EMSO: Environment for Modeling, Simulation and Optimization of
Biorefineries
Solutions for Process Control and Optimization
COPPE/UFRJ
11th World Bioenergy Symposium – WBS 2018
Argimiro R. Secchi
Chemical Engineering Program – COPPE
Universidade Federal do Rio de Janeiro
Technological Center, Rio de Janeiro – RJ, Brazil
Rio de Janeiro, June 19th, 2018
3
Founded in 1963
First department of COPPE, origin of the Institute
First to offer Graduate Courses in Engineering in Brazil
Chemical Engineering Program
COPPE - UFRJ
2018
4
Kinetics and Catalysis
Modeling, Simulation and Control of Chemical Processes
Biotechnology and Environmental Processes
Membrane Separation Processes
Polymer Engineering
Applied Thermodynamic
Thermofluidynamics
Interfacial Phenomena
Chemical Engineering Program
Research Areas
6
OutlineOutline
• Motivation
• Environment for Modeling, Simulation and Optimization
• Biorefinery Modeling
• Biorefinery Simulation
• Biorefinery Optimization for 1G+2G Bioethanol Production
• Final Remarks
EMSO: Environment for Modeling, Simulation and Optimization of Biorefineries
7
High dependence on fossil fuels
Source: IEA, 2013
Growing energy needs
World total final consumption by
fuel (2011)
World total final consumption by
fuel (2011)
Energy consumption by sector
Energy consumption by sector
Source: BP, 2014
Energy demand increases 41% between 2012 and 2035
Increasing GHG emissions Opportunities for renewable resource
Waste and wastewater,
3%Energy supply,
26% Transport, 13%
Residential & Commercial
buildings , 8%
Industry, 19%
Agriculture, 14%
Forestry, 17%
Global GHG emissions by source (2004)
Global GHG emissions by source (2004)
Source: IPCC, 2007; EPA, 2012
U.S., 13300,
57%
Brazil, 6267, 27%
Europe; 1371; 6%
China; 696; 3%
India; 545; 2%
Canada; 523; 2%
ROW; 727; 3%
Source: USDA-FAZ, 2014
Global ethanol production 2013
(million gallons)
Global ethanol production 2013
(million gallons)
Total Emissions in 2012 = 6,526Million Metric Tons of CO2 equivalent
MotivationMotivation
• Angarita et al. (2015), Biochemical Engineering Journal, 104, 10-19.
8• DOE (2016). Bioenergy Technologies Office: Multi-year program plan – Energy Dept., USA.
Global production of biofuels Industrial routes for conversion of biomass
MotivationMotivation
9
SHF: Separate Hydrolysis and fermentation. Low efficiency. Older technology.SSF: Prevents enzyme inhibition by product. Disadvantages in operating conditions (≠ optimal cond.)SSCF: Only one bioreactor.CBP: Only one microorganism and one bioreactor.
• Cardona et al., (2010). Bioresource Technology, 101, 4751-4766.
Many process synthesis and design alternatives
MotivationMotivation
Ionic liquid Extrusion Milling
10
batch continuous
FinFout
fedbatch
Fin
continuous with product separation
Fin
Fout
FinFout
continuous tubular ... and many others!
fluidized bed
MotivationMotivation
Several equipment types and operating modes
11
MotivationMotivation
Opportunities for process optimization
• Furlan et al. (2015), Computer Aided Chemical Engineering, 37, 1349-1354.
Need Computer Aided Process Engineering (CAPE) Tools!
12
tools
applications
Environment for Modeling, Simulation and OptimizationEnvironment for Modeling,
Simulation and Optimization
PSE CAPE
13
Environment for Modeling, Simulation and OptimizationEnvironment for Modeling,
Simulation and Optimization
EMSO stands for “Environment for Modeling, Simulation, and Optimization”
Development started in 2001 (by Rafael P. Soares), written in C++ language
Available in Windows and Linux
Models are written in an object-oriented modeling language
Equation-oriented simulator and optimizer
Computationally efficient for dynamic and steady-state simulations
Continuous improvements through ALSOC project:
http://www.enq.ufrgs.br/alsoc
14
Thermodynamic andPhysical Properties – Plugin
Thermodynamic andPhysical Properties – Plugin
Data bank with about 2000 pure compounds
Mixture properties calculation
Ideal GasRK
SRKPR
APRASRKCPA
GERG2008Ideal Liquid
WilsonNRTL
UNIFACUNIQUAQ
F-SAC. . .
Thermodynamic models:
15
All equipments or modules are simultaneously evaluated(Block decomposition can be used to explore sequential solution)
Open-source Modeling Equipment contain only chemistry
and physics of the model
Equation-Oriented SimulatorEquation-Oriented Simulator
ex: EMSO, Ascend, Jacobian, gPROMS, AspenDynamics, EcosimPro
16
A process flowsheet model can be hierarchically decomposed:
Plant
Separa
tion
Sys
tem
Pretreat. System
Reaction System
Separation System
Colu
mn 1
Colu
mn
2
Colu
mn
3
Column
Feed Tray
Linked Trays
Linked Trays
Condenser
Splitter
Pump
Rebolier
Linked Trays
Tray
Tray
Tray
Tray
Object-Oriented ModelingObject-Oriented Modeling
17
Tray
mass balance
energy balance
thermodynamic equilibrium
mol fraction normalization
ab
stra
ct m
od
el
liquid flow model
vapor flow model con
cre
te m
od
el (
ide
al t
ray)
efficiency model
con
cre
te m
od
el (
rea
l tra
y)
Object-Oriented ModelingObject-Oriented Modeling
18
• ABACUSS / JACOBIAN (Barton, 1999)
• ASCEND (Piela, 1989)
• Dymola (Elmqvist, 1978)
• EcosimPro (EA Int. & ESA, 1999)
• EMSO (Soares and Secchi, 2003)
• gPROMS/Speedup (Barton and Pantelides, 1994)
• Modelica (Modelica Association, 1996)
• ModKit (Bogusch et al., 2001)
• MPROSIM (Rao et al., 2004)
• Omola (Andersson, 1994)
• ProMoT (Tränkle et al., 1997)
Examples of general-purpose object-oriented modeling languages:
Object-Oriented ModelingObject-Oriented Modeling
19
• Separation Systems– Dynamic Flash
– Steady-State Flash
– Dynamic Condenser
– Steady-State Condenser
– Dynamic Reboiler
– Steady-State Reboiler
– Partial Reboiler
– Equilibrium Stage - Tray
– Splitter
– Mixer
– Cylindrical Tank
– Horizontal Cylindrical
– Column Section
– Distillation Column with Dynamic Condenser and Reboiler
– Distillation Column with Thermosyphon Reboiler and Sub-cooling
– Distillation Column with Thermosyphon Reboiler and Dynamic Condenser
– Distillation Column with Kettle Reboiler and Sub-cooling
– Rectifier Column
– Rectifier Column with Sub-cooling
– Stripping Column with Reflux
– Stripping Column with Sub-cooled Reflux
– Absorption Column with Reflux
– Absorption Column with Sub-cooled Reflux
– Stripping Column with Kettle Reboiler
– Stripping Column with Thermosyphon Reboiler
– Absorption Column with Kettle Reboiler
– Absorption Column with Thermosyphon Reboiler
• Controllers –PID Controllers (series, parallel, AW, AWBT)–Incremental PID Controllers (series, parallel, AW, AWBT)–Lead-Lag, Lag–Comparator, Sum, Ratio, Multiply, HiLoSelect–IAE –ISE
• Heat Exchangers –Simplified Shell-Tube Heat Exchanger –Rigorous Shell-Tube Heat Exchanger –Discretized Shell-Tube Heat Exchanger –Multi-Streams Heat Exchanger - MHeatex–Heat Exchanger – Heater and Cooler –Double Pipe Heat Exchanger–Plate Heat Exchanger
• Reactors–CSTR –PFR–Gibbs–Equilibrium–Batch–Fed Batch
• Pressure Changers –Pumps–Turbines–Compressors–Valves
• Didactic Models –Fogler’s book Exercises
EML – OO Model LibraryEML – OO Model Library
Petrochemical-oriented model library (2001 – 2012):
20
Dynamic and steady-state simulation
Steady-state optimization (NLP, MINLP)
Dynamic and steady-state parameter estimation
Steady-state data reconciliation
Process monitoring and inferences with OPC communication
Build bifurcation diagram (interface with AUTO for DAEs)
Sensitivity analysis and case study (surface response)
Linearization of nonlinear dynamic system
State estimation and model updating (EMSO-CEKF)
Dynamic simulation with SIMULINK/SCICOS (interface with MATLAB/SCILAB)
Add new solvers (DAE, NLA, NLP)
Add external routines using the Plugins resource
What can I do with EMSO?What can I do with EMSO?
21
Biorefinery ModelingBiorefinery Modeling
EMSO as platform for Sugarcane Virtual Biorefinery (since 2012)
22
Biorefinery ModelingBiorefinery Modeling
# Component # Component
1 water 17 HMF (hydroxymethylfurfural)
2 sucrose 18 glycerol
3 glucose 19 unknown sugars
4 xylose 20 MEG (ethylene glycol)
5 ethanol 21 sulfuric acid
6 CO2 22 phosphoric acid
7 CO 23 impurities
9 oxygen 24 cellobiose
9 nitrogen 25 ammonium hydroxide
10 hydrogen 26 cellulose
11 methane 27 hemicellulose
12 ammonia 28 ash
13 lignin 29 enzyme
14 xylan 30 yeast
15 acetic acid 31 calcium hydroxide
16 furfural 32 calcium phosphate
Physico-chemical properties
Many chemical species not present in petrochemical-oriented database
VRTherm
23
Biomodel
Accessories
Heat exchangers
Mixers and splitters
Pressure changers
Reactors
Separators
Flowsheets
1. assumptions2. energy stream3. main stream4. water stream
1. duplicator and selector2. tank3. valve
1. heater2. water heater3. heat exchanger
4. hybrid heat exchanger5. water heat exchanger
1. mixer and splitter2. water mixer and splitter3. hybrid mixer
4. splitter heat5. splitter power
1. compressor2. isenthalpic valve3. pump
4. water pump5. turbine6. steam turbine
1. boiler2. digester3. enz. hydrolysis
4. fermenter5. liming tank6. phosphate tank
7. pre-treatment8. stoic. reactor
1. centrifuge2. cleaning3. column4. decanter5. dry cleaning
6. evaporator7. filter8. flash9. water flash10. mill
11. separator12. sieve13. absorption tower
1. Examples of flowsheets
Biorefinery ModelingBiorefinery Modeling
Biorefinery Model Library
25• Furlan et al. (2012), Computers and Chemical Engineering, 43 , 1–9.
Biorefinery SimulationBiorefinery Simulation
1G + 2G ethanol-from-sugarcane production plant
27
Biorefinery OptimizationBiorefinery Optimization
• Carpio et al. (2017), Computer Aided Chemical Engineering, 40, 2065-2070.
28
Biorefinery OptimizationBiorefinery Optimization
Retro-Techno Economic Analysis
Rea
ctio
n yi
eld
iso-economics NPV = 0
Bio
cata
lyst
yie
ld
iso-economics NPV = 0
• Furlan et al. (2016), Industrial & Engineering Chemistry Research, 55 (37), 9865–9872.
29
• CAPE tools are available for bioprocess synthesis and design;
• Simulation and optimization of biorefineries illustrate the potential of the developed model library;
• Physico-chemical properties database need to be enlarged;
• The power and availability of computer hardware and software have increased our ability to model complex phenomena in biochemical processes. In fact, we are probably limited now more by what can or cannot be measured experimentally, than by techniques for solving equations;
• Particular importance are model-based state estimation techniques which compensate for the scarcity of online sensors for bioprocesses.
Final RemarksFinal Remarks
30
TeamTeam
ProfessorsAntônio J.G. Cruz (UFSCar)Argimiro R. SecchiCaliane B.B. Costa (UEM)Elba P.S. Bon (Bioetanol)Felipe F. Furlan (UFSCar)Frederico W. TavaresKese P.F. Alberton, EQRoberto C. Giordano (UFSCar)Maurício B. de Souza Jr.Príamo A. Melo Jr.Tito L.A. Moitinho
Engineers:Bruno L. NogueiraDasciana Rodrigues (Embrapa)Jurgen L. BregadoRicardo S. Teixeira (Bioetanol)Rodrigo R.O. Barros (Bioetanol)Rossano Gambetta (Embrapa)
Secretariat: Rosemary Cezar
PhD students:Alex F. TeixeiraAtaíde S. AndradeCaio F. C. MarcellosDaniel M. ThomazEliza H. C. ItoFelipe C. CunhaJavier A. AngaritaJeiveison G. S. S. MaiaLeonardo D. RibeiroMaria Rosa T. GoesRafael B. DemunerReinaldo C. SpelanoRoymel R. CarpioSergio A. C. GiraldoThamires A. L. Guedes
Post-Docs:José Mauel G. T. PerezLeonardo S. SouzaSimone C. Miyoshi
MSc students:Allyne M. dos SantosAndré F. F. SouzaMaría Jimena F. QuagliataMariana CarvalhoMariana K. MoroMario G. Neves Nt.Otávio F. IvoPedro C. N. FerreiraThiago C. d’ÁvilaVitor P. Paixão
Undergrad. students:Bruno BezCarlos M. M. FonsecaIsabella Q. SouzaLucas F. BernardinoLucas MarquesPedro DelouSilvio Cisneiros Nt.Thales S. M. GamaVictor C. Gomes
32
... thank you for your attention!
Process Modeling, Simulation and Control Lab
• Prof. Argimiro R. Secchi, D.Sc.
• Phone: +55-21-3938-8307
• E-mail: [email protected]
Prof. Maurício B. de Souza Jr., D.Sc.
• Phone: + 55-21-3938-7315
• E-mail: [email protected]
• http://portal.peq.coppe.ufrj.br/index.php/areas-de-pesquisa/modelagem-simulacao-e-processos
http://www.enq.ufrgs.br/alsoc
Solutions for Process Control and Optimization
COPPE/UFRJ
33
ReferencesReferences• Barton, P.I., Pantelides, C.C., 1994, The modeling of combined discrete-continuous processes, AIChE J., 40, 966–979.
• Bogusch, R., Marquardt, W. 1997. A formal representation of process model equations. Comp. & Chem. Engng. 21 (10) 1105-111.
• Cardona, C.A., Quintero, J. A., Paz, I. C., 2010. Production of Bioethanol from Sugarcane Bagasse: Status and Perspectives, Bioresource Technology, 101, 4754-4766.
• Carpio, R.R., Giordano, R.C., Secchi, A.R., 2017. Enhanced Surrogate Assisted Global Optimization Algorithm Based on Maximizing Probability of Improvement Method, Computer Aided Chemical Engineering, 40, 2065-2070.
• EA International, ESA. 1999. EcosimPro ver. 3.0: Getting started, users manual, modeling language (EL), modeling and simulation guide, and mathematical algorithms. Madrid, Spain: EA International.
• Elmqvist, H., Bruck, D., Otter, M. 1999. Dymola: Dynamic modeling laboratory: User’s manual. Version 4.0. Lund, Sweden: Dynamic AB.
• Elnashaie, S.S.E.H., Chen, Z., Garhyan, P., Prasad, P., Mahecha-Botero, A.. 2006. Practical Implications of Bifurcation and Chaos in Chemical and Biological Reaction Engineering. Int. J. Chemical Reactor Engineering, 4, 1-41.
• Furlan, F.F., Costa, C.B.B., Fonseca, G.C., Soares, R.P., Secchi, A.R., Cruz, A.J.G., Giordano, R.C., 2012. Assessing the Production of First and Second Generation Bioethanol from Sugarcane through the Integration of Global Optimization and Process Detailed Modeling, Computers and Chemical Engineering, 43, 1–9.
• Furlan, F.F., Giordano, R.C., Costa, C.B.B, Secchi, A.R., Woodley, J.M., 2015. Process Alternatives for Second Generation Ethanol Production from Sugarcane Bagasse, Computer Aided Chemical Engineering, 37, 1349-1354.
• Furlan, F.F., Costa, C.B.B., Secchi, A.R., Woodley, J.M., R.C. Giordano, 2016. Retro-Techno-Economic Analysis (RTEA): using (bio)process systems engineering tools to attain process target values, Ind. & Engineering Chemistry Research, 55 (37), 9865–9872.
• Angarita, J.D., Souza, R.B.A., Cruz, A.J.G., Biscaia Jr., E.C., Secchi, A.R., 2015. Kinetic Modeling for Enzymatic Hydrolysis of Pretreated Sugarcane Straw, Biochemical Engineering Journal, 104, 10-19.
34
ReferencesReferences
• Modelica Association. 1996. Modelica: A unified object-oriented language for physical systems modeling: Tutorial, rationale and language specification. Retrieved from http://www.modelica.org.
• Piela, P.C. 1989. ASCEND:Anobject-oriented environment for modeling and analysis. Ph.D. diss., Engineering Design Research Center, Carnegie Mellon University, Pittsburgh, PA.
• Rao, R.M., Rengaswamy, R., Suresh, A.K. and Balaraman, K.S. 2004. Industrial Experience with Object-Oriented Modelling FCC Case Study. Chem. Engng. Res. & Des., 82 (A4) 527–552.
• Rodrigues, R., Soares, R.P. and Secchi, A.R. 2010. Teaching Chemical Reaction Engineering Using EMSO Simulator. Computer Applications in Engineering Education, 18 (4) 607-618.
• Soares, R.P. and Secchi, A.R. 2003. EMSO: A New Environment for Modeling, Simulation and Optimization. ESCAPE 13, Lappeenranta, Finlândia, 947 – 952.
• Soares, R.P. and Secchi, A.R. 2005. Direct Initialisation and Solution of High-Index DAE Systems, ESCAPE 15, Barcelona, Spain, 157–162.
• Valle, E.C., Soares, R.P., Finkler, T.F. and Secchi, A.R. 2008. A New Tool Providing an Integrated Framework for Process Optimization, EngOpt 2008 - International Conference on Engineering Optimization, Rio de Janeiro, Brazil.