.97 11 - 2014 𝑦 = 7044.6𝑥0.9359 0.91 4 - - - -
Legend: Overview of the economies of scale for the heat recovery subset, measured for the entire sample and for one-year
subsamples to minimize the impact of learning-by-doing. Note the poor goodness-of-fit for the 2009 project sample.
Note: (1)
The module sample contains only price quotes from the manufacturer Electratherm.
The biomass ORC systems in the database have capacities ranging from 200 kW to 2.4 MW. The
project prices are generally higher and display a stronger scattering than those of the modules
(Figure 5.24). Considering all biomass ORC together yields statistically insignificant results (Table
5.3). Therefore, the scale effects of the biomass ORC systems are again measured by considering
one-year samples. The results of these subsamples display very high statistical explanatory
powers and significantly positive scale effects. Note that nearly all the considered subsamples
contain only price quotes published by the manufacturer Turboden (only the 2008 project
sample contains one other case). In other words, it appears that the manufacturer itself entails
positive economies of scale and reflects these in its costs. To investigate this finding more
profoundly, the scale effects are analysed for separate manufacturers below. Finally, Figure 5.25
shows that the most common price for biomass ORC projects lies between 2400 and 2600
€/kW. The biomass ORC modules in the database cost mostly around 1000 or 1600 €/kW.
Figure 5.24. Specific investment costs as a function of installed capacity: biomass subset.
Legend: Illustration of the specific investment costs of ORC projects and modules, using biomass as thermal source, as a function
of their installed capacity (NTOT = 40; NP = 38; NM = 15). Project scale factor = 0.36 (R² = 0.24);
module scale factor = 0.44 (R² = 0.37).
-
5.000,00
10.000,00
15.000,00
20.000,00
25.000,00
- 500 1.000 1.500 2.000 2.500 3.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Biomass
M Biomass
Power (P Biomass)
Power (M Biomass)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
161
Figure 5.25. Cumulative frequency of ORC specific investment costs: biomass subset.
Legend: Illustration of the cumulative frequency of specific investment cost ranges for ORC projects and modules using biomass
as thermal source (NTOT = 40; NP = 38; NM = 15).
Table 5.3. Economies of scale for ORC projects and modules: biomass subset.
Project Module Note
Data sample
Curve fitting costs vs. size
R² Sample size
Curve fitting costs vs. size
R² Sample size
All data 𝑦 = 351679𝑥0.3573 0.24 38 𝑦 = 56533𝑥0.4419 0.37 15 - 2002 𝑦 = 98859𝑥0.4798 0.99 4 - - - (2)
2008 𝑦 = 177988𝑥0.4346 0.91 10 𝑦 = 35785𝑥0.505 0.93 9 (3) 2009 𝑦 = 195432𝑥0.4873 1 7 - - - (2)
Legend: Overview of the economies of scale for the biomass subset, measured for the entire sample and for one-year subsamples
to minimize the impact of learning-by-doing. Note the poor goodness-of-fit for overall project and module samples. Notes: (2)
The
project sample contains only biomass price quotes from the manufacturer Turboden. (3)
The module sample contains only price
quotes from Turboden; the project sample contains the same price quotes from Turboden and one real case from Turboden.
The geothermal ORC projects span a very broad capacity range: the smallest record in the
dataset has an installed capacity of 400 kW, whereas the largest has a capacity of 42 MW. Figure
5.26 shows higher investment costs for the smaller projects, but otherwise the course is
relatively flat. There is no information about the costs of geothermal ORC modules in the
database. Also, the curve fitting for all the geothermal projects has a very low goodness-of-fit
(Table 5.4) and there is not sufficient information available to subdivide the geothermal set into
useful subsamples. Hence, no conclusions can be made about the scale effects of geothermal
ORC systems. Finally, Figure 5.27 does not give a strong indications for a most likely geothermal
project investment cost, but values around 1400 €/kW and 2400 €/kW appear slightly more
often.
05
10152025303540
Co
un
t
SIC range [k€/kW]
M biomass
P Biomass
CHAPTER FIVE
162
Figure 5.26. Specific investment costs as a function of installed capacity: geothermal subset.
Legend: Illustration of the specific investment costs of ORC projects and modules, using geothermal energy as thermal source, as
a function of their installed capacity (NTOT = 15; NP = 15; NM = 0). Project scale factor = 0.62 (R² = 0.09).
Figure 5.27. Economies of scale for ORC technology: cumulative frequency curves for the geothermal subset.
Legend: Illustration of the cumulative frequency of specific investment cost ranges for ORC projects and modules using
geothermal energy as thermal source (NTOT = 15; NP = 15; NM = 0).
Table 5.4. Economies of scale for ORC projects and modules: geothermal subset.
Project Module Notes
Data sample
Curve fitting costs vs. size
R² Sample size
Curve fitting costs vs. size
R² Sample size
All data 𝑦 = 5𝐸 + 06𝑥0.6188 0.10 15 - - - - Legend: Overview of the economies of scale for the geothermal subset. The sample is too small to consider separate one-year
subsamples. Note the dramatically poor goodness-of-fit for the project curve fitting.
Finally, the economies of scale for ORC technology are investigated by considering the
manufacturer that constructed the system instead of the heat source that is used for its
operation. Of the 112 records in the database, nearly half of the records stem from the
manufacturer Turboden and only for this manufacturer there is sufficient data to make sensible
-
1.000,00
2.000,00
3.000,00
4.000,00
5.000,00
6.000,00
7.000,00
- 10.000 20.000 30.000 40.000 50.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Geothermal
Power (P Geothermal)
02468
10121416
Co
un
t
SIC range [k€/kW]
P Geothermal
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
163
analyses. There are 53 records for Turboden in the database, with information about 46 projects
and 31 modules. Turboden’s specific project costs are, logically, higher than those of the
modules (Figure 5.28). The project costs range from as high as 11,000 €/kW for a 400 kW system
to 760 €/kW for one of 6 MW. In general, the quality of the curve fitting for Turboden’s projects
is somewhat ambiguous (Table 5.5). The scale effects for all Turboden data do not allow to draw
conclusions, as well as some of the subsamples. Other subsamples with a very strong statistical
significance suggest scale factors in the range 0.43-0.49, and 0.97 for the systems installed in
2012 and operating on excess heat. The most expensive Turboden ORC module is a 500 kW heat
recovery module (2770 €/kW), whereas the cheapest is a 6 MW heat recovery system (425
€/kW). The overall curve fitting for the Turboden modules suggests positive economies of scale
with a scale factor of 0.76 (Table 5.5). A similar result is suggested for the data samples from
2008, and from 2012 using excess heat. However, the biomass subsample from 2008 has a very
strong statistical significance and displays a scale factor of 0.5. Finally, Figure 5.29 displays that
most Turboden projects have a price between 2200 and 2800 €/kW, or around 4400 €/kW. The
Turboden modules in the database mostly have a SIC around 1000 €/kW or 1600 €/kW.
Figure 5.28. Specific investment costs as a function of installed capacity: Turboden subset.
Legend: Illustration of the specific investment costs of ORC projects and modules, constructed by the manufacturer Turboden, as
a function of their installed capacity (NTOT = 53; NP = 46; NM = 31). Project scale factor = 0.59 (R² = 0.49);
module scale factor = 0.76 (R² = 0.81).
-
2.000,00
4.000,00
6.000,00
8.000,00
10.000,00
12.000,00
- 1.000 2.000 3.000 4.000 5.000 6.000 7.000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Power [kW]
P Turboden
M Turboden
Power (P Turboden)
Power (M Turboden)
CHAPTER FIVE
164
Figure 5.29. Economies of scale for ORC technology: cumulative frequency curves for the Turboden subset.
Legend: Illustration of the cumulative frequency of specific investment cost ranges for ORC projects and modules constructed by
the manufacturer Turboden (NTOT = 53; NP = 46; NM = 31).
Table 5.5. Economies of scale for ORC projects and modules: Turboden subset.
Project Module Note
Data sample
Curve fitting costs vs. size
R² Sample size
Curve fitting costs vs. size
R² Sample size
All data 𝑦 = 64679𝑥0.587 0.49 46 𝑦 = 6185.5𝑥0.7619 0.81 31 - 2002 𝑦 = 98859𝑥0.4798 0.88 4 - - - (2) 2008 𝑦 = 156256𝑥0.4549 0.92 11 𝑦 = 7764.9𝑥0.7296 0.74 10 -
2008 BIO 𝑦 = 177988𝑥0.4346 0.91 10 𝑦 = 35785𝑥0.505 0.93 9 (4) 2009 𝑦
= 1𝐸 + 06𝑥0.1931
0.24 11 - - - -
2009 BIO 𝑦 = 195432𝑥0.4873 1 7 - - - (5) 2009 HR 𝑦 = 2𝐸 + 06𝑥0.1187 0.16 4 - - - (6)
2012 𝑦 = 7813.2𝑥0.8406 0.65 10 𝑦 = 2345.5𝑥−0.109 0.06 11 -
2012 HR 𝑦 = 3238.1𝑥0.9712 0.84 9 𝑦 = 8248.2𝑥0.7424 0.92 7 (7) Legend: Overview of the economies of scale for the Turboden subset, measured for the entire sample and for one-year
subsamples to minimize the impact of learning-by-doing. Note the poor goodness-of-fit several project (all data, 2009, 2009 HR
and 2012) and module (2012) samples. Notes: (2)
The project sample contains only biomass price quotes from the manufacturer
Turboden. (4)
A subset of the 2008 Turboden sample, which contains only those using biomass. The module sample contains only
price quotes. (5)
A subset of the 2009 Turboden sample, which contains only those using biomass. The project sample contains
only price quotes. (6)
A subset of the 2009 Turboden sample, which contains only those using recovered heat. (7)
A subset of the
2012 Turboden sample, which contains only those using recovered heat.
0
10
20
30
40
50
Co
un
t
SIC range [k€/kW]
M Turboden
P Turboden
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
165
5.3.3. Cost development of ORC technology: the impact of experience
An important feature of technological change is knowledge, and more particularly the growth of
knowledge over time. It is intuitively clear that one becomes better at performing a task when
one’s knowledge about how to do so expands. This process of knowledge acquisition is
commonly referred to as learning. (Arrow, 1962) From an economic point of view, learning
entails efficiency improvements and thus potential for cost reduction.
Learning by doing and the experience curve
Considering learning as the growth of knowledge over time, one approach is to investigate the
time trend of the performance measure at hand, such as product costs. However, time trends
say little about the forces influencing the variable under investigation and, moreover, time is
not a mouldable parameter (Arrow, 1962). An alternative approach is an examination of the
technology’s performance not as a function of time, but as a function of knowledge
accumulation itself, measured as the cumulative quantities produced. The relation between
experience and increasing productivity was first observed by Wright (1936) for airframe
production. He noted that the required hours of labour decreased as the amount of previously
produced airframes of the same type increased. This experience-productivity relation has been
evidenced through many empirical investigations since then, and is comprised in the so-called
learning curve.
The learning curve is most commonly formulated as follows:
𝐶𝐶𝑢𝑚 = 𝐶𝑒0𝐶𝑢𝑚−𝑚
(5.2)
with 𝐶𝑒0 the cost of the first produced unit;
𝐶𝐶𝑢𝑚 the cost per unit; Cum the total cumulative production; m the experience parameter (the learning elasticity).
When plotted on a log-log scale, the learning curve takes the form of a downward sloping
straight line (see
Figure 5.30) (Junginger, van Sark, Kahouli-Brahmi, & Schaeffer, 2010; OECD/IEA, 2000):
𝑙𝑜𝑔𝐶𝐶𝑢𝑚 = 𝑙𝑜𝑔𝐶𝑒0 − 𝑚𝑙𝑜𝑔𝐶𝑢𝑚.
(5.3)
with 𝐶𝑒0 the cost of the first produced unit;
𝐶𝐶𝑢𝑚 the cost per unit; Cum the total cumulative production; m the experience parameter (the learning elasticity).
CHAPTER FIVE
166
Figure 5.30. The one-factor learning curve on (a) a normal scale and (b) a log-log scale.
Learning curves demonstrate the direction of the experience-productivity relationship. The size
of the effect is quantified by the experience parameter m, the slope of the double-log linear
curve. With the experience parameter m the progress ratio (PR) is calculated:
𝑃𝑅 =
𝐶𝑒0 ∗ (2𝐶𝑢𝑚)−𝑚
𝐶𝑒0𝐶𝑢𝑚−𝑚
= 2−𝑚.
(5.4)
with 𝐶𝑒0 the cost of the first produced unit;
Cum the total cumulative production; m the experience parameter (the learning elasticity); 𝑃𝑅 the progress rate.
The progress ratio reflects the decrease in costs due to cumulative doubling of the experience
index (mostly delivered production or installed capacity). The learning rate (LR) is the
complement of the progress ratio:
𝐿𝑅 = 1 − 𝑃𝑅.
(5.5)
with 𝐿𝑅 the learning rate; 𝑃𝑅 the progress rate.
The learning curve represents the finding that the unit cost of a product decreases with a
constant parameter for each doubling of the experience index (production or capacity). A
progress ratio of, for instance, 80% implies a learning rate of 20%, or a unit cost reduction of
20% for every doubling of cumulative experience (Junginger, van Sark, Kahouli-Brahmi, et al.,
2010; OECD/IEA, 2000).
010203040506070
1 2 4 8
16
32
64
12
8
25
6
51
2
10
24
20
48
Co
st/u
nit
(a) Cumulative production
1
10
100
1 10 100 1000 10000
log
(Co
st/u
nit
)
(b) log (Cumulative production)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
167
The process comprised in the learning curve is known as learning-by-doing. Note the
importance of ‘doing’ in the expression: true learning occurs only when actively solving an issue
or working on a matter. Or as stated by Arrow (1962): “learning is the product of experience”.
This implies that laboratory R&D by itself is not sufficient for a technology to benefit from
experience and become cost-efficient (OECD/IEA, 2000). Moreover, to achieve continuous
improvement the stimulus for learning must itself be evolving instead of just repeating. The
learning in situations of mere repetition diminishes sharply. (Arrow, 1962) This effect is likewise
demonstrated by the progress ratio. A capacity doubling from 1 MW to 2 MW reduces costs by
20%, but when the installed capacity amounts 100 MW an additional 100 MW of capacity is
required to achieve the same effect. (Berglund & Söderholm, 2006) Finally, experience provides
opportunities for cost reduction but it does not by definition cause them. The opposite occurs,
for instance, when cost increases from design changes or performance improvements are not as
quickly offset with cost reductions from learning, standardization, specialization or scale effects.
(Neij, 1997)
The learning curve introduced in equation (5.2) is a single-factor relation, representing unit cost
reduction as a function of accumulated experience. The original annotation of ‘learning’
referred to accumulating experience of workers in the manufacturing process, which reduces
the required labour force. But there are other origins of learning, and associated cost
reductions. For instance, managers improve their role with increasing know-how (Berglund &
Söderholm, 2006) and research and development induce innovation. To distinguish worker
learning from other causes of learning, a distinction is commonly made among learning,
experience and progress curves. Learning curves specifically denote the learning by individual
workers or in a production process, progress curves measure experience at the firm level,
including managerial action, and experience curves are generally used to denote progress at the
level of the industry (Dutton & Thomas, 1984). In the remainder of this work the broader
denotation experience curve will be used.
The effect measured in experience curves is thus the result of a combination of different factors.
Neij (1997) distinguishes between production changes, product changes and changes in input
prices as drivers behind experience curves. Production changes refer to process innovations,
true learning effects and scaling effects, whereas product changes refer to product innovations,
redesigns and standardizations (Neij, 1997). Dutton and Thomas (1984) analyse more than 100
estimated progress ratios, from all kinds of manufacturing industries, and distinguished four
underlying dynamics causing the progress: (1) technological change effects, (2) labour learning
effects, (3) characteristics of the local industry and the firm and (4) scale effects. The first factor,
technological change, refers to the depiction of the suitable experience parameter. Empirical
evidence supports the finding that cumulative gross investment is a better representative of
experience than cumulative production. The second factor relates to the so-called Horndal
CHAPTER FIVE
168
effect, which got its name after discovering an annual 2% labour productivity in the Horndal iron
works plant although no new investments were undertaken in the last 15 years (Lundberg,
1961). The learning is thus due to learning by the labour forces, either direct or indirect,
because the capital stock is fixed. Thirdly, learning effects depend on the organizational
structure of the firm and its local industry. For instance, mechanization degrees, assembly
versus machining labour, continuous versus batch processes and the length of cycle times are
found to influence progress rates. Finally, scale economies refer to average cost reductions due
to increasing scale. Progress curves aggregate scale and non-scale effects so that some of the
progress effects may be due to scale effects. Two factors of cost reduction can be distinguished
here: caused by increased knowledge that is the result of increased cumulative output, and as a
result of a change in expected production volume. The former refers to the experience effect,
whereas the latter points at changing production techniques for larger output volumes. These
four causes of progress take different forms according to their origin (endogenous or
exogenous) and to their type (autonomous or induced). (Dutton & Thomas, 1984) Exogenous
causes of learning stem from external sources of information or as benefits from external
parties, including suppliers, customers, competitors and governments, whereas endogenous
learning is the result of learning within the firm. (Dutton & Thomas, 1984) Induced learning is
learning caused by investment, or resources that have not been available in the past and that
now become accessible. Autonomous learning refers to automatic improvements are a result
from sustained production over a longer time span. (Dutton & Thomas, 1984)
Experience curves: types, variable selection and interpretation
The experience curve is generally defined as the relation between experience and productivity,
but the practical implementation of the concept can take many forms. Firstly, experience curves
can be distinguished in terms of perspective. Experience curves with a production perspective
concern the experience process of the production itself, measured from e.g., different
manufacturers, whereas experience curves with a market perspective measure market
experience, e.g., in different countries (Neij et al., 2003). Alternatively, experience curves differ
in the system scope they cover. An (energy) technology is the aggregate of many
subcomponents, so the learning that occurs is the result of the accumulation of many learning
effects. Experience curves can cover the entire system, or parts thereof. (Neij et al., 2003)
Experience curves also differ by the choice of variables used to estimate the relation. First there
is the choice of the experience parameter. The experience gained by the manufacturer is mostly
approached by cumulative production or cumulative installed capacity. Arrow (1962) argues the
cumulative gross investment should be used to denote experience, because the cumulative
output is not completely representative. In case, for instance, the cumulative output is constant,
this would imply the learning rate to be constant as well, and thus suggest that learning
stabilizes to equilibrium (Arrow, 1962). This is not automatically the case and therefore
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
169
cumulative output is not the best parameter. In practice however, reliable numbers on
technology investment are hard to obtain so cumulative production is mostly used as best
proxy. Second, the choice of the cost parameter impacts the estimation. The best performance
indicator is the cost of producing the technology. Because costs are often not available, prices
are commonly used as a proxy. When using prices instead of costs the analysis requires more
caution. Between costs and prices lies the margin for the manufacturer. This margin is often
unknown but may also vary over time and with competition in the market. Fluctuations in prices
do thus not necessarily reflect changes in production costs, particularly if only short-period
analyses are made. (Junginger, van Sark, & Faaij, 2010) In cases where the return rate for the
producer is rather constant, such as in stable markets, the relation between prices and
production costs is steadier, and the experience effect is the same when measured for either
costs or prices (OECD/IEA, 2000). In practice however, cost information is mostly not available
and moreover, end-users make decisions based on prices in the market and not on production
costs. (Junginger, van Sark, & Faaij, 2010) Experience curves are therefore commonly expressed
in prices, and manufacturer progress curves in unit costs. (Dutton & Thomas, 1984)
An important note is that experience by itself is no guarantee for cost reduction. A study by
(Grubler, 2010) demonstrates how accumulated experience with nuclear power generation
technology is paired by cost increases in the French case. Short-term barriers for cost reduction
are, for instance, market barriers such as high investment costs, low product performance,
information gaps, limited product availability or limited access to capital, but also non-market
barriers such as the rate at which increased production allows for actual cost reduction (Neij et
al., 2003). Long-term limitations for cost reduction exist in the form of physical technology
development limits, limits in the potential to reduce costs further or ceiled market potential
(Neij et al., 2003). Finally, although experience curves correlate the costs of a technology to its
cumulative experience, this does not necessarily entail a causal relationship (Rubin, Azevedo,
Jaramillo, & Yeh, 2015). In other words, merely assuming a continuation of past trends into the
future is a precarious, uncertain practice which requires a thorough understanding of not only
the progress ratios but also other underlying drivers.
CHAPTER FIVE
170
Experience curves for ORC-generated electricity
The experience curves estimated in this work are constructed following the method of Neij et al.
(2003). Experience curves relate the annual average investment costs for a system with the total
installed capacity and are expressed in real terms, i.e. corrected for inflation (Junginger, van
Sark, Kahouli-Brahmi, et al., 2010). The annual average specific investment cost pt, thus the
price of the system, is calculated based on the equation:
pt =
∑ pi. nii=1
∑ ci. nii=1
(5.6)
with ci the capacity of model i; ni the number of systems with model I; pi the unit price of a specific model i; pt The annual average specific investment cost.
Estimating the experience curve of a technology assumes that each implementation of this
technology is constructed in the same way. This is not always the case. For instance, the
practical shape and implementation of ORC technology varies due to the use of different
expander types or working fluids, the characteristics of other selected subcomponents, etc. The
estimations in this chapter make no distinction between the different technological types of
ORCs, which implies the analysis concerns the experience effects of ORC-generated electricity
rather than those of a particular ORC technology implementation. The experience curves are
estimated using the same database (cfr. section 5.3.1) as for the analysis of the scale effects. By
knowledge of the author, the experience effects of ORC technology have not been investigated
at this scale before.
First of all, the experience effects of the complete dataset are investigated: 112 ORC records in
total, with data for 50 modules and 90 projects. The curve fittings are displayed in Figure 5.31
and the numerical results in Table 5.6. The experience curves of the modules and the projects
are plotted on the same graph, on a double-log scale. The quality of the curve fitting is
dramatically low for both curves. This implies that the explanatory power of the data is
insufficient and it is not possible to make sensible conclusions based on these curves. However,
this first rough approach makes no distinction among for instance the heat source used by the
system or the market effects. Note that there is no such thing as the experience curve for one
technology. Different experience curves are obtained when selecting different input variables,
time frames, scopes, etc. So, this overall experience curve shows the overarching pattern of the
data but says nothing about the underlying dynamics. For instance, a closer look shows a
remarkable pattern with two peaks in the annual average investment costs. To investigate the
underlying dynamics of experience with ORC technology more profoundly, the analysis is
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
171
repeated for several subsamples of the dataset: according to the heat source used and
according to the manufacturer of the ORC system.
Figure 5.31. Experience curves for ORC-generated electricity: complete dataset.
Legend: The experience curve fits the average specific investment costs as a function of the worldwide cumulative installed
capacity (2015 Euros; NTOT = 112; NP = 90; NM = 50). Project progress ratio = 113% (R² = 0.12; period: 1984-2016);
module progress ratio = 102% (R² = 0; period: 1984-2016).
Table 5.6. Experience curve fitting for ORC-generated electricity: complete dataset.
Projects Modules
Curve fitting 𝑦 = 1041.6𝑥0.1798 𝑦 = 1247𝑥0.0247
R² 0.12 0.00 Progress ratio (PR) 113 % 102 %
Learning rate (LR) -13 % -2 %
Legend: Overview of the experience curve fitting for the complete dataset (NTOT = 112; NP = 90; NM = 50). Note the very poor
goodness-of-fit for both project and module curve fittings.
Similarly as for the scale effects, the experience analysis is repeated for each of the four heat source
categories separately. First, the ORC systems operating on recovered heat are considered. The database
contains useful data for 53 heat recovery ORCs, with price information for 34 projects, 34 modules and
15 records with insight in both the module and the project price. The database contains sufficient data to
estimate the experience curves for both heat recovery ORCs modules and projects and both data
samples span the period from 1999 to 2015. The experience effect is investigated with respect to the
total worldwide ORC capacity as well as to the worldwide installed capacity of ORCs that operate on
recovered heat. The experience curves are presented in Figure 5.32 and the curve fitting in Table 5.7.
The R² values are very low for all the curve fittings, i.e. no conclusive analysis can be made based on this
data. Considering the estimated learning curves merely indicatively, Figure 5.32 suggests slightly
negative experience effects for both the heat recovery modules and projects. Similarly as with the
complete dataset experience curves, there appears to be a cost increase particularly in the most recent
years. However, these ORCs with (very) large investment costs should not be classified as outliers.
100
1000
10000
100000
10,00 100,00 1000,00 10000,00
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Cumulative installed capacity (MW)
P complete dataset
M Complete dataset
Power (P completedataset)
Power (M Completedataset)
CHAPTER FIVE
172
Rather, they represent the fact that new manufacturers enter the market and charge higher prices that
the experienced ones.
Figure 5.32. Experience curves for ORC-generated electricity: heat recovery subsample.
Legend: The experience curve fits the average specific investment costs as a function of the worldwide cumulative installed heat
recovery ORC capacity (2015 Euros; NTOT = 53; NP = 34; NM = 34). Project progress ratio = 106% (R² = 0.06; period: 1984-2016);
module progress ratio = 101% (R² = 0; period: 1984-2016).
Table 5.7. Experience curve fitting for ORC-generated electricity: heat recovery subsample.
Projects Modules
Experience relative to worldwide cumulative installed heat recovery ORC capacity
Curve fitting 𝑦 = 2131.3𝑥0.0838 𝑦 = 1384.3𝑥0.0142
R² 0.06 0.00 Progress ratio (PR) 106 % 101 %
Learning rate (LR) -6 % -1 %
Experience relative to total worldwide cumulative installed capacity
Curve fitting 𝑦 = 1316.1𝑥0.1184 𝑦 = 1999.4𝑥0.0293
R² 0.05 0.01 Progress ratio (PR) 108 % 102 %
Learning rate (LR) -8 % -2 %
Legend: Overview of the experience curve fitting for the heat recovery subset, as a function of worldwide heat recovery ORC
capacity and total worldwide ORC capacity (NTOT = 53; NP = 34; NM = 34). Note the very poor goodness-of-fit for all project and
module curve fittings.
Secondly, the experience effects of ORC systems operating on biomass sources are investigated.
The cost database contains 40 biomass ORC systems, with data for 38 projects and 15 modules.
The results are displayed in Figure 5.33 and Table 5.8. Note that the quality of the curve fitting is
mediocre for the modules, but not at all for the ORC projects. I.e., for the modules the data
100
1000
10000
0,10 1,00 10,00 100,00 1000,00
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Cumulative installed capacity (MW)
P Heat recovery
M Heat recovery
Power (P Heat recovery)
Power (M Heat recovery)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
173
suggest that there exist learning effects, but for the projects it is not possible to draw concise
conclusions because the data do not show a valuable relation between the costs and the
cumulative installed capacity. These findings are not surprising, since the modules can be
standardized, but the integration requirements for completion of the project vary case by case.
The average costs of biomass ORC modules have decreased at a rate of 7 % per doubling of
worldwide installed biomass ORC capacity and at a rate of 40 % when the worldwide installed
ORC capacity is considered, regardless of the heat source used.
Figure 5.33. Experience curves for ORC-generated electricity: biomass subsample.
Legend: The experience curve fits the average specific investment costs as a function of the worldwide cumulative installed
biomass ORC capacity (2015 Euros; NTOT = 40; NP = 38; NM = 15). Project progress ratio = 100% (R² = 0; period: 1999-2015);
module progress ratio = 93% (R² = 0.69; period: 2001-2012).
Table 5.8. Experience curve fitting for ORC-generated electricity: biomass subsample.
Projects Modules
Experience relative to worldwide cumulative installed biomass ORC capacity
Curve fitting 𝑦 = 5061.8𝑥−0.005 𝑦 = 1959.3𝑥−0.111
R² 0.00 0.69 Progress ratio (PR) 100 % 93 %
Learning rate (LR) 0 % 7 %
Experience relative to total worldwide cumulative installed capacity
Curve fitting 𝑦 = 1311𝑥0.1872 𝑦 = 250313𝑥−0.7373
R² 0.02 0.55 Progress ratio (PR) 114 % 60 %
Learning rate (LR) -14 % 40 %
Legend: Overview of the experience curve fitting for the biomass subset, as a function of worldwide biomass ORC capacity and
total worldwide ORC capacity (NTOT = 40; NP = 38; NM = 15). Note the very poor goodness-of-fit for both project curve fittings. The
curve fitting for the modules is low with respect to the total ORC capacity, and mediocre relative to the biomass ORC capacity.
100
1000
10000
100000
0,10 1,00 10,00 100,00 1000,00
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Cumulative installed capacity (MW)
P Biomass
M Biomass
Power (P Biomass)
Power (M Biomass)
CHAPTER FIVE
174
Thirdly, the dataset contains information about 15 geothermal ORC projects, spanning the
period from 1989 to 2014. The course of the annual average geothermal project costs is again
observed with respect to both the worldwide installed geothermal ORC capacity and the total
worldwide ORC capacity. The resulting experience curve fitting for the projects is available in
Table 5.9 and Figure 5.34 shows the experience curves for geothermal ORC projects, plotted in
function of the geothermal ORC capacity. Given the low goodness of fit for the geothermal
curve fitting, the available data does not allow to make valid conclusions about the learning of
geothermal ORC projects. In a purely indicative way, the available data suggest the investment
costs for geothermal ORC projects have increased strongly with accumulating experience.
Plotted against the total cumulative installed ORC capacity worldwide, the learning would
become less negative at a rate of -5%.
Figure 5.34. Experience curves for ORC-generated electricity: geothermal subsample.
Legend: The experience curve fits the average specific investment costs as a function of the worldwide cumulative installed
geothermal ORC capacity (2015 Euros; NTOT = 15; NP = 15; NM = 0). Project progress ratio = 144% (R² = 0.33; period: 1989-2014).
100
1000
10000
100 1000 10000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Year
P Geothermal
Power (P Geothermal)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
175
Table 5.9. Experience curve fitting for ORC-generated electricity: geothermal subsample.
Projects Modules
Experience relative to worldwide cumulative installed geothermal ORC
capacity
Curve fitting 𝑦 = 71.439𝑥0.522 -
R² 0.33 - Progress ratio (PR) 144 % -
Learning rate (LR) -44 % -
Experience relative to total worldwide cumulative installed capacity
Curve fitting 𝑦 = 139.04𝑥0.416 -
R² 0.28 - Progress ratio (PR) 133 % -
Learning rate (LR) -33 % -
Legend: Overview of the experience curve fitting for the geothermal subset, as a function of worldwide geothermal ORC capacity
and total worldwide ORC capacity (NTOT = 15; NP = 15; NM = 0). Note the poor goodness-of-fit for both project curve fittings.
Finally, the effect of accumulated experience is considered for the manufacturers separately.
Again, only the manufacturer Turboden there is sufficient data available so for the other
manufacturers it is not possible to perform the analysis. The cost database contains 53 records
of Turboden ORC systems for investigation (with data about 46 projects and 31 modules). To
obtain a valid representation of the manufacturer’s experience gain, the experience curve is
estimated using the cumulative installed ORC capacity of Turboden itself as well as with respect
to the total worldwide installed ORC capacity. Because there is no information about the
installed ORC capacity of Turboden before 1999, one record from 1984 had to be removed for
the analysis. The curve fitting and the resulting progress rates of are shown in Table 5.10. Figure
5.35 shows the experience curve with respect to the cumulative installed capacity of Turboden
itself. The data available on the Turboden projects do not allow to make conclusions about the
effect of experience of the costs of complete projects. For the ORC modules, the data suggests
that experience accumulated within the firm appears to pay off, with a PR of 91 %. The costs of
Turboden’s ORC modules have declined with 5% for every doubling of installed capacity since
launch of the company’s first projects. Where the first ORCs were individually tailor-made for
the client, Turboden has been able to drive production up and today offers several standardized
product lines. Obviously, the systems are adapted for the specific needs of the end-user and the
on-site construction and integration cannot be standardized. This explains why the learning
effect of the projects is less apparent than that of the modules. The effect of increased
production volumes and standardized product lines is visible in the module experience curve.
CHAPTER FIVE
176
Figure 5.35. Experience curves for ORC-generated electricity: Turboden subsample.
Legend: The experience curve fits the average specific investment costs as a function of the worldwide cumulative installed
Turboden ORC capacity, on a log-log scale (2015 Euros; NTOT = 52; NP = 45; NM = 30). Project progress ratio = 98% (R² = 0.02;
period: 1999-2015); module progress ratio = 91% (R² = 0.53; period: 2001-2012).
Table 5.10. Experience curve fitting for ORC-generated electricity: Turboden subsample.
Projects Modules
Experience relative to worldwide cumulative installed capacity of Turboden
Curve fitting 𝑦 = 4668.9𝑥−0.033 𝑦 = 1961.2𝑥−0.136
R² 0.02 0.53 Progress ratio (PR) 98 % 91 %
Learning rate (LR) 2 % 9 %
Experience relative to total worldwide cumulative installed capacity
Curve fitting 𝑦 = 1000.7𝑥0.2 𝑦 = 270612𝑥−0.764
R² 0.03 0.30 Progress ratio (PR) 115 % 59 %
Learning rate (LR) -15 % 41 %
Legend: Overview of the experience curve fitting for the Turboden subset, as a function of worldwide Turboden ORC capacity and
total worldwide ORC capacity (NTOT = 52; NP = 45; NM = 30). Note the very poor goodness-of-fit for the project curve fittings. The
module curve fitting has poor significance when the total ORC capacity is considered, and very mediocre with respect to
Turboden’s own capacity.
100
1000
10000
100000
0,1 1 10 100 1000
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Cumulative installed capacity (MW)
P Turboden
M Turboden
Power (P Turboden)
Power (M Turboden)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
177
5.4. Discussion of the results
5.4.1. ORC technology diffusion and the distribution of market shares
The diffusion of new technologies is typically envisioned in the form of a technology life-cycle,
where the technology goes from the initial technology adoption over a more extensive diffusion
and followed by a stage of market maturity. However, not all inventions or emerging
technologies pass through all these stages. ORC technology is closely related to the more
conventional, classic steam Rankine cycle, but the use of alternative working fluids instead of
water is only investigated more thoroughly since the 1960s. At that point, the technical
feasibility of the technology was demonstrated. Both research efforts and practical applications
were undertaken since then, albeit at a very modest scale. The real take-off of ORC technology
took place only after the start of the new millennium. Today, there are more than 610 ORCs
operational and at least another 110 sold. ORCs are commissioned in every region of the world,
but the large majority (69 %) is situated in Europe. The majority operates on biomass sources
(46 %) and on recovered heat (35 %), but the systems using geothermal heat sources are
responsible for the bulk of the total installed capacity (76 %). Biomass and heat recovery ORCs
tend to have smaller capacities (smaller than or around 1 MW), whereas geothermal systems
typically have a capacity of multiple MW. There exist ORC systems that use solar energy to
operate, but these represent a small minority.
The ORC market has historically been dominated by two companies: Ormat and Turboden.
Ormat exists since 1965 and has its foundations in an Israeli-USA research collaboration
between Harry Zvi Tabor and Lucien Bronicki. Building its first systems about half a century ago,
the company was responsible for the first commercialization of ORC technology and it continues
to install new systems today. Ormat has mostly experience with larger-scale (> 1 MW) systems,
mainly in geothermal applications but also using recovered heat. Remote heat recovery
systems, mainly for gas pipelines, were installed since the early days and other heat recovery
units since the end of the 1990s. Turboden exists since 1980 as a spin-off founded by Prof.
Mario Gaia from the Politecnico di Milano, Italy. The company installed its first commercial ORC
at the end of the 1990s and has been leading the smaller-scale ORC market since then. The
company is complementary to Ormat and has most experience with biomass applications of
about 0.5 to 1.5 MW. Turboden installed a few geothermal systems, but with generally lower
capacities than the ones built by Ormat, and in 2005 Turboden entered the heat recovery
market. Although Turboden is younger, they are leading in number of units sold: the company is
responsible for 44 % of the ORC systems worldwide. Other manufacturers followed shortly after
Turboden, but they have not yet managed to acquire a similar market share. However, since
about 2010 the smaller ORC manufacturers are growing faster. Table 5.11 displays for each of
the ORC manufacturers in the references database the suitable heat source temperature range,
CHAPTER FIVE
178
the power range of the installed references and the share of each type of heat source in the
manufacturer’s references. The younger manufacturers expand their activities and with
increased experience their potential competition for the vested companies could increase. The
third largest in terms of installed systems count (8.7 %) is Electratherm, a US-based ORC
manufacturer established in 2005. Electratherm’s ORC units have smaller capacities of 65 to 100
kW. Most applications operate on recovered heat, but there are also biomass systems, and a
few geothermal and solar ones. The company Exergy was founded in Italy in 2000 by Claudio
Spadacini, educated at the Politecnico di Milano, Italy. The company finished the design of its
low-speed radial outflow turbine in 2009 and installed 6.1 % of all existing systems today.
Exergy’s references are mainly geothermal-based or utilize recovered heat, a few apply biomass
sources and one plant uses solar energy. The applications have a wide size range, from 125 kW
for biomass systems up to 25 MW for geothermal units. The heat recovery applications range
from 100 kW to 5.5 MW. Kaishan is the largest domestic manufacturer of air compressors and
the largest manufacturer of screw air compressors in Asia. The Chinese company installed its
first ORC system in 2012 and has most of its ORC references in China. Triogen is an ORC
manufacturer founded in 2001 as a spin-off from the Technical University of Delft, the
Netherlands, and the ORC was developed in cooperation with Lappeenranta University of
Technology, Finland. The pilot installation was built in 2008, standardized after tests, and
commercialized as of 2009. The Triogen ORC requires a heat source of minimum 350°C and
produces 165 kWe gross from 900 kW of thermal input. In the Triogen ORC systems the working
fluid toluene is heated directly by the heat source, without the requirement of intermediate
heat transfer equipment. The ORC system is built on a single shaft, including a high-speed
turbine, electric generator and a pump. The majority of the Triogen references operate on
excess heat, only a few run on biomass. The manufacturer Adoratec was founded in 2004 in
Germany, has built its first plants in 2006 and is now part of Maxxtec. The company has only
biomass references, with sizes in the 250 kW to 3 MW range. The references installed under the
name Maxxtec (about 100) use Turboden units and all have biomass inputs. BEP Europe is a
Belgium-based firm specialized in heat recovery ORC systems. Their E-RATIONAL single screw
radial inflow ORC machine is suitable for heat sources with temperatures in the 80 – 150°C
range. THE ORC-1000 model generates 55 kWe to 132 kWe from 1000 kWth heat input and the
ORC-4000 model obtains 250 to 500 kWe from 4000 kWth, depending on the operation
conditions. The first BEP systems were operational in 2011. Enogia was established in 2009 and
manufactures small scale ORC systems. The ORCs operate with an in-house designed micro-
turboexpander and a high-speed generator. The company is mainly involved in heat recovery
ORCs, but has also several solar systems operational and under construction.
Entry into the ORC market is not easy because of technological intricacy. Several of the ORC
manufacturers have emerged as a spin-off from universities or in close cooperation with
research institutes. Developing an ORC prototype can take several years and even longer before
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
179
the first unit is sold. This encourages concentration on particular applications. New
manufacturers entering the ORC market can attempt to distinguish themselves by proposing a
new type of cycle architecture or by using different working fluids, aiming to be more efficient in
terms of either costs or production, or by aiming for a temperature or capacity range that is not
yet served. In summary, Table 5.11 suggests that the ORC manufacturer market is segmented,
with each of the manufacturers specialized in a different type of heat source and different scale
range. There are only a limited number of ORC market players, most of them dedicated to the
ORC market, and within each market segment there is limited competition.
Table 5.11. Suitable temperature range of the heat source, capacity range of installed systems and percentage of applications in each heat source category for each manufacturer in the database.
Heat source temperature [°C]
Power range [kW]
Heat sources [%]
GEO HR BIO SOL
Turboden 100 – 300 100 – 16,500 3.1 11.0 84.7 1.2 Ormat 150 – 300 1 – 330,000 57.4 40.9 - 1.7 Electratherm > 93 35 – 100 7.8 71.9 17.2 3.1 Exergy NA 100 – 25,000 46.7 40.0 11.1 2.2 Kaishan 80 – NA 100 – 7,445 10.7 89.3 - - Triogen > 350 95 – 170 - 76.5 23.5 - Adoratec/Maxxtec 300 270 – 3,100 - - 100 - BEP Europe 80 – 150 11 – 740 9.1 90.9 - - Enogia 80 – 400 5 – 100 - 64.3 - 35.7
Factors influencing the diffusion of ORC technology
The technical appeal of ORC technology is encompassed in its very nature. There are
tremendous amounts of heat sources that can be converted into work using ORC technology
and the technology is relatively compact and well-suited for use in remote areas or as add-on
for existing facilities. The very first ORC systems were developed with the goal in mind to
provide accessible energy in off-grid areas using solar-ORC solutions. But the actual diffusion of
a technology depends on more than technical appeal only, or as stated by Rosenberg (1996a, p.
191): “the diffusion of inventions is an essentially economic phenomenon, the timing of which
can be largely explained by expected profits”. The diffusion of new techniques is commonly
understood in light of existing solutions, i.e. their superiority with respect to the old technique
and the rate at which they replace it (Rosenberg, 1996a). ORC technology does not necessarily
replace an existing technology, but rather it opens the market for low-grade heat-to-power
conversion, which is until today underexploited. Nevertheless, the diffusion of ORC technology
was launched only relatively recently. Several factors can be at the origin of this trend.
First, an ORC principally generates electricity and exhausts low-grade heat. The return of
investing in an ORC system is forthcoming from the sales of electricity (and potentially low-
CHAPTER FIVE
180
grade heat) or from the savings on the own commodity bills when the production is used in-
house. Hence, the economic appeal of ORC technology relies on the energy prices prevailing in
the market. Figure 5.36 shows the industrial electricity price for countries in Europe. Electricity
is a more locally traded commodity, which implies that electricity prices are influenced by
primary fuel prices, but also by other factors such as the cost of carbon dioxide emission
certificates (Eurostat, 2016) or country-specific circumstances such as taxes, the local energy
supply mix, local market conditions, etc. There is a large variability in the prices for different
countries, but an overall increase in electricity prices since the new millennium. This can explain
the increasing interest in ORC technology. The highest electricity prices are paid in Cyprus,
Malta, Italy, Germany, but also for instance Liechtenstein, the UK and Ireland have seen prices
increase over the last decade (Figure 5.36). Germany and Italy are among the countries with the
highest electricity prices, and both have more than 25% of the European ORCs installed. Austria
had one of the highest electricity prices in Europe in the 1990s and early 2000s, but since 2009
the Austrian electricity price declines, against the general trends. Austria was one of the
forerunners and had many of the first European ORCs installed, but since 2011 no new ORC
projects were undertaken. The Czech Republic has more than 5 % of European ORCs. Electricity
prices were historically among the lowest in Europe, have rocketed to above average in the
period 2008-2011, and declined again since then to one of the lowest levels in Europe. The
majority of the ORCs were installed in the period 2011-2014. The electricity prices in France
have always been and continue to be relatively low, but still there has been a steady ORC
deployment since 2011. The UK had low to moderate electricity prices, saw a sharp increase
after 2003 and has peaked above average since 2013. Interest for ORC systems launched only
really after 2011, but today the UK ORC market is booming with several systems installed in
2015 and a multiple thereof under construction. These findings suggest a link between ORC
deployment and local electricity prices.
Nevertheless, the electricity price by itself will not be the only explanatory factor. For instance,
islands like Cyprus and Malta have very high electricity prices, but the ORC market there is
(nearly) unexploited. Both islands are poor in traditional energy reserves, but have high solar
potential. Other countries, such as France, have fairly low electricity prices but still saw an
increase in ORC deployment. It seems that investors react more on the increase in prices, than
on the height of the price level. Other factors influencing the deployment of ORC technology in
a country may be the availability of suitable heat sources, the possibility to utilize the ORC for
cogeneration, awareness of the existence of ORC technology as a solution, but also local stimuli
for energy efficiency and renewable energy deployment. Awareness concerning the state of the
environment and the impact of human activities on energy resources are topics raised since the
1970s. The oil crises induced significantly higher energy prices, the Club of Rome raised public
attention for environmental issues with the publication of ‘Limits to Growth’ (Meadows,
Meadows, Randers, & Behrens, 1972) and at the institutional level the European Union (then
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
181
European Community) decided to intensify cooperation on energy issues (see the Council
Resolution concerning a new energy policy strategy for the Community (Council of the European
Communities, 1974)). The 1980s were characterized by economic recessions and the interest for
energy and environmental matters was pushed back on the agenda. The Intergovernmental
Panel on Climate Change (IPCC) published its first report in 1988, thereby drawing attention
towards climate issues, the Earth Summit in Rio de Janeiro in 1992 lead to the adoption of the
United Nations Framework Convention on Climate Change (UNFCCC), and in 1997 the Kyoto
Protocol was agreed upon. The European Union adopted legislation regarding the internal
market for electricity (European Parliament & Council of the European Union, 1996) and gas
(European Parliament & Council of the European Union, 1998). But, focussing on the European
situation, the concrete trend towards more integrated action was initiated in 2007 with the first
EU energy action plan (Commission of the European Communities, 2007). Today, energy policy
at the EU level encapsulates many domains, with the 20-20-20 targets and legislation on, among
others, renewable energy (European Parliament & Council of the European Union, 2009) and
the promotion of energy efficiency (European Parliament & Council of the European Union,
2012). Member states of the EU are bound to implement EU legislation, but are free to decide
which strategies, mechanisms or incentives will be used to achieve the goals. This causes a wide
variation in local support for energy technologies. For instance, Germany has been at the
forefront of innovative energy policy with its 1991 Electricity Feed-in Act and its successor the
2000 Renewable Energy Act (Erneuerbare-Energien-Gesetz (EEG)), encouraging power
generation from renewable energy sources. Italy utilizes a white certificate scheme for energy
efficiency enhancement since 2005. A green certificate scheme was in effect but is now being
replaced by a system with feed-in tariffs, premiums and other incentives. The UK supports the
uptake of renewable heat technologies via the 2011 Renewable Heat Incentive (RHI), which
explains the large contemporary interest and sales of ORC systems in the UK. Hence, the
country-specific uptake of renewable energy and efficiency technologies depends also on the
local policy framework. The overall deployment pattern observed for ORC technology, slowly at
first but rapidly increasing in the new millennium, has also been observed for other renewable
energy categories, such as onshore wind energy, offshore wind energy and photovoltaic solar
energy (Junginger, van Sark, & Faaij, 2010). Many of these technologies were technically proves
several decades ago, but were only applied to a limited extent until the awareness for
alternative energy systems became more prominent.
CHAPTER FIVE
182
Figure 5.36. Electricity prices in European countries for industrial consumers.
Legend: Industrial consumers with electricity use between 500 and 2000 MWh (excluding VAT and other recoverable taxes and levies). Percentages between brackets indicate the
share of ORC systems installed in the country compared to all ORC systems installed in Europe. Source: Eurostat (2013, 2016).
0
0,05
0,1
0,15
0,2
0,25In
du
stri
al e
lect
rici
ty p
rice
[Eu
ro/k
Wh
]
Time
Bosnia and HerzegovinaIrelandKosovoLiechtensteinLuxembourgLithuaniaMacedonia, the FYR ofMaltaMoldovaMontenegroSerbiaSlovenia (0.2 %)Norway (0.2 %)Cyprus (0.2 %)Hungary (0.4 %)Estonia (0.4 %)Denmark (0.4 %)Bulgaria (0.4 %)Greece (0.6 %)Sweden (0.8 %)Finland (0.8 %)Portugal (1 %)Romania (1.4 %)Netherlands (1,6 %)Spain (1.8 %)Belgium (2 %)Poland (2.2 %)Croatia (2.2 %)Slovakia (2.9 %)Latvia (3.1 %)United Kingdom (4.7 %)France (4.9 %)Czech Republic (5.5 %)Austria (6.9 %)Germany (25.5 %)Italy (25.9 %)Euro areaEuropean Union
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
183
5.4.2. Scale economies for ORC technology
Economies of scale refer to the situation where an increase in size leads to a less proportionate
increase in the costs of the product. For energy technologies, the size of the system is usually
understood as its installed capacity. The result of upscaling for ORC technology is measured for
the complete dataset, as well as for multiple representative subsets. The numerical results of
the analysis are summarized in Table 5.12. The estimation of the general scale effect, i.e.
demonstrated for the complete dataset, shows and overall beneficially influence of scaling.
Subdividing the database according to the thermal energy input used for the operation yields
most significant results for the biomass and heat recovery subsamples. The biomass samples
show consistent scale factors around 0.43-0.5. The scale factors of the heat recovery ORCs show
more variability, between 0.46 and 0.89 for the modules and up to 0.94 for the heat recovery
projects. The geothermal ORC systems are typically large in scale, but reveal limited cost
advantages within their size range. For the smaller-scale geothermal ORC projects there is more
variation in the costs and an overall decline, but once the systems pass the 10 MW threshold
the cost effect of increasing scale (up to 40 MW) is relatively limited. Overall, the quality of the
curve fitting is too low to draw results from the data.
Finally, the economies of scale are investigated within the production frame of a specific
manufacturer, Turboden. The results are somewhat more ambiguous than for the heat source
categories, with several subsamples with a too low goodness-of-fit. The statistically significant
results for the Turboden projects show scale factors consistently between 0.43 and 0.49. The
scale factors for Turboden’s modules are less consistent, but the most significant results include
a scale factor of 0.5 for Turboden’s biomass systems and a factor of 0.74 for Turboden’s heat
recovery factors. Note that the 2002 project sample, the 2008 biomass module sample and the
2009 biomass project sample contains only price quotes provided by Turboden. Each of them
has a scale factor of approximately 0.45, suggesting that this is the internal scale factor applied
by the company itself.
CHAPTER FIVE
184
Table 5.12. Overview of economies of scale of ORC projects and modules, estimated for different categories.
Project Module Notes
Data sample
Scale factor
R² Scale factor
R²
Complete dataset
All data 0.82 0.77 0.75 0.90 -
2002 0.48 0.99 - - -
2006 0.55 0.90 - - -
2007 - - 0.89 1 -
2008 0.64 0.59 0.73 0.74 -
2009 0.68 0.63 0.44 0.89 -
2010 1 0.96 - - -
2011 0.94 0.94 0.79 0.95 -
2012 0.87 0.83 0.74 0.86 -
Heat recovery
All data 0.85 0.74 0.78 0.94 -
2007 - - 0.89 1 (1)
2009 0.32 0.20 0.46 0.92 -
2011 - - 0.80 0.95 -
2012 0.83 0.78 0.76 0.97 -
2014 0.94 0.91 - - -
Biomass All data 0.36 0.24 0.44 0.37 - 2002 0.48 0.99 - - (2) 2008 0.43 0.91 0.51 0.93 (3) 2009 0.49 1 - - (2)
Geothermal All data 0.62 0.10 - - -
Turboden All data 0.59 0.49 0.76 0.81 - 2002 0.48 0.99 - - (2) 2008 0.45 0.92 0.73 0.74 - 2008 BIO 0.43 0.91 0.51 0.93 (4)
2009 0.19 0.24 - - - 2009 BIO 0.49 1 - - (5)
2009 HR 0.12 0.16 - - (6)
2012 0.84 0.65 0.89 0.81 - 2012 HR 0.97 0.84 0.74 0.92 (7)
Legend: Summary of estimated scale factors for different subsamples. Scale factor estimates with a good statistical significance
(R² > 0.9) are displayed in bold. Notes: (1)
The module sample contains only price quotes from the manufacturer Electratherm. (2)
The project sample contains only biomass price quotes from the manufacturer Turboden. (3)
The module sample contains only
price quotes from Turboden; the project sample contains the same price quotes from Turboden and one real case from Turboden. (4)
A subset of the 2008 Turboden sample, which contains only those using biomass. The module sample contains only price
quotes. (5)
A subset of the 2009 Turboden sample, which contains only those using biomass. The project sample contains only
price quotes. (6)
A subset of the 2009 Turboden sample, which contains only those using recovered heat. (7)
A subset of the 2012
Turboden sample, which contains only those using recovered heat.
Upscaling of the installed capacity has been a major driver for cost reductions for other
technologies. In the case of wind power for instance, the positive experience effect is to a large
extent the result of a progressive increase in turbine size over the years (Neij, 1999). Figure 5.37
shows the average capacity of the ORC systems installed in each year. The figure shows large to
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
185
very large average capacities installed before the year 2000, and only low ones thereafter. Many
references before 2000 were installed by the company Ormat, specialized in large-scale ORC
systems. To see whether this trend of upscaling also exists for ORC manufacturing, Figure 5.38
shows the trend of the average size of all installed ORC systems and the average size of ORC
systems installed by the company Turboden, for the period after 2001. The variability in the
average capacity of all ORCs is higher than of Turboden ORCs. Turboden’s systems show a
moderate size increase over time, suggesting the company is exploring the benefits of upscaling.
Moreover, Turboden’s ORC deployment trajectory not only involves upscaling, but also
advances with returns to scale. The company is one of the early firms to commercialize ORC
technology successfully and acquire an extensive share of the market. This is evidently the result
of many factors, including company-internal factors which are entrepreneurial, managerial and
financial in nature, but also the fact that Turboden relatively quickly offered standardized ORC
modules. Today, the company offers a standard range of ORC modules for different
applications.
Figure 5.37. Average capacity [MW] of ORC systems installed.
Figure 5.38. Average capacity [MW] ORC systems installed after 2000, for all systems and for Turboden.
0
20
40
60
80
100
120
140
Average Capacity [MW]
0
1
2
3
4
5
6
Average Capacity [MW]
Average Turboden Capacity [MW]
CHAPTER FIVE
186
5.4.3. On learning-by-doing in the ORC market
The relation between accumulating experience and the average costs of a technology is
conceptualized in the experience curve. This study investigated whether the increased
application of ORC technology results in corresponding price changes. A summary of the
resulting progress rates is provided in Table 5.13. For most of the subcategories considered, the
dataset does not allow to draw meaningful conclusions about the impact of experience.
Particularly for the ORC projects the quality of the curve fitting is insufficient. There are strong
differences in the experience effects for the ORC module subcategories, but the results are
significant only for the biomass systems and the ORCs installed by the manufacturer Turboden.
The biomass ORC modules comprise the only submarket with indisputable positive learning
effects. The biomass ORCs have encountered an average reduction in module costs of 7 % for
each doubling of the installed biomass ORC capacity. Compared to the total worldwide installed
ORC capacity, this number increases drastically to 40 %. In other words, the biomass ORCs have
gained cost advantages from accumulating experience, especially in comparison with other ORC
systems. For Turboden, the experience curve of the modules suggests a cost decrease of 9 % for
each doubling of the company’s installed capacity (although with a relatively low R²). The
similarity between the experience curves of the biomass and the Turboden ORC modules is not
surprising. Nearly half of the ORC systems installed worldwide use biomass as heat source, 82%
of the biomass ORCs is installed by Turboden and 85% of Turboden’s ORCs use biomass.
Turboden is one of the most experienced manufacturers, that managed to reduce the costs of
its systems with accumulating experience. Hence, the effects of the company’s experience are
reflected in the progress of biomass ORCs.
Table 5.13. Progress rates for ORC projects and modules, estimated for different categories.
Experience relative to total cumulative installed capacity in subcategory
Experience relative to total worldwide cumulative installed capacity
Projects R² Modules R² Projects R² Modules R²
Complete dataset
113 % 0.12 102 % 0.01 113 % 0.12 102 % 0.01
Heat recovery 106 % 0.06 101 % 0.00 108 % 0.05 102 % 0.01
Biomass 100 % 0.00 93 % 0.69 114 % 0.02 60 % 0.55
Geothermal 144 % 0.33 - - 133 % 0.28 - -
Turboden 98 % 0.02 91 % 0.53 115 % 0.03 59 % 0.30 Legend: Summary of the experience curve fittings, for different subsamples. Estimates with an R² above 0.5 are highlighted in
bold.
The experience curve of ORC technology has not been estimated before, so the results of this
study cannot be compared to previous work. Most experience curve research until today has
focussed on modular technologies such as photovoltaic and wind energy. Searching the
literature for experience analyses on geothermal power generation yielded no results, so the
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
187
findings from this study cannot be put in perspective to that of other works. Similarly, excess
heat recovery with ORC technology is a niche application, which has not been investigated in
terms of experience before. Power generation systems that operate on bioenergy differ from
other renewable power technologies by the fact that they require fuel input. Estimating reliable
experience curves for biomass plants has proven to be difficult due to the wide variation in
conversion technologies, plant layouts, and feedstock costs (M. Junginger, Faaij, Björheden, &
Turkenburg, 2005; Martin Junginger, van Sark, & Faaij, 2010). Therefore, bioenergy systems are
generally considered as compound systems with three subcomponents, each of them subject to
a different experience system: the investment costs, the O&M costs and the fuel costs (M.
Junginger et al., 2005). The focus in this work is on the experience measured in the investment
costs; the other two learning systems cannot be investigated based on the data available. The
effect of experience on the investment costs of bioenergy systems has been studied by, for
instance, Martin Junginger et al. (2006) for biomass CHP plants in Sweden. The PR was
estimated between 75 % and 91 %, using different approaches, but the statistical significance of
the fits was too low (Martin Junginger et al., 2006). The progress ratio for Danish biogas plants
was estimated at 88 % (R² = 0.69) (Martin Junginger et al., 2006). Koornneef, Junginger, and
Faaij (2007) found progress ratios between 42 % and 93 % for fluidized bed combustion plants.
Berghout (2008) studied rapeseed for biodiesel production in Germany but found no statistically
significant progress rates for the investment costs. Hence, the statistically most significant
progress ratio estimated in the current study (93 % for biomass ORC projects) is in line with the
findings from the literature, based on the limited number of analyses for bioenergy. For
comparison, more classic power generation technologies such as coal-fired plants have progress
ratios estimated at 92.8 % between 1957 and 1976 (Ostwald & Reisdorf, 1979), at 92.4 % for the
period 1975-1993 (Kouvaritakis, Soria, & Isoard, 2000; McDonald & Schrattenholzer, 2001) and
at 88 % between 1902 and 2006 (McNerney, Doyne Farmer, & Trancik, 2011). Classic natural-gas
fired plants have estimated progress ratios in the range of 78 % to 90 % (MacGregor, Maslak, &
Stoll, 1991; Rubin, Azevedo, Jaramillo, & Yeh, 2015) and estimated at 88.8 % for the period
1957-1968 (Ostwald & Reisdorf, 1979).
Recall that learning-by-doing is just one driver of cost reductions in technology development.
Other mechanisms of influence include learning-by searching, i.e. research, development and
demonstration activities, learning-by-using, involving feedback from end-users, learning-by-
interacting, stemming from network interactions in the diffusion phase, upsizing of the
technology, or economies of scale, including standardization and large-scale production (Martin
Junginger et al., 2006). These mechanisms occur in varying proportions throughout the different
phases of technology deployment and it is often not easy to isolate and study the single effects
separately. Hence, the experience curves estimated in section 5.3.3 likely include multiple cost
drivers. One approach to make abstraction of the effects of upscaling in an experience curve
analysis is to apply a scale factor and convert the systems under investigation to a reference size
CHAPTER FIVE
188
(Martin Junginger et al., 2006; Martin Junginger et al., 2010). In this case, the variation in
specific costs for different capacities is adjusted by using the scale function previously presented
in equation (5.1). When the reference size equals unity, the scaling function becomes:
cadj
(Padj)𝑆𝐹 = cref
(5.7)
with cref the production costs of the reference system; cadj the production costs of the system for which the scale factor will be applied;
Padj the size of the system for which the scale factor will be applied;
SF the scale factor.
Hence, to exclude the impact of upscaling from the experience curves estimated in this chapter,
the analysis is repeated for scale-adjusted data. For each of the subcategories, the data has
been rescaled using equation (5.7) and the scale factors fitted in section 5.3.2. The results are
presented in Table 5.14. The scale transformation improves the quality of the curve fitting for
most of the estimates, but the majority remains statistically insignificant. The PR of the biomass
modules increased with one per cent to 94 % and has an R² of approximately 0.9, when the
installed capacity of the biomass ORC systems is taken as reference. Compared to the total
worldwide installed ORC capacity, the quality of the biomass ORC module curve fitting improves
and suggests a PR of 66 % instead of the previously estimated 60 %. Finally, the quality of the
curve fittings for the Turboden projects improved drastically and results in a PR of 94 % (instead
of the 91 % estimated before). The curve fitting for Turboden’s modules improves, in two cases,
to an acceptable R² of 0.7 and progress ratios of approximately 66-67 %.
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
189
Table 5.14. Progress rates for the adjusted ORC projects and modules, for different categories.
Experience relative to total cumulative installed capacity in subcategory
Experience relative to total worldwide cumulative installed capacity
Projects Modules Projects Modules
SF PR [%]
R² SF PR [%]
R² SF PR [%]
R² SF PR [%]
R²
Complete dataset
0.82 117 0.28 0.75 107 0.30 0.82 117 0.28 0.75 107 0.30
Heat recovery
0.94 107 0.10 0.89 104 0.14 0.94 110 0.09 0.89 106 0.15
- - - 0.76 106 0.24 - - - 0.76 109 0.24
- - - 0.81 106 0.22 - - - 0.81 106 0.22
0.46 112 0.24 0.46 116 0.22
Biomass 0.48 103 0.05 0.51 94 0.89 0.48 130 0.16 0.51 66 0.70
Geoth. - - - - - - - - - - - -
Turbod. 0.46 103 0.05 0.76 94 0.94 0.46 143 0.26 0.76 66 0.69
- - - 0.51 95 0.60 - - - 0.50 75 0.35
- - - 0.74 94 0.92 - - - 0.74 67 0.66 Legend: Summary of the experience curve fittings for different subsamples, adjusted to minimize the influence of scale effects.
Estimates with an R² above 0.5 are highlighted in bold.
Factors influencing the underlying cost dynamics
Although most of the estimated experience curves have insufficient explanatory power, a closer
look at the data could provide insight in the underlying cost dynamics. For instance, many of the
experience curves estimated display a decrease in average SIC at first, followed by an upsurge in
the most recent years. One explanation for this pattern could be the experience of the
individual manufacturers, since the upward-driving forces originate from ORC systems sold by
less experienced manufacturers. This hypothesis is strengthened by the beneficial experience
curve estimated for the incumbent manufacturer Turboden. For Ormat we would likewise
expect a positive learning effect because of their decades-long experience with ORC systems,
but there was not enough data available (only 7 records) to perform a decent investigation for
this manufacturer. In other words, the experience curves suggest that the ORC market is not
stable yet, but in turbulent development. New entrants have not yet been able to benefit from
experience, or they introduce new ORC designs, so that prices charged by younger
manufacturers are commonly higher than those from more established manufacturers. Also, the
costs of a product in an early phase of market diffusion are often even higher than the costs of
demonstration projects (Martin Junginger et al., 2010). The relation between the price and the
costs of a new product typically undergoes four stages (see Figure 5.39). At first, the price is
lower than the production costs to create a market for the product. Then, the costs decrease
but the price declines at a slower rate. In this so-called umbrella phase, the dominant
manufacturer can determine the price in the market and new manufacturers are attracted by
the increasing profit margins. In the shakeout phase, prices drop faster than costs and in the
final, stable phase there is a constant margin between costs and prices. (Martin Junginger et al.,
CHAPTER FIVE
190
2010; OECD/IEA, 2000) Since the prices charged by the incumbent manufacturer drop much
stronger than those of the other manufacturers (cfr. Figure 5.40) it is possible that the ORC
manufacturer market, which is already quite oligopolistic, is currently in a shakeout phase.
Figure 5.39. Relation between the price and costs of a new product.
Legend: The price-cost relation for a new product typically goes in four stages. Source: (OECD/IEA, 2000).
Figure 5.40. Experience curves for ORC-generated electricity: Turboden subsample compared to the rest of the data.
Legend: The experience curves for Turboden and for the complete dataset, compared to the total worldwide installed ORC
capacity. Turboden project progress ratio = 115% (R² = 0.03; period: 1999-2015); module progress ratio = 59% (R² = 0.3; period:
2001-2012). Complete dataset project progress ratio = 113% (R² = 0.12; period: 1984-2016);
module progress ratio = 102% (R² = 0; period: 1984-2016).
A second explanation for the recently increasing ORC costs is the similarity with increases in
market prices for renewable energy technologies. For instance, the costs of wind generated
have dropped in the past, but saw strong increases in the period 2005-2008 (Martin Junginger
100
1000
10000
100000
100,00 1000,00 10000,00
Spe
cifi
c in
vest
me
nt
cost
s [€
20
15
/kW
]
Cumulative installed capacity (MW)
P Turboden
M Turboden
P Complete dataset
M Complete dataset
Power (P Turboden)
Power (M Turboden)
Power (P Completedataset)Power (M Completedataset)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
191
et al., 2010). In the case of wind power, this can be attributed to increasing raw material costs –
including steel, copper, lead, cement and carbon fibre – in the period 2004-2008 and a
significant growth of wind turbine demand (Blanco, 2009). Similarly, the global market prices for
PV modules have increased in the period 2003-2008, despite the decrease expected from the
experience curve. Again, these price increases were annotated to a strongly increasing demand,
leading to production shortage. But also scarcity in purified silicon was at the basis of the
observed price increases, as well as by certain manufacturers using the silicon scarcity as excuse
for additional price growths, to increase their profit margins. (Martin Junginger et al., 2010) Also
the capital and operating costs of new coal plants increased significantly in the period 2000-
2007, also partially due to augmenting material prices (Hamilton, Herzog, & Parsons, 2009;
Martin Junginger et al., 2010). A shortage in ORC supply due to a rapidly increasing demand is
probably not the reason for ORC price increases, but rising material prices are likely to have an
influence. Moreover, recall that the ORC manufacturer market is oligopolistic, so that the
manufacturers can set their prices more freely than in a market with full competition.
Finally, the low statistical significance of the experience curve fittings can be the result of a wide
variability in ORC design. Similarly as in the case of biomass power plants (Martin Junginger et
al., 2006), the layout of ORC systems often depends on local conditions and site-specific
requirements. Hence, the integration and other project costs are very specific to the projects at
hand and nearly impossible to standardize. The module costs can be standardized (cfr. the case
of Turboden) but the extent of learning in this case depends, among other things, on the
learning potential of the subcomponents. Comparing again with wind power, wind turbines are
for a large part composed of components that have been developed for use in other
applications. This implies the costs of these components may have declined previously. (Neij,
1999) This is similar for ORC technology. Components like heat exchangers and pumps have
been developed and used extensively so they can be considered mature. The novelty of ORC
technology lies in exploring appropriate working fluids, designing correspondingly efficient
expanders and establishing a working cycle. Such conditions have two potential implications.
Either there is a positive experience gain for the new component, but it is not reflected in the
total system’s costs since it constitutes only a minor part of the whole. This observation holds
for instance for wind turbines (Neij, 1999), but also for advanced thermodynamic cycles
(OECD/IEA, 2000). The technological advance in the novelty may be substantial, but as most
other components are already standardized this implies that the experience effect of such
technologies will be less strong than for technologies like photovoltaics with more technology-
specific development. Another explanation is that cost increases induced by design
modifications are not quickly enough offset by cost reductions from the accumulation of
experience, from product standardization, upscaling, etc. (Neij, 1999). A well-known example
for this phenomenon is the case of nuclear power. Even in favourable deployment conditions,
the technology has proven to be so inherently unwieldy that the real costs continued to escalate
CHAPTER FIVE
192
substantially (Grubler, 2010). Also for ORC technology there may be a case of increasing costs
due to design changes. Although an ORC is conceptually straightforward, the practical
realization of a certain prototype is preceded by intensive R&D activities. Also, there are still
plenty of architecture design modifications, specialized expander designs or new working fluids
to be explored in the quest for improved cycle efficiency. However, there is no data available to
analyse the experience effects of the individual subcomponents for ORC technology.
Remarks on the experience curves for ORC-generated electricity
Because the experience curve is influenced by plenty of parameters, there may be fluctuations
in the short term. Therefore an experience curve should span sufficient doublings in production
to extricate the actual underlying pattern. (Neij et al., 2003) Although the installed ORC capacity
has seen several doublings over the observed time span, the ORC market and the technology
itself are still in turbulent development. The observed learning rates vary considerably according
to the application under consideration and the available amount of data was generally not
sufficient to make robust conclusions. Also, progress ratios do not necessarily remain constant,
but can change over time (Martin Junginger et al., 2010). Therefore, it should be interesting to
monitor the ORC market closely in the coming years and repeat the experience curve analysis as
more data becomes available.
Moreover, the experience curves in this chapter have been estimated using ORC prices instead
of production costs. Cost experience curves and price experience curves are related, but the
latter also includes manufacturer strategies, bargaining power of the investor and the
implications of public policy on prices. Hence, an experience curve estimated using prices entails
a different interpretation than one based on production costs. Cost experience curves are useful
to investigate changes in the cost structure of the technology itself, whereas price experience
curves serve to identify changes in the market. These are different things and a structural
change in the ORC market does not necessarily imply a structural change in the costs of the
technology. For instance, a kink in the price curve may reflect a shakeout phase of price
competitors from the market but this has little relation to the cost evolution of the technology
itself. (OECD/IEA, 2000) In stable markets, the difference between the production costs of a
technology and the price charged tends to be stable as well, meaning that the curves would
follow the same pattern (cfr. Figure 5.39). The progress ratio of the price is then similar to that
of the costs. (OECD/IEA, 2000) However, the ORC manufacturer market is oligopolistic and still
in full development. This means that the observed price experience effects cannot simply be
interpreted as evolutions in the technology’s costs (or vice versa). When the goal is to assess the
impact of experience in production, the experience curves estimated for Turboden are more
representative. By considering just one incumbent manufacturer, the impact of less experienced
manufacturers with higher costs is eliminated and the measured learning is more representative
for the actual increased production experience. (Neij, 1999)
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
193
Finally, experience curves are often investigated with the goal to forecast technology costs, or
to determine the learning investments required for the technology’s deployment (Schaeffer et
al., 2004). Forecasting the price of ORC technology requires reasonable assumptions about the
development of the market and the future progress ratio. With the currently available data, this
is a very precarious exercise. If the experience curve could be used to predict cost development,
this would be a useful tool for policy makers. Particularly for emerging, sustainable technologies
such an instrument is useful, because it allows to evaluate the investments required, i.e. the
learning investments, to make the technology a valid competitor for conventional technologies.
This question has been addressed previously for, for instance, solar PV technology (Schaeffer et
al. (2004); van der Zwaan and Rabl (2003)). Making a quantification of the necessary learning
investments for ORC technology requires a better insight in the technology’s cost dynamics than
available today. Nevertheless, more general insights in policy support for emerging technologies
include that policy measures, often specifically adapted for the local circumstances, cannot
simply be transposed to other countries (Schaeffer et al., 2004). Moreover, most research
suggests that a combination of approaches is key for successful learning. Both measures to
stimulate technology development (technology push) and investments in learning should be
accompanied by others to support market penetration (market pull) and learning investments
(Martin Junginger et al., 2010; Schaeffer et al., 2004).
5.5. Chapter conclusions
This chapter investigated the technological innovation path of ORC technology. Technological
innovation entails many facets, from research and development to end-user feedback. The goal
of this chapter was to obtain insight in the diffusion of ORC technology and to assess how this
development has influenced its costs. To answer this question, the focus lies on three,
quantifiable aspects: the diffusion of ORC technology, the relation between the scale of the
system and its price and the impact of accumulating experience on the price of ORC technology.
The investigation was supported empirically with information on academic research output, on
the structure of the ORC market and on market prices for ORC system. Two extensive databases
have been composed for this research. The first concerns a nearly exhaustive, global coverage
of the currently contracted ORC systems. The second includes information on the costs of ORC
systems from various manufacturers, sizes, using different heat sources and over multiple years.
By knowledge of the author, this amount of empirical data collection is unprecedented for ORC
technology.
The most important findings from this study are the following. Although ORC technology was
technically proven in the 1960s, the diffusion of ORC technology launched only truly after the
start of the new millennium. This pattern is similar to that of other renewable energy
technologies. This augmented interest in renewable energy technologies, including ORC
CHAPTER FIVE
194
technology, can be linked to sharply grown electricity prices and an increased awareness of
energy issues and its reflection in energy policies, particularly in Europe. Today, ORC systems
exist in every region worldwide, but the majority is operational in Europe. Furthermore, the
most-used heat source is biomass and the market is rather oligopolistic. Two companies, Ormat
and Turboden, dominate the market in terms of total installed capacity and amount of
references operational. Entrance into the market is not easy because the intensive R&D
involved with developing a working prototype and the high costs associated with producing the
first units. Yet, new manufacturers are entering the market and gain an increasing share of it.
The development of the costs of ORC technology is strongly influenced by economies of scale.
The impact of upscaling is measured for the complete dataset as well as for the dataset
subdivided according to the thermal energy source and the manufacturer who constructed the
ORC system. The measured scale factors are the lowest for the biomass ORC systems
(constructed by Turboden), which implies that upscaling of these systems corresponds to a less
proportionate increase in costs compared to the other ORC applications. The existing biomass
ORC systems are for the large majority installed by the manufacturer Turboden and, the other
way around, the majority of Turboden’s ORC systems operate on biomass. This explains the
close correspondence between these two subgroups.
Finally, the impact of the accumulating experience with ORC technology was investigated by
exploring the relation between the worldwide cumulative installed capacity and the
development of the average market price. Overall, the ORC projects show a negative impact of
experience and the ORC modules a neutral to positive impact, but the statistical significance of
the analysis in general is too low to draw valid conclusions. The best results are obtained for the
biomass ORC projects and modules and the Turboden projects. Eliminating the impact of scale
economies by rescaling the data improves the quality of these estimates further, and suggests a
lightly positive effect of experience for biomass and Turboden ORC modules. The learning rate
of 6 % for both categories signifies that the price of these systems have decreased with 6 %, for
each doubling of installed capacity in the subgroup. Compared to the worldwide installed ORC
capacity, the costs of biomass ORC modules have likely been reduced at a rate of 34 %. In other
words, most learning has occurred within the company Turboden and for ORC systems that
operate on biomass.
Moreover, the data underlying the general experience curves suggest that recent increases in
average technology prices are the result of new, less experienced manufacturers entering the
market. Recall that gaining experience with a technology is inherently interwoven with its
application. There is no learning-by-doing without doing. ORC modules are composed of many
well-established components. The innovation lies primarily in the development of new cycle
designs and technology-specific components such as expanders, a cost-intensive activity. The
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
195
empirical data suggests that potential for future cost reductions lies in the upscaling of system
sizes and in offering standardized product lines to benefit from returns to scale in production. It
is not possible to say how much each of these, and other, factors will influence the future cost
evolution of ORC technology. External factors of relevance include sufficiently high electricity
prices and supportive policy measures to encourage the technology’s development and market
diffusion.
CHAPTER FIVE
196
Bibliography
Arrow, K. J. (1962). The Economic Implications of Learning by Doing. The Review of Economic Studies, 29(3), 155-173.
Berglund, C., & Söderholm, P. (2006). Modeling technical change in energy system analysis: analyzing the introduction of learning-by-doing in bottom-up energy models. Energy Policy, 34(12), 1344-1356. doi:10.1016/j.enpol.2004.09.002
Berghout, N. A. (2008). Technological Learning in the German Biodiesel Industry: An Experience Curve Approach to Quantify Reductions in Prodcution Costs, Energy Use and Greenhouse Gas Emissions. (Master Thesis), Utrecht University, Copernicus Institute, Utrecht, The Netherlands. (NWS-S-2008-12)
Besanko, D. A., & Braeutigam, R. R. (2005). Microeconomics (Second ed.). Hoboken, New Jersey, USA: John Wiley & Sons, Inc., 706 pp.
Blanco, M. I. (2009). The economics of wind energy. Renewable and Sustainable Energy Reviews, 13(6-7), 1372-1382. doi:10.1016/j.rser.2008.09.004
Christensen, L. R., & Greene, W. H. (1976). Economies of Scale in U.S. Electric Power Generation. The Journal of Political Economy, 84(4), 655-676.
Commission of the European Communities. (2007). Communication from the Commission to the European Council and the European Parliament: An energy policy for europe. Brussels, Belgium,
Commission of the European Communities: 28 pp. Council resolution of 17 September 1974 concerning a new energy policy strategy for the Community,
No C 153 C.F.R. (1974). Cowing, T. G., & Smith, V. K. (1978). The Estimation of a Production Technology: A Survey of Econometric
Analyses of Steam Electric Generation. Land Economics, 54(2), 156-186. Dutton, J. M., & Thomas, A. (1984). Treating Progress Functions as a Managerial Opportunity. The
Academy of Management Review, 9(2), 235-247. Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning
common rules for the internal market in electricity L 027 C.F.R. (1996). Directive 98/30/EC of the European Parliament and of the Council of 22 June 1998 concerning common
rules for the internal market in natural gas, L 204 C.F.R. (1998). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion
of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC, L 140 C.F.R. (2009).
Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC, L 315 C.F.R. (2012).
Enos, J. L. (1962). Invention and Innovation in the Petroleum Refining Industry. In Universities-National Bureau Committee for Economic Research & Committee on Economic Growth of the Social Science Research Council (Eds.), The Rate and Direction of Inventive Activity: Economic and Social Factors (pp. 299-322): Princeton University Press.
Eurostat. (2016). Energy price statistics. Eurostat: Statistics Explained. Retrieved on 26/09/2016 from http://ec.europa.eu/eurostat/statistics-explained/index.php/Energy_price_statistics#Further_Eurostat_information.
Eurostat. (2013). Electricity prices for industrial consumers - bi-annual data (until 2007) [nrg_pc_205_h]. Retrieved on 26/09/2016 from http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nrg_pc_205_h&lang=en.
THE DYNAMICS OF ORC TECHNOLOGY: TECHNOLOGICAL INNOVATION, ECONOMIES OF SCALE AND LEARNING BY DOING
197
Eurostat. (2016). Electricity prices for industrial consumers - bi-annual data (from 2007 onwards) [nrg_pc_205]. Retrieved on 26/09/2016 from http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nrg_pc_205&lang=en.
Grübler, A. (1998). Technology and global change. Cambridge, UK: Cambridge University Press. Grübler, A., Nakicenovic, N., & Victor, D. G. (1999). Dynamics of energy technologies and global change.
Energy Policy, 27, 247-280. Grubler, A. (2010). The costs of the French nuclear scale-up: A case of negative learning by doing. Energy
Policy, 38(9), 5174-5188. doi:10.1016/j.enpol.2010.05.003 Hamilton, M. R., Herzog, H. J., & Parsons, J. E. (2009). Cost and U.S. public policy for new coal power
plants with carbon capture and sequestration. Energy Procedia, 1(1), 4487-4494. doi:10.1016/j.egypro.2009.02.266
Jaffe, A. B., Newell, R. G., & Stavins, R. N. (2004). Economics of Energy Efficiency. Encyclopedia of Energy, 2, 79-90.
Jaffe, A. B., & Stavins, R. N. (1994). The energy paradox and the diffusion of conservation technology. Resource and Energy Economics, 16, 91-122.
Junginger, M., de Visser, E., Hjort-Gregersen, K., Koornneef, J., Raven, R., Faaij, A., & Turkenburg, W. (2006). Technological learning in bioenergy systems. Energy Policy, 34(18), 4024-4041. doi:10.1016/j.enpol.2005.09.012
Junginger, M., Faaij, A., Björheden, R., & Turkenburg, W. C. (2005). Technological learning and cost reductions in wood fuel supply chains in Sweden. Biomass and Bioenergy, 29(6), 399-418. doi:10.1016/j.biombioe.2005.06.006
Junginger, M., van Sark, W., Kahouli-Brahmi, S., & Schaeffer, G. J. (2010). The experience curve approach: history, methodological aspects and applications. In M. Junginger, W. van Sark, & A. Faaij (Eds.), Technological Learning in the Energy Sector: Lessons for Policy, Industry and Science (pp. 9-17). Cheltenham, UK: Edward Elgar Publishing Limited.
Junginger, M., van Sark, W., & Faaij, A. (2010). Technological Learning in the Energy Sector: Lessons for Policy, Industry and Science. Cheltenham, UK: Edward Elgar Publishing Limited, 332 pp.
Koornneef, J., Junginger, M., & Faaij, A. (2007). Development of fluidized bed combustion—An overview of trends, performance and cost. Progress in Energy and Combustion Science, 33(1), 19-55. doi:10.1016/j.pecs.2006.07.001
Kouvaritakis, N., Soria, A., & Isoard, S. (2000). Modelling energy technology dynamics: methodology for adaptive expectations models with learning by doing and learning by searching. International Journal of Global Energy Issues, 14, 1-4.
Lemmens, S. (2016). Learning processes for energy efficiency technologies: case study organic Rankine cycles. Antwerp, Belgium, University of Antwerp: 30 pp.
Lundberg, E. (1961). Produktivitet och räntabilitet: studier i kapitalets betydelse inom svenskt näringliv. Stockholm, Sweden: P.A. Norstedt and Soner.
MacGregor, P. R., Maslak, C. E., & Stoll, H. G. (1991). The market outlook for integrated gasification combined cycle technology. Paper presented at the 4th International Exhibition and Conference for the Power Generation Industries, Houston, Texas, USA.
McDonald, A., & Schrattenholzer, L. (2001). Learning rates for energy technologies. Energy Policy, 29, 255-261.
McNerney, J., Doyne Farmer, J., & Trancik, J. E. (2011). Historical costs of coal-fired electricity and implications for the future. Energy Policy, 39(6), 3042-3054. doi:10.1016/j.enpol.2011.01.037
Neij, L. (1999). Cost dynamics of wind power. Energy, 24, 375-389. Meadows, D. H., Meadows, D. L., Randers, J., & Behrens, W. W. I. (1972). The limits to growth. New York,
USA: Universe Books, 205 pp.
CHAPTER FIVE
198
Neij, L. (1997). Use of experience curves to analyse the prospects for diffusion and adoption of renewable energy technology. Energy Policy, 23(13), 1099-1107.
Neij, L., Dannemand Andersen, P., Durstewitz, M., Helby, P., Hoppe-Kilpper, M., & Morthorst, P. E. (2003). Experience Curves: A Tool For Energy Policy Assessment (EXTOOL). Lund, Sweden,
Environmental and Energy Systems Studies, Lund University: 146 pp. Nerlove, M. (1963). Returns to Scale in Electricity Supply. In C. F. Christ (Ed.), Measurement in Economics:
Studies in Honor of Yehuda Grunfeld (pp. 167-198). Stanford, California, USA: Stanford University Press.
OECD/IEA. (2000). Experience Curves for Energy Technology Policy. Retrieved from Paris, France: Ostwald, P. F., & Reisdorf, J. B. (1979). Measurement of technology progress and capital cost for nuclear,
coal-fired, and gas-fired power plants using the learning curve. Engineering and Process Economics, 4, 435-454.
Quoilin, S., van den Broek, M., Declaye, S., Dewallef, P., & Lemort, V. (2013). Techno-economic survey of Organic Rankine Cycle (ORC) systems. Renewable and Sustainable Energy Reviews, 22, 168-186. doi:10.1016/j.rser.2013.01.028
Rogers, E. M. (2003). Diffusion of Innovations (Fifth ed.). New York, NY, USA: Free Press, 551 pp. Rosenberg, N. (1996a). Perspectives on Technology. New York, USA: Cambridge University Press, 353 pp. Rosenberg, N. (1996b). Uncertainty and Technological Change. Conference Series ; [Proceedings], 40, 91-
125. Rubin, E. S., Azevedo, I. M. L., Jaramillo, P., & Yeh, S. (2015). A review of learning rates for electricity
supply technologies. Energy Policy, 86, 198-218. doi:10.1016/j.enpol.2015.06.011 Schaeffer, G. J., Alsema, E., Seebregts, A., Beurskens, L., de Moor, H., van Sark, W., . . . Zuccaro, C. (2004).
Learning from the Sun. Analysis of the use of experience curves for energy policy purposes: The case of photovoltaic power. Final report of the Photex project. Energy research Centre of the Netherlands (ECN): 99 pp.
Tartière, T. (2016). ORC World Map. Retrieved on 15/03/2016 from http://orc-world-map.org/. Thomson Reuters. (2016). Web of Science [v.5.22.1]. Retrieved on 13/07/2016 from
https://apps.webofknowledge.com/UA_GeneralSearch_input.do?product=UA&search_mode=GeneralSearch&SID=V1JrMWu2JF6SKZ5uZ4Y&preferencesSaved=.
van der Zwaan, B., & Rabl, A. (2003). Prospects for PV: a learning curve analysis. Solar Energy, 74(1), 19-31. doi:10.1016/s0038-092x(03)00112-9
Vélez, F., Segovia, J. J., Martín, M. C., Antolín, G., Chejne, F., & Quijano, A. (2012). A technical, economical and market review of organic Rankine cycles for the conversion of low-grade heat for power generation. Renewable and Sustainable Energy Reviews, 16(6), 4175-4189. doi:10.1016/j.rser.2012.03.022
Wright, T. P. (1936). Factors Affecting the Cost of Airplanes. Journal of Aeronautical Sciences, 3(4), 122-128.
6. Concluding chapter:
synthesis, implications and
recommendations
CONCLUDING CHAPTER: SYNTHESIS, IMPLICATIONS AND RECOMMENDATIONS
201
6.1. Recapitulation of the key results
The starting point of this thesis is the interaction between technological innovation and current
challenges in energy sustainability. The approach in this work is technology-specific and the
technology under consideration is the organic Rankine cycle (ORC), adapted from the
conventional steam Rankine cycle. The ORC operates with alternative working fluids instead of
water and is suitable to generate power from lower-temperature thermal flows, including
renewable energy sources and excess heat streams. In other words, the technology has
potential to contribute to worldwide renewable energy and energy efficiency goals and is
therefore the subject of various research and development activities. The general setup of this
thesis was to obtain insight into this potential of ORC technology. The diffusion of an emerging
technology is influenced by many factors, of which the technical feasibility is just dimension.
Other factors include the economics of the technology itself, its interactions with other
technologies and among the actors in the diffusion network or the influence of policy makers on
research and development and on market formation. In the academic literature, there are at
least two frameworks that offer an outline to study the development path of an emerging
technology: strategic niche management and the technological innovation systems. Each of
these frameworks discusses the essential drivers and processes for the emergence of a novel
technology, or an entire technological system. Within the scope of this thesis, it was not
possible to address each of these aspects. The engineering aspects of ORC technology are well-
studied but the economic point of view is a largely unexploited strand of research. Therefore,
this thesis starts the investigation at the very basis and expands the scope of the analysis
gradually as new insights emerge. Central in this research is the technology itself. The focus
expands to the basic dimensions of economics, influence of the market and the economic
context, impact of the policy framework and, finally, the diffusion of the technology and the
effect of accumulated experience.
The first question formulated for this research was: “What are the merits of ORC technology?
And what is the state-of-the-art knowledge about its economics?”. These essentials were
investigated in chapter 2. The merits of ORC technology are generally understood as its ability to
process energy sources of smaller scales and with lower temperatures than commonly feasible
with the steam Rankine cycle. Therefore, the ORC is typically associated with renewable energy
sources such as biomass, geothermal or solar energy, or with industrial excess thermal energy
flows. Despite its perceived potential, there was until now no comprehensive overview of the
economics of the technology or its market development. The contribution of chapter 2 consists
of a thorough literature review of the literature to disclose the state-of-the-art knowledge on
the economics of ORC systems. The most important findings from this chapter are the following.
Firstly, it is sensible to categorize ORC technology according to the thermal energy source
applied. The practical context and the implementation of ORC systems vary strongly according
CHAPTER SIX
202
to the thermal source. Secondly, the technical perspective on ORC technology dominates the
literature. In the studies that do investigate other aspects focus primarily on the capital costs. In
general, the published costs vary considerably among and within each of the four heat source
categories. This is due to the fact that there is very little information published on the
economics of real ORC systems. A large share of the literature discusses the economics using
costs that have been estimated, based on newly proposed ORC system designs. There are
various approaches to estimate the costs of industrial processes, from scaling methods to
bottom-up cost estimation with component-specific cost correlations. The resulting ORC specific
investment costs in the literature are very diverse, but nearly none of the studies assess to what
extent these estimates are representative. The question here is to what extent these estimated
costs can be considered representative for the actual system costs. Furthermore, the
investment costs are mostly at the core of the economic study, because they are directly related
to the technical system setup as well as the financial feasibility of the investment. But the
competitiveness of an electricity generation technology is measured by the cost at which
electricity can be produced, i.e. the LCOE. Nonetheless, there are only very few studies that
assess the LCOE of ORC applications, with very diverging results. Again, the majority of these
LCOEs have been calculated based on estimated, and not real, ORC system costs. In summary,
the state-of-the-art knowledge on ORC economics is to a large extent based on projected costs,
because there is an overall lack of insights from commissioned systems.
Therefore, chapter 3 explored the principles of cost engineering. The goal of this chapter was to
get insight in the methodologies that are so widely used for ORC cost estimation, but not so
often critically assessed. The guiding question was: “To what extent can cost engineering
techniques be used to get insight in the costs of ORC systems?”. Several methods have been
developed to estimate the costs of industrial projects. Simpler approaches require less effort
but yield lower accuracy. High accuracies are possible with definitive and detailed estimates, but
these require a far-reaching detail of plant design. Cost estimates pursued in the frame of
research generally achieve accuracies in the range of order-of-magnitude (-50 % to +100 %),
study (-30 % to +50 %) or preliminary (-20 % to +30 %) estimates. This chapter investigated
several cost estimation techniques by applying them to estimate the costs of a heat recovery
ORC case study. The resulting values varied considerably according to the method used and with
respect to the actual costs. Part of this variation is due to the inherent inaccuracies of the
different cost estimation methods, but probably also due to the fact that the scaling factors,
correlations, multiplication factors, etc. have been established for process industries in general
and not for ORC technology in particular. One approach to improve the quality of future cost
estimates is to expand insights in real ORC costs and establish technology-specific scale factors
or component correlations.
CONCLUDING CHAPTER: SYNTHESIS, IMPLICATIONS AND RECOMMENDATIONS
203
Beyond this narrow technology-costs perspective, the diffusion of an emerging technology is
influenced by external factors such as the prevailing market conditions or the extent of policy
support. Chapter 4 addresses the quest for more insight in the economics of real ORC systems
and expands the scope of the investigation to the matter of project appraisal, the importance of
the parameters of influence and the impact of public policy. The central question of this chapter
reads as follows: “Which factors influence the financial feasibility of an investment in ORC
technology?”. The question is answered by means of an in-depth case study, concerning an
industrial heat recovery ORC system installed in the Flanders Region, Belgium. The most
important results are the following. Firstly, the project was evaluated positively with an IRR of
12.6 %. However, these theoretical projections were nuanced by the plant manager, who
reported that due to recurring defaults the produced electricity is lower than expected and the
IRR reaches approximately 8 % in practice. This positive evaluation is the result of the beneficial
impact of supportive government measures. In a scenario without any form of subsidy, but
accounting for corporate income taxation, the projected IRR amounts 5.4 %. For many industrial
firms this result may be on the lower side of their expected return, but with the currently
prevailing low interest rates this is investment is worthwhile. Considering the sensitivity of the
results, the strongest influence is measured for changes in the annual load hours and the
electricity price. The Monte Carlo analysis demonstrates that there is a probability of nearly 70
% that this is a good investment, when the expected return amounts 6%. Achieving a positive
return depends largely on attaining a sufficient amount of load hours, which depend on the
production of the main process but also on a good O&M strategy. The costs of O&M have a
negative impact on the overall project assessment, but it may be worthwhile to fine-tune the
O&M strategy, maybe even incurring additional costs. These additional costs will have only a
marginal influence compared to the additional benefits of an optimized ORC production time.
Finally, a central process for the development of emerging technologies is technology learning.
In order to learn, or gain experience with the new technology, it has to be applied in practice.
Therefore, chapter 5 broadens the scope of the investigation and is built around the question:
“How did ORC technology diffuse and how does the deployment of the technology influence its
costs?”. The contribution of this chapter exists on three key fronts. First of all, the knowledge on
the real costs of ORC technology is extended drastically by composing a database including
more than 100 relevant cases. Secondly, the diffusion of ORC technology in the market is
investigated thoroughly after compiling a second database which includes approximately 95 %
of all ORC systems worldwide. By knowledge of the author, this amount of empirical data has
never been collected for ORC technology. Thirdly, combining these two databases allows
investigating the dynamics of ORC technology innovation and competition. The most important
findings from this chapter are the following. The diffusion of ORC technology increased rapidly
after the start of the new millennium, a similar pattern compared to that of several renewable
energy technologies. This interest for renewable energy technologies, including ORC technology,
CHAPTER SIX
204
can be linked to sharply grown electricity prices and an increased awareness of energy issues
and its reflection in energy policies, particularly in Europe. Secondly, ORC systems exist in every
corner of the world, but the majority is commissioned in Europe, and biomass energy is the
most-used thermal energy source. The ORC manufacturer market is rather oligopolistic: two
companies, Ormat and Turboden, dominate the market in terms of total installed capacity and
amount of references. The intensive R&D required to develop a working prototype and the high
costs associated with producing the first units make market entrance difficult. Yet, new
manufacturers are entering the market and gain an increasing share of it. Lastly, the costs of
ORC technology are strongly subject to economies of scale. The scale effects are stronger for
ORC projects than for the modules, and the strongest for biomass ORCs. The effect of learning-
by-doing in the ORC market, measured as cost reduction corresponding to accumulating
installed capacity, is somewhat ambiguous.
6.2. Implications and recommendations
This research focussed on the potential of ORC technology to contribute to worldwide energy
sustainability challenges. The thesis investigated several dimensions in the development of an
emerging technology, including the technology’s economics, the influence of market conditions,
the impact of public policy as well as the dynamics of the technology itself and its costs. This
final section discusses the implications of the research and sets out the perspectives for the
future. This focusses is on three key parties of interest: the research community, the industry
and the policy makers.
For the research community, the most important implications of this study are the following. It
is a positive evolution that the study of ORC technology is increasingly expanded beyond the
narrow technical-engineering approach, mostly with an inclusion of a perspective on the
system’s costs. An important remark for future ORC research is the need for responsible
consideration of the economic data. In order to enhance comparability of future research
output, it is recommended to enhance the clarification of the scope of the economic data, both
in terms of the system that they represent as in terms of the origin of the costs. Moreover, few
publications go beyond the step of disclosing the estimated or real system costs. Estimated
costs should always be accompanied by a discussion on the accuracy of the results; real costs
situated in time and space. Finally, knowledge on the capital costs of an ORC system is useful
when for comparison of different system designs or to assess the financial appeal of the system.
To estimate ORC technology with respect to other technologies and competitiveness in the
market, the scope has to be expanded to a calculation of the electricity production costs of the
system. This thesis took large steps in the review of existing ORC economic insights and the
collection of new data, but there remain several questions open for investigation. Future lines
of research include a more profound investigation into the share of each of the components in
CONCLUDING CHAPTER: SYNTHESIS, IMPLICATIONS AND RECOMMENDATIONS
205
the ORC’s costs. The identification of component-specific economic data is an issue that the
current work was not able to solve, despite the excessive amount of data collected. Knowledge
on the cost characteristics of the separate components should allow making better cost
estimations in the future. Most of the components used in an ORC system concern standard
industrial equipment, but the expander is mostly technology-specific and smaller-scale
applications may require new designs, so that existing cost correlations are not suitable.
Expanding the scope by looking for typical multiplication factors between the costs of ORC
modules and projects is an option, but due to the diverse integration requirements such general
factors will never achieve high accuracies. One aspect that this thesis was able to solve is the
determination of valid scale factors for ORC technology, which can be used to estimate the costs
of ORC systems based on scale relations. Another, quite specific, suggestion for future research
is a detailed investigation of the actual potential of excess heat recovery. This potential is often
assumed substantial, but not so well quantified. Only when the volume of the wasted thermal
energy is known, valid projections about the contribution of heat recovery for the achievement
of energy efficiency goals can be made.
The implications and recommendations for the industry relate to both the manufacturers of
ORC technology and to the end-users. ORC technology has been proven economically feasible,
but under certain conditions. One of the – self-evident – key influences is the height of the
electricity price, but also the attainable load hours have a major importance. Another factor of
major importance is financial support for the investment, i.e. government intervention. The
investment costs of an ORC system remain high compared to its technical efficiency and, in a
profit-driven world, the return generated by an ORC investment is not always sufficient by itself.
The expansion of ORC technology is thus most likely to be successful in markets with high
electricity prices and a supportive regulatory framework. Alternatively, the future of ORC
technology may exist in the combined supply of ORC technology with other industrial
equipment, i.e. where ORC manufacturers integrate their system directly when new industrial
plants are constructed. In this way, the ORC is just one of the components in a larger project
and its overhead costs with be more limited compared to the case where it has to be
constructed as an ex-post add-on.
From the end-user’s perspective, the questions concern whether it is interesting to recover the
excess heat or to utilize the renewable energy source in the first place, and secondly, whether
ORC technology is the most suitable option to do so. In several markets the industry is
confronted with top-down requirements for renewable energy production and energy efficiency
enhancement. In such a context, the success of ORC technology depends on its competition
with other technologies. Industries confronted with excess heat streams may find more efficient
ways to reutilize these, such as process integration or heat delivery to district heating networks.
But in cases where these options are not possible, ORC is certainly a valid option. ORC
CHAPTER SIX
206
technology is currently the only commercially proven technology to generate power from low-
grade thermal sources. In markets where the regulatory framework is less stringent, i.e. where
the industry is not forced to take action, the success of ORC technology is subject to the
competitive rules of the market. In such cases, industries will probably only engage in ORC
technology investment when the electricity price is so high that the industry looks for
alternatives by itself, or when the capital cost for the technology decreases significantly so that
the investment in ORC technology is a more attractive option in the firm’s portfolio. Future lines
of research in this context include a more thorough comparison of the economics of ORC
technology and its (emerging) competitors. The competition concerns both the alternative uses
for the heat source, such as process integration or district heating, and the other technologies
that generate power from low-grade thermal sources.
With respect to the implications and recommendations for policy makers, the importance of
financial support for the diffusion of ORC technology has been proven. The question concerns
whether government support for ORC technology is worthwhile or not. Financial support helps
to overcome the burden of high initial investment costs for the end-user, but may also induce
the so-called ‘ride down the experience curve’. A technology has to be implemented in practice
and experienced has to be gained before the investment costs can reduce. Public policy can
support this process by supporting the initial uptake of the technology, gradually decreasing the
size of the intervention as the technology gains from experience. The overall experience curve
for ORC-generated electricity is somewhat ambiguous, but the curves estimated for biomass
applications and for the systems of the experienced manufacturer Turboden suggest that the
costs of ORC technology decrease with accumulating experience. Whether this policy support
has to focus on certain technologies, assessed to have suitable characteristics, or whether it
should be more generally oriented towards the more general achievement of renewable energy
and energy efficiency goals is another discussion. Future research questions in this domain
involve a deeper investigation of the required type and extent of policy support.
As a final conclusion: in light of the challenges to change the structure of energy supply and use,
the uptake of ORC technology will not be world-changing by itself. Other actions will have to be
taken to induce more pervasive changes. On the other hand, every major change is the result of
innumerable minor actions – each of which to be appreciated for its own contribution.
Appendix A. Summary of the literature
review on the economics
of ORC technology
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
209
A.1. Summary of the heat recovery ORC literature (section 2.3.2)
System scope
Module Project
Economic scope
Real
Tumen Ozdil and Segmen (2016): What: ORC for heat recovery in
Adana, Turkey Describes: investment costs, details
on component costs Financial analysis: PP
Leslie, Sweetser, Zimron, and Stovall (2009): What: ORC for heat recovery from
gas turbine in North Dakota, USA Describes: specific investment costs Financial analysis: NPV, IRR
/ Tchanche, Declaye, Quoilin, Papadakis, and Lemort (2010): What: heat recovery ORC,
laboratory prototype Describes: detailed investment
costs, costs of larger system estimated through extrapolation
Financial analysis: NPV, LCOE
Quote Gaia and Macchi (1981): What: ORC for heat recovery from
ceramic tunnel oven Describes: specific investment costs,
no details Financial analysis: PP
Angelino, Gaia, and Macchi (1984): What: heat recovery ORC Describes: detailed investment
costs Financial analysis: PP
Vescovo (2009): What: ORC for heat recovery in
cement, glass and steel industry Describes: investment costs of ORC
modules and projects for the three cases
Financial analysis: NPV, IRR
Vescovo (2009): What: ORC for heat recovery in
cement, glass and steel industry Describes: investment costs of ORC
modules and projects for the three cases
Financial analysis: NPV, IRR Forni, Vaccari, Di Santo, Rossetti, and
Baresi (2012): What: ORC for heat recovery in
cement, glass, steel and oil and gas industry
Describes: investment costs of ORC modules and projects for the four cases
Financial analysis: NPV, IRR, PP
Forni et al. (2012): What: ORC for heat recovery in
cement, glass, steel and oil and gas industry
Describes: investment costs of ORC modules and projects for the four cases
Financial analysis: NPV, IRR, PP
David, Michel, and Sanchez (2011): What: ORC for heat recovery from
cokes oven and from biogas engine Describes: detailed project
investment costs Financial analysis: PP
David et al. (2011): What: ORC for heat recovery from
cokes oven and from biogas engine Describes: detailed project
investment costs Financial analysis: PP
Ghirardo, Santin, Traverso, and Massardo (2011):
/
APPENDIX A
210
What: ORC for heat recovery from an on-board solid oxide fuel cell, compared with other technologies
Describes: project investment costs, no details, estimated by scaling from a manufacturer module quote
Financial analysis: cost of electricity Vanslambrouck, Vankeirsbilck,
Gusev, and De Paepe (2011): What: heat recovery ORCs Describes: indicative investment
costs, based on accumulated experience with ORC projects
Financial analysis: PP
/
Estimated Lee, Kuo, Chien, and Shih (1988): What: ORC for heat recovery Describes: bottom-up estimate of
investment, annual and other costs, details per component given
Financial analysis: net present benefit, cost-benefit ratio, PP
Tchanche et al. (2010): What: heat recovery ORC,
laboratory prototype Describes: detailed investment
costs, costs of larger system estimated through extrapolation
Financial analysis: NPV, LCOE F. Yang et al. (2015):
What: ORC for heat recovery from an internal combustion diesel engine
Describes: bottom-up estimate of investment costs, no details
Financial analysis: electricity production cost
Ghirardo et al. (2011): What: ORC for heat recovery from
an on-board solid oxide fuel cell, compared with other technologies
Describes: project investment costs, no details, estimated by scaling from a manufacturer module quote
Financial analysis: cost of electricity S. Lecompte, Lazova, van den Broek,
and De Paepe (2014): What: compare subcritical and
transcritical heat recovery ORC Describes: investment costs,
graphical display of details, bottom-up estimate of investment costs using component correlations
Financial analysis: no
Walsh and Thornley (2013): What: heat recovery ORC compared
with condensing boiler Describes: specific investment costs,
no details, power law method used but not explained
Financial analysis: NPV, discounted PP
M.-H. Yang (2015): What: transcritical ORC for heat
recovery from marine diesel engine Describes: investment costs, no
details, bottom-up estimate using component correlations
Financial analysis: PP, LCOE
Quoilin, Declaye, Tchanche, and Lemort (2011): What: heat recovery ORC, different
working fluids Describes: investment costs,
graphical display of details, estimated with cost correlations from literature and estimated by authors themselves
Financial analysis: no
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
211
/ S. Lecompte, Huisseune, van den Broek, De Schampheleire, and De Paepe (2013): What: heat recovery ORC for
cogeneration under real part-load conditions
Describes: investment costs, graphical display of details, bottom-up estimate of investment costs using component correlations
Financial analysis: no / Steven Lecompte, Lemmens,
Huisseune, van den Broek, and De Paepe (2015): What: compare subcritical and
transcritical heat recovery ORC Describes: investment costs,
graphical display of details, using component correlations
Financial analysis: NPV, PP / Pierobon, Nguyen, Larsen, Haglind,
and Elmegaard (2013): What: ORC for heat recovery from a
gas turbine on an offshore platform Describes: detailed investment
costs, using component correlations Financial analysis: NPV, DPP
/ Le, Kheiri, Feidt, and Pelloux-Prayer (2014): What: ORC for heat recovery,
different fluids Describes: detailed investment
costs, bottom-up estimate using component correlations
Financial analysis: LCOE, PP, IRR / Kwak, Binns, and Kim (2014):
What: heat recovery ORC, compared with other technologies
Describes: investment, costs, no details, bottom-up estimate using component correlations
Financial analysis: cost of electricity
APPENDIX A
212
/ Amini, Mirkhani, Pakjesm Pourfard, Ashjaee, and Khodkar (2015): What: transcritical ORC for heat
recovery from combined-cycle power plant
Describes: investment costs, no details, bottom-up estimate using component correlations
Financial analysis: no / M.-H. Yang and Yeh (2015):
What: ORC for heat recovery from marine diesel engines, different fluids
Describes: investment costs, no details, bottom-up estimate using component correlations
Financial analysis: net power output index
/ Di Maria and Micale (2015): What: ORC for exhaust recovery
from aerobic treatment of organic waste
Describes: detailed investment costs, bottom-up estimate using component correlations
Financial analysis: no / Heberle and Brüggemann (2016):
What: heat recovery ORC Describes: investment costs, no
details, bottom-up estimate using component correlations, project costs are estimated based on module costs for different multiplications factors
Financial analysis: LCOE / Najjar and Radhwan (1988):
What: ORC for heat recovery from gas turbine
Describes: detailed investment costs, estimation method not clear
Financial analysis: PP / Kalina (2011):
What: ORC for heat recovery in biomass gasification internal combustion engine
Describes: investment costs, estimation method not clear
Financial analysis: DPB, IRR
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
213
/ Law, Harvey, and Reay (2013): What: heat recovery ORC compared
with heat pump Describes: investment costs, no
details, estimation method unclear Financial analysis: net savings
/ Lukawski (2009): What: heat recovery ORC Describes: detailed bottom-up
estimate of purchased equipment costs and investment costs, power output unclear
Financial analysis: LCOE / Imran et al. (2014):
What: basic and regenerative ORC for heat recovery
Describes: specific investment costs, no details, power output unclear
Financial analysis: no / Papadopoulos, Stijepovic, and Linke
(2010): What: heat recovery ORC, screening
of working fluids Describes: estimate of costs,
numerical results SIC not provided Financial analysis: no
/ Cayer, Galanis, and Nesreddine (2010): What: transcritical ORC Describes: bottom-up estimate of
investment costs, numerical results SIC not provided
Financial analysis: no / Grabińsky (2011):
What: ORC for heat recovery from bio-liquid diesel power plant
Describes: bottom-up estimate of investment costs, numerical results SIC not provided
Financial analysis: NPV, PP
APPENDIX A
214
/ Meinel, Wieland, and Spliethoff (2014): What: ORC for heat recovery from
biomass digestion plant, compare standard cycle with recuperator cycle and two-stage regenerative pre-heating cycle
Describes: bottom-up estimate of investment costs, numerical results SIC not provided
Financial analysis: no / Z. Q. Wang, Zhou, Guo, and Wang
(2012): What: heat recovery ORC Describes: heat exchanger costs
estimated using correlation, results not revealed, optimization using ratio of heat transfer area and net power output
Financial analysis: PP / Y.-R. Li, Wang, and Du (2012):
What: heat recovery ORC Describes: no costs estimated, heat
exchanger area used as proxy Financial analysis: economic
performance assessed as ratio of net power output and heat transfer area
/ D. Wang, Ling, and Peng (2013): What: heat recovery ORC Describes: no costs estimated, heat
exchanger area used as proxy Financial analysis: economic
performance assessed as ratio of net power output and heat transfer area
/ J. Wang, Yan, Wang, Ma, and Dai (2013): What: heat recovery ORC Describes: no costs estimated, heat
exchanger area used as proxy Financial analysis: economic
performance assessed as ratio of net power output and heat transfer area
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
215
/ Wu, Zhu, and Yu (2016): What: heat recovery ORC Describes: no costs estimated Financial analysis: economic
performance assessed as ratio of net power output and UA (product of heat transfer coefficient and heat transfer area)
Assumed / Yari and Mahmoudi (2010): What: ORC for heat recovery from
gas turbine-modular helium reactor Describes: specific investment costs
assumed Financial analysis: cost of electricity
Arvay, Muller, Ramdeen, and Cunningham (2011): What: ORC for heat recovery from
biopolymer plant and ORC for bottoming in cogeneration
Describes: specific investment costs assumed
Financial analysis: PP
Arvay et al. (2011): What: ORC for heat recovery from
biopolymer plant and ORC for bottoming in cogeneration
Describes: specific investment costs assumed
Financial analysis: PP
/ Schuster, Karellas, Kakaras, and Spliethoff (2009): What: ORC for heat recovery from
biogas digestion plant Describes: specific investment costs
assumed Financial analysis: electricity
production costs
APPENDIX A
216
A.2. Summary of the biomass ORC literature (section 2.3.3)
System scope
Module Project
Economic scope
Real
/ STIA - Holzindustrie Ges.m.b.H. (2001): What: biomass CHP ORC in Admont,
Austria Describes: investment costs,
detailed annual and other costs Financial analysis: PP
Ingwald Obernberger, Thonhofer, and Reisenhofer (2002) & I. Obernberger, Carlsen, and Biedermann (2003): What: biomass CHP ORC in Lienz,
Austria Describes: detailed investment,
annual and other costs Financial analysis: specific
electricity production costs
Ingwald Obernberger et al. (2002) & I. Obernberger et al. (2003): What: biomass CHP ORC in Lienz,
Austria Describes: detailed investment,
annual and other costs Financial analysis: specific
electricity production costs
/
Barz (2008): What: biomass CHP ORC in
Friedland, Germany Describes: investment costs, no
details Financial analysis: no
/ Bini et al. (2004): What: biomass CHP ORC in Tirano,
Italy Describes: investment costs Financial analysis: NPV, DPP, IRR
/ Tańczuk and Ulbrich (2013): What: biomass ORC in Stuttgart-
Ostfildern, Germany Describes: investment costs, no
details, costs of smaller-scale system estimated through scaling
Financial analysis: NPV, DPP
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
217
/ Bolhar-Nordenkampf, Pröll, Aichernig, and Hofbauer (2004): What: compare six technologies for
biomass cogeneration , location of ORC unknown
Describes: SIC, no details Financial analysis: no
/ Duvia, Bini, Spanring, and Portenkirchner (2007): What: biomass CHP ORC in Salzburg,
Austria Describes: investment costs, no
details Financial analysis: PP
Quote Peretti (2008): What: biomass CHP ORCs for
sawmills Describes: quotes of investment
costs of three ORC projects, module costs can be calculated from these
Financial analysis: PP
Peretti (2008): What: biomass CHP ORCs for
sawmills Describes: quotes of investment
costs of three ORC projects, module costs can be calculated from these
Financial analysis: PP Duvia and Tavolo (2008):
What: biomass CHP ORCs for pellet industry
Describes: quotes of investment costs of three ORC projects, module costs can be calculated from these
Financial analysis: PP
Duvia and Tavolo (2008): What: biomass CHP ORCs for pellet
industry Describes: quotes of investment
costs of three ORC projects, module costs can be calculated from these
Financial analysis: PP / Duvia, Guercio, and Rossi di Schio
(2009): What: biomass CHP ORCs for
different industries Describes: quotes of investment
costs of ORC projects Financial analysis: PP
Estimated / Chinese, Meneghetti, and Nardin (2004): What: biomass CHP ORC for
furniture manufacturing district Describes: investment costs,
obtained via interpolation of data from literature and manufacturer data
Financial analysis: net annual profit, PP
APPENDIX A
218
/ Uris, Linares, and Arenas (2014): What: biomass CHP ORC, different
designs Describes: investment costs,
obtained via interpolation of data from existing plant
Financial analysis: IRR
/ Huang, Wang, et al. (2013): What: trigeneration ORC, using
three types of biomass Describes: bottom-up estimate of
investment costs, detailed annual and other costs
Financial analysis: PP and break-even electricity selling price
/ Huang, McIlveen-Wright, et al. (2013): What: CHP ORC, using two types of
biomass Describes: bottom-up estimate of
investment costs, detailed annual and other costs
Financial analysis: break-even electricity selling price
Tańczuk and Ulbrich (2013): What: biomass ORC in Stuttgart-
Ostfildern, Germany Describes: investment costs, no
details, costs of smaller-scale system estimated through scaling
Financial analysis: NPV, DPP / Ahmadi, Dincer, and Rosen (2014):
What: biomass system for electricity, cooling, heating, hydrogen, domestic hot water and desalinated fresh water
Describes: bottom-up estimate of investment costs, but costs not revealed
Financial analysis: total cost rate
Assumed Gard (2008): What: biomass CHP ORC Describes: investment costs,
assumed based on data from literature
Financial analysis: cost of electricity, cost of heat
Gard (2008): What: biomass CHP ORC Describes: investment costs,
assumed based on data from literature
Financial analysis: cost of electricity, cost of heat
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
219
Moro, Pinamonti, and Reini (2008): What: biomass CHP ORC for
furniture industry Describes: investment costs,
obtained via interpolation of data from literature and manufacturer data
Financial analysis: NPV, IRR, PP
Wood and Rowley (2011): What: compare six biomass CHP
technologies Describes: assumed investment
costs, no details Financial analysis: NPV
/ Rentizelas, Karellas, Kakaras, and Tatsiopoulos (2009): What: compare ORC with
gasification for biomass trigeneration
Describes: assumed investment costs, no details
Financial analysis: NPV, IRR, PP /
Kempegowda, Skreiberg, and Tran (2012): What: compare multiple biomass
CHP technologies in the Norwegian context
Describes: investment costs, assumed based on communication with manufacturers
Financial analysis: NPV, IRR, cost of generation
/ Maraver, Sin, Royo, and Sebastián (2013): What: biomass CCHP ORC,
compared with Stirling engine Describes: investment costs,
assumed based on literature Financial analysis: annual total cost
savings / Gonçalves, Faias, and de Sousa
(2012): What: biomass CHP ORC in sawmills Describes: investment costs,
assumed based on literature Financial analysis: NPV
/ Maraver, Uche, and Royo (2012): What: biomass polygeneration ORC,
combined with multi-effect water distillation
Describes: investment costs, assumed based on literature
Financial analysis: NPV, PP
APPENDIX A
220
/ Algieri and Morrone (2014): What: biomass CCHP ORC for
single-family application in Italy Describes: investment costs,
assumed Financial analysis: NPV, PP
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
221
A.3. Summary of the geothermal ORC literature (section 2.3.4)
System scope
Module Project
Economic scope
Real
/ Entingh and F. (2003): What: review of geothermal power
system costs Describes: specific investment costs
of two real cases and quotes for two other cases, but installed power capacities not provided
Financial analysis: no / Pernecker (1999):
What: geothermal ORC in Altheim, Austria
Describes: investment costs Financial analysis: no
/ Holdmann (2007): What: geothermal ORC in Fairbanks,
Alaska, USA Describes: investment costs, no
details Financial analysis: no
/ Lazzaretto, Toffolo, Manente, Rossi, and Paci (2011): What: geothermal ORC Describes: detailed investment
costs, compare results of bottom-up estimate with costs of known geothermal plant
Financial analysis: LCOE / Toffolo, Lazzaretto, Manente, and
Paci (2014): What: geothermal ORC Describes: detailed investment
costs, compare results of bottom-up estimate with costs of known geothermal plant
Financial analysis: LCOE
Quote / Forsha and Nichols (1991): What: geothermal ORC Describes: detailed specific
investment costs Financial analysis: no
/ Entingh and F. (2003): What: review of geothermal power
system costs
APPENDIX A
222
Describes: specific investment costs of two real cases and quotes for two other cases, but installed power capacities not provided
Financial analysis: no
Estimated / I.K. Smith, Stosic, Kovacevic, and Langson (2007): What: geothermal ORC Describes: detailed specific
investment costs, estimated based on scaling of known costs of laboratory prototype
Financial analysis: no / Campos Rodríguez et al. (2013):
What: compare ORC and Kalina cycle for low temperature enhanced geothermal system in Brazil
Describes: detailed investment costs, estimated based on past purchase orders and quotations from experienced professional cost estimators
Financial analysis: LCOE Astolfi, Romano, Bombarda, and
Macchi (2014a, 2014b): What: ORC for low-medium
enthalpy geothermal sources Describes: specific investment costs,
graphical details, estimated using cost correlations
Financial analysis: no
Astolfi et al. (2014a, 2014b): What: ORC for low-medium
enthalpy geothermal sources Describes: specific investment costs,
graphical details, estimated using cost correlations
Financial analysis: no
Lazzaretto et al. (2011): What: geothermal ORC Describes: detailed investment
costs, compare results of bottom-up estimate with costs of known geothermal plant
Financial analysis: LCOE
Lazzaretto et al. (2011): What: geothermal ORC Describes: detailed investment
costs, compare results of bottom-up estimate with costs of known geothermal plant
Financial analysis: LCOE Toffolo et al. (2014):
What: geothermal ORC Describes: detailed investment
costs, compare results of bottom-up estimate with costs of known geothermal plant
Financial analysis: LCOE
Toffolo et al. (2014): What: geothermal ORC Describes: detailed investment
costs, compare results of bottom-up estimate with costs of known geothermal plant
Financial analysis: LCOE M. Li et al. (2014):
What: ORC compared to COC2 transcritical cycle for geothermal
/
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
223
sources Describes: specific investment costs,
no details, estimated using cost correlations
Financial analysis: ratio of the heat exchanger’s costs to the whole system’s costs
Heberle and Brüggemann (2015): What: geothermal ORC,
investigating zeotropic working fluids
Describes: detailed specific investment costs, estimated using cost correlations
Financial analysis: electricity generation costs
Heberle and Brüggemann (2015): What: geothermal ORC,
investigating zeotropic working fluids
Describes: detailed specific investment costs, estimated using cost correlations
Financial analysis: electricity generation costs
Zare (2015): What: compare simple,
regenerative and ORC with internal heat exchanger for binary geothermal plants
Describes: specific investment costs graphically displayed, bottom-up estimate
Financial analysis: PP
Walraven, Laenen, and D'Haeseleer (2015): What: geothermal ORC with air-
cooled condenser Describes: detailed specific
investment costs, graphical details, estimated using cost correlations
Financial analysis: NPV
/
Walraven, Laenen, and D’haeseleer (2015): What: geothermal ORC, comparing
air-cooled and water-cooled condensation
Describes: detailed specific investment costs, graphical details, estimated using cost correlations
Financial analysis: LCOE / M.-H. Yang and Yeh (2016):
What: geothermal ORC Describes: detailed specific
investment costs Financial analysis: net power output
index / Madhawa Hettiarachchi, Golubovic,
Worek, and Ikegami (2007): What: geothermal ORC Describes: no costs estimated Financial analysis: economic
performance assessed as ratio of heat exchanger area to net power output
APPENDIX A
224
Assumed Preißinger, Heberle, and Bruggemann (2013): What: sub- and transcritical ORC for
geothermal applications, using different fluids
Describes: specific investment costs, no details, assumed
Financial analysis: PP
Ian K. Smith, Stosic, and Kovacevic (2005): What: geothermal binary ORC using
screw expander Describes: investment costs of
generator, condenser and cooling not included, assumed
Financial analysis: no / Oguz Arslan and Yetik (2011):
What: supercritical binary geothermal ORC for the Simav geothermal field, Turkey
Describes: investment costs, detailed specific investment costs assumed
Financial analysis: no / O. Arslan, Ozgur, and Kose (2012):
What: geothermal ORC for the Simav geothermal field, Turkey
Describes: investment costs, detailed specific investment costs assumed
Financial analysis: compares costs to benefits
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
225
A.2. Summary of the solar ORC literature (section 2.3.5)
System scope
Module Project
Economic scope
Real
Baral and Kim (2015): What: solar ORC, laboratory
prototype Describes: detailed investment
costs, most costs from lab ORC, solar collector and fluid costs estimated
Financial analysis: PP
Manolakos, Mohamed, Karagiannis, and Papadakis (2008): What: solar ORC, laboratory
prototype Describes: investment costs,
probably real costs from laboratory prototype
Financial analysis: PP / Canada, Cohen, Cable, Brosseau, and
Price (2004); Vélez et al. (2012): What: solar ORC with parabolic
through collectors, installed in Saguaro, Arizona, USA
Describes: specific investment costs, not given in paper itself, but by Vélez et al. (2012)
Financial analysis: no
Quote Prabhu (2006): What: solar ORC with parabolic
trough collectors, Describes: detailed investment
costs, estimated as pre-study for real case
Financial analysis: no
Prabhu (2006): What: solar ORC with parabolic
trough collectors, Describes: detailed investment
costs, estimated as pre-study for real case
Financial analysis: no / Barber (1978):
What: solar ORC Describes: specific investment costs,
graphically, estimation method not clear
Financial analysis: no
Estimated Georges, Declaye, Dumont, Quoilin, and Lemort (2013): What: solar ORC, prototype Describes: detailed investment
costs, origin not clear Financial analysis: no
Nafey and Sharaf (2010): What: solar ORC for reverse osmosis
desalination Describes: costs estimated using
component correlations, but numerical results SIC not provided
Financial analysis: no Oliveira et al. (2002):
What: solar trigeneration ORC Describes: specific investment costs,
no details Financial analysis: no
Nafey, Sharaf, and García-Rodríguez (2010): What: solar ORC for reverse osmosis
desalination Describes: costs estimated using
component correlations, but numerical results SIC not provided
APPENDIX A
226
Financial analysis: no Freeman, Hellgardt, and Markides
(2015): What: solar cogeneration ORC for
domestic use in the UK Describes: investment costs,
estimated from component costs Financial analysis: LCOE, DPB
Sharaf, Nafey, and García-Rodríguez (2011): What: compare two solar ORC
configurations for multi effect distillation
Describes: costs estimated using component correlations, but numerical results SIC not provided
Financial analysis: no Kosmadakis, Manolakos, Kyritsis, and
Papadakis (2009): What: two-stage solar ORC for
reverse osmosis desalination Describes: detailed investment
costs, estimation method not clear Financial analysis: NPV, benefit-cost
ratio, PP, IRR
Kosmadakis et al. (2009): What: two-stage solar ORC for
reverse osmosis desalination Describes: detailed investment
costs, estimation method not clear Financial analysis: NPV, benefit-cost
ratio, PP, IRR
Yamaguchi, Zhang, Fujima, Enomoto, and Sawada (2006): What: supercritical solar ORC Describes: numerical results SIC not
provided, estimation method not clear
Financial analysis: no
Assumed Villarini et al. (2013): What: solar collector field, ORC,
absorber unit and thermal storage unit
Describes: detailed investment costs, assumed
Financial analysis: NPV, IRR
Villarini et al. (2013): What: solar collector field, ORC,
absorber unit and thermal storage unit
Describes: detailed investment costs, assumed
Financial analysis: NPV, IRR / Bruno, López-Villada, Letelier,
Romera, and Coronas (2008): What: solar thermal ORC for reverse
osmosis desalination Describes: specific investment costs,
assumed Financial analysis: no
/ Kosmadakis, Manolakos, and Papadakis (2011): What: solar concentrating
photovoltaic system combined with ORC
Describes: specific investment costs, no details, assumed
Financial analysis: NPV, PP
SUMMARY OF THE LITERATURE REVIEW ON THE ECONOMICS OF ORC TECHNOLOGY
227
Karellas, Terzis, and Manolakos (2011): What: solar thermal ORC
photovoltaic desalination system for the Chalki Island, Greece
Describes: detailed specific investment and annual costs, probably assumed
Financial analysis: NPV, IRR, DPP, Levelized Water Cost
228
Appendix B. Numerical results for the
financial appraisal of
the case study
APPENDIX B: NUMERICAL RESULTS FOR THE FINANCIAL APPRAISAL OF THE HEAT RECOVERY CASE STUDY
231
B.1. Results of the financial appraisal (section 4.3.3)
Table B. 1. Numerical results of the project appraisal, for high, medium and low discount rates.
Discount rate
scenario
NPV
[k€]
IRR
[%]
PP
[y]
LCOE
[€/MWh]
Policy LCOE
[€/MWh]
High (𝐫 = 𝟏𝟐%) 33 12.6 6 147 66
Medium (𝐫 = 𝟔%) 504 12.6 6 102 44
Low (𝐫 = 𝟑%) 898 12.6 6 83 35
B.2. Impact of public policy (section 4.4.1)
Table B. 2. Numerical results of the project appraisal for the ‘No Policy’ scenario, for high, medium and low discount rates.
Discount rate
scenario
NPV
[k€]
IRR
[%]
PP
[y]
LCOE
[€/MWh]
Policy LCOE
[€/MWh]
High (𝐫 = 𝟏𝟐%) -522 6.7 11 147 147
Medium (𝐫 = 𝟔%) 99 6.7 11 102 102
Low (𝐫 = 𝟑%) 638 6.7 11 83 83
Table B. 3. Numerical results of the project appraisal for the ‘Tax & Premium’ scenario, for high, medium and low discount rates.
Discount rate
scenario
NPV
[k€]
IRR
[%]
PP
[y]
LCOE
[€/MWh]
Policy LCOE
[€/MWh]
High (𝐫 = 𝟏𝟐%) -37 11.4 6 147 72
Medium (𝐫 = 𝟔%) 431 11.4 6 102 48
Low (𝐫 = 𝟑%) 822 11.4 6 83 38
Table B. 4. Numerical results of the project appraisal for the ‘Tax & Deduction’ scenario, for high, medium and low discount
rates.
Discount rate
scenario
NPV
[k€]
IRR
[%]
PP
[y]
LCOE
[€/MWh]
Policy LCOE
[€/MWh]
High (𝐫 = 𝟏𝟐%) -469 6 11 147 108
Medium (𝐫 = 𝟔%) 3 6 11 102 71
Low (𝐫 = 𝟑%) 396 6 11 83 56
Table B. 5. Numerical results of the project appraisal for the ‘Only Taxes’ scenario, for high, medium and low discount rates.
Discount rate
scenario
NPV
[k€]
IRR
[%]
PP
[y]
LCOE
[€/MWh]
Policy LCOE
[€/MWh]
High (𝐫 = 𝟏𝟐%) -538 5.4 12 147 113
Medium (𝐫 = 𝟔%) -71 5.4 12 102 75
Low (𝐫 = 𝟑%) 321 5.4 12 83 59
232
233
Appendix C. Survey from ORC end-users
SURVEY FROM ORC END-USERS
235
C.1. Introduction
Thank you for participating! The survey starts on the next page. First some brief instructions and definitions.
1) The setup of an ORC system differs for each application and on-site requirements or limitations. The following terminology is used to
facilitate the scope of the questions:
Module: refers to only the ORC unit itself. It is the foundation of the ORC system (evaporator, expander, condenser, pump, generator).
ORC Integration: refers to the equipment needed to install the ORC module. The integration requirements differ strongly per type of heat source
and according to the requirements and limitations on-site.
Project costs: refer to the additional costs that are involved in the completion of the ORC project. These include labor costs, costs for start-up,
engineering, ….
ORC Project: refers to the complete ORC power system, including the necessary piping, heat exchangers, construction labour, start-up,
contingencies, … These integration requirements are very site-specific and change per application.
ORC System: generally refers to the ORC power system, when the distinction between module or project is not relevant.
2) An ORC project which contains multiple ORC modules is considered as one block. For instance, a 100 kW ORC project which is composed
of two 50kW modules is seen as one 100 kW project.
ORC Module Equipment such as:
- Heat recovery heat exchanger;
- Biomass combustion equipment;
- Geothermal well piping & pumps;
- Solar installation;
- …
ORC Integration Non-equipment costs such as:
- Costs for transportation;
- Labor costs for on-site
construction;
- Costs for start-up;
- Costs for engineering or project
planning;
- …
Project costs
ORC Project
APPENDIX C
236
3) If you have multiple ORC projects in your firm, please fill out this questionnaire for one, representative project. You can complete this
survey multiple times for the other ORC projects in your firm.
C.2. Questions
C.2.1 Location and application
CATEGORY QUESTION ANSWERS GOAL
SCOPE Location *
1. Where is your ORC system installed?
Indicate the country where the ORC is located (disclosing the region or city is optional, but not mandatory). TEXTBOX: Country *
Identifying the geographical distribution of ORC systems.
SCOPE Sector *
2. My company is active in the following sector:
Select from the following options (more than one answer possible).
Identifying the sectors where ORCs are used.
Steel industry
Cement industry
Glass industry
Chemical industry
Energy industry
Wood industry
Waste processing
Biomass industry
District heating
Public sector
Recreation
Other: …
SCOPE Heat source *
3. Which thermal input is used for the ORC system?
Select from the following options (more than one answer possible).
Identifying how much each of the possible heat sources are used to operate ORCs. Biomass
Geothermal
Recovered (waste) heat
Solar
Other: …
SURVEY FROM ORC END-USERS
237
SCOPE Application *
4. The ORC is used for: Select from the following options (more than one answer possible).
Electricity generation
Heat delivery to another process
Heat delivery to a district heating network
Heat delivery to buildings
Other: …
Identifying how much of the ORCs are used for more than electricity generation.
APPENDIX C
238
C.2.2 Manufacturer and seller CATEGORY QUESTION ANSWERS GOAL
SCOPE Manufacturer **
5. a) Who is the manufacturer of your ORC system?
Select from the following options (only one answer possible). (manufacturers are displayed with logo)
Identifying the manufacturer of the ORC system.
Turboden
Ormat
Electratherm
Exergy
Triogen
Adoratec/Maxxtec
BEP Europe
Enertime
Other: …
SCOPE Model
b) Is the ORC module a standard model (from a manufacturer product line) or is it custom made?
Select from the following options (only one answer possible) and, if applicable, specify the model.
Standard model from a manufacturer product line: …
Custom made
I don’t know
Identifying whether ORCs in the market are mainly custom-made or from a standard product line.
MARKET Seller
c) The ORC system was purchased:
Select from the following options (only one answer possible).
directly from the ORC manufacturer itself (e.g., Turboden, Ormat, …)
from a third party (e.g., engineering firm)
other: …
Identifying whether the ORCs are sold primarily by the manufacturers themselves or by other parties.
SURVEY FROM ORC END-USERS
239
C.2.3 Motivation and alternatives CATEGORY QUESTION ANSWERS GOAL
MOTIVATION Acquaintance
6. How did you learn about ORC systems as a solution for your firm?
Select from the following options (more than one answer possible).
I was contacted by an ORC installer/manufacturer
I learned about ORC systems from a branch colleague
I learned about ORC systems at a conference
I learned about ORC systems from a sector federation
I learned about ORC systems from a government institution or document
I learned about ORC systems on the internet
Other: …
Identifying where/how firms learn about ORCs.
MOTIVATION Incentives/ Advantages
7. Which were the main reasons to invest in the ORC system?
Rank the following options in order of importance by typing numbers (1, 2, 3, ..) into the boxes (1 is the most important). Choose as many options as you want, leave the irrelevant options blank.
High energy price on the market
The investment costs of the ORC were low
Financial support (subsidies, green/white certificates, tax reductions, …)
We had to find a useful application for the heat source to comply with quality certifications (such as ISO / EMAS)
We had to find a useful application for the heat source because our firm participates in a voluntary agreement (agreements with the government on energy use and efficiency)
Identifying the most important incentives for ORC investments.
APPENDIX C
240
Other firms in the sector have been installing ORC systems
We did not purchase the ORC system, but I have an energy services agreement (ESCO) with the manufacturer (the manufacturer pays for the system, and you have the advantage of the ORC production)
We want generate more renewable energy to contribute to global energy & climate challenges
We want to improve the energy use (efficiency) of our firm to contribute to global energy & climate challenges
We want to generate more renewable energy because we want to make our firm ‘greener’
We want to improve the energy use (efficiency) of my firm because I want to make our firm ‘greener’
Other: …
MOTIVATION Concerns/ Barriers/ Disadvantages
8. Did you have concerns while evaluating ORC systems as a solution for your firm?
Select from the following options (more than one answer possible).
No, we had no concerns
The technology is new
We were concerned that the production from the ORC would be lower than expected
We were concerned that the ORC would interrupt the operation of our principal processes (e.g., in the case of heat recovery)
We were concerned about the price of the working fluid
We were concerned about the safety of the working fluid
Other: …
Identifying barriers impeding ORC investments.
SURVEY FROM ORC END-USERS
241
ALTERNATIVES Existence
9. a) Did you have other options to utilize the heat source (e.g., the biomass or the waste heat), instead of using it with the ORC?
Select from the following options (more than one answer possible).
We did not look for other options, we considered only the ORC
We looked for alternatives to use the heat source (instead the ORC), but there were no other possibilities
We looked for alternatives to use the heat source (instead the ORC), but the ORC was the only technically feasible solution
Yes, a feasible alternative was to install a steam turbine
Yes, a feasible alternative was to deliver the heat to another process (without ORC)
Yes, a feasible alternative was to deliver the heat to a district heating network (without ORC)
Yes, a feasible alternative was to deliver the heat to buildings (without ORC)
Other: …
Identifying whether or not firms that installed ORCs also had other options.
ALTERNATIVES Choice ORC
Display only if ‘yes’ to question 9. a).
b) What were the main reasons to choose for the ORC system, and not for the alternative(s)?
Rank the following options in order of importance by typing numbers (1, 2, 3, ..) into the boxes. Choose as many options as you want, leave the unimportant options blank.
We wanted to produce electricity
Better return on investment
More financial support (subsidies, green/white certificates, tax reductions, …)
The time required to install the ORC was shorter than the alternatives
The simplicity of the ORC
Other: …
Identifying why firms invest in ORCs when there are alternatives.
APPENDIX C
242
C.2.4 Timing CATEGORY QUESTION ANSWERS GOAL
TIMING Purchase *
10. When did you purchase the ORC system (i.e. when was the order signed)?
Indicate the year when the ORC was purchased (indicating the month is optional).
Year: …
Month: …
Dating the purchase.
TIMING Start operation *
11. a) Has the ORC system started operating?
Indicate whether the ORC started operating or not. If yes, indicate the year when the ORC was started (indicating the month is optional).
Yes, the system was started in (month/year): …
No, the ORC is not in operation yet.
Identifying the construction time.
TIMING End operation *
Display only if ‘yes’ to question 11. a).
b) Is the ORC system still in operation?
Select from the following options (only one answer possible).
Yes
No, the ORC is no longer in use. The operation was ended in (year):
Knowing whether the system is still in use
Display only if ‘no’ to question 11.b).
c) Why was the operation ended?
TEXTBOX Identifying reasons why ORCs stop operating.
SURVEY FROM ORC END-USERS
243
C.2.5 Power output and performance CATEGORY QUESTION ANSWERS GOAL
POWER Capacity * POWER Net power output
12. What is the power output of the ORC system?
Example: in case your system has an installed capacity of 300 kW, but is composed of two 150 kW modules, you can indicate this either as '300' or as '2 * 150'.
Indicate the nominal power of the ORC and, if available, indicate the net power output (Please answer in kW).
Nominal power output (gross installed capacity) [in kW]: …
Net power output on data sheet [in kW]: …
Achieved (measured) net power output [in kW]: …
Knowing the installed capacity. Identifying the difference between the installed capacity and the actual net output.
PERFORMANCE 13. Is the net power output (electricity generation) of the ORC in line with your expectations?
Select from the following options (only one answer possible).
Yes
No, it is lower than expected
No, it is higher than expected
Other: …
Revealing customer satisfaction on the actual power output.
PERFORMANCE 14. Are you, in general, satisfied with the operation of the ORC?
Yes, the ORC system works as it should.
No, there were/are some issues (e.g., start-up, interruption with other processes, leakage…): …
Identifying practical problems when the ORCs are in use.
APPENDIX C
244
C.2.6 Investment costs and subsidies CATEGORY QUESTION ANSWERS GOAL
CAPITAL COSTS *
15. How much did you pay for: Indicate the investment costs per category. Indicate at least either a) the ORC module costs; or d) the complete ORC project costs. Indicating b) the integration costs; c) other project costs is optional. Don't forget to specify the currency. (Drawing of the definitions included)
Identifying the investment costs for ORC systems.
a) the ORC module itself? … b) the integration of the ORC module? … c) other project costs? … d) the complete ORC project? … e) other: … Specify the currency: …
POLICY Investment support *
16. Did you get financial subsidies for the investment costs of the ORC?
Example: The government wants to encourage investments in ORCs and offers financial support for the investment. However, the subsidy can only be used for the investment costs of the ORC Module, and not for the complete ORC project.
Select the entries that received subsidies (more than one answer possible) . If possible, specify the amount and the unit in the textbox (e.g., 10000 €, 10%, ...). a) Yes, for the investment costs of the ORC module itself: … b) Yes, for the investment costs for integration of the ORC
module: … c) Yes, for the other project costs … d) Yes, for the investment costs of the complete ORC
project: … e) Yes, other: … f) No, we received no subsidies for the investment costs of
the ORC system
Identifying whether ORCs get investment subsidies and, if possible, how much.
POLICY Production support *
17. Did/do you receive subsidies for the production (power, heat) of the ORC system?
Example: The government
Select from the following options (more than one answer possible).
Yes, green certificates.
Yes, white certificates.
Yes, other: …
Identifying whether ORCs are subsidized for their green electricity production, for their contribution in energy efficiency,
SURVEY FROM ORC END-USERS
245
wants to encourage green electricity production. You get one green certificate per MWh electricity you produce with the ORC.
No, the production of the ORC was not subsidized. ….
POLICY Fiscal support *
18. Did/do you get fiscal advantages for installing the ORC system?
Example: a reduction on the corporate income taxation.
Select from the following options (only one answer possible).
Yes.
No.
I don’t know.
Identifying whether governments grant fiscal advantages for installing ORC systems.
APPENDIX C
246
C.2.7 Annual costs CATEGORY QUESTION ANSWERS GOAL
ANNUAL COSTS Contract
19. Do you have a contract with your ORC supplier for annual maintenance of the ORC system?
Select from the following options (only one answer possible).
Yes
No
Identifying how many ORCs have a maintenance contract.
ANNUAL COSTS Maintenance
20. How much do you spend annually on maintenance of the ORC system?
If possible, specify the amount and the unit (e.g., €, $, ...).
Amount per year: …
Currency: …
Identifying how much is paid annually for maintenance of the system.
ANNUAL COSTS Other
21. Are there other annual costs for the ORC system?
Select from the following options (more than one answer possible).
No, there are no other annual costs for the ORC
Working fluid: …
Safety: …
Operating the system (labor hours): …
Environmental regulation: …
Other:…
Identifying other annual costs.
OTHER COSTS Replacement
22. Did you encounter any costs to replace parts of the ORC system?
Select from the following options (only one answer possible).
No, there has been no need to replace parts
No, the replacement costs are included in the maintenance contract with the supplier
Yes: …
Other: …
Identifying which equipment is prone to break down.
SURVEY FROM ORC END-USERS
247
C.2.8 Closure
Thank you very much for completing this survey! On this final page, you can leave comments and your contact details.
Comments 23. Do you have any suggestions or comments?
TEXTBOX
Contact * 24. Can we contact you in case we want to discuss your case more in detail?
Yes. You can contact me as follows: …
No, rather not
Results * You can leave you contact details here in case you are interested in a brief summary of the results of this study. Your contact details will not be used for other purposes.
TEXTBOX
SATISFACTION In general, how satisfied are you with the purchase of the ORC?
Bar scale.
