comparison of methods for thermal storage sizing in ...€¦ · 3rdinternational conference on...

3rd International Conference on Smart Energy Systems and 4th Generation District HeatingCopenhagen, 12‐13 September 2017

Comparison of methods for thermal storage sizing in district heating networks

Olatz TerrerosAIT Austrian Institute of Technology GmbH

1

Background

23rd International Conference on Smart Energy Systems and 4th Generation District Heating

Copenhagen, 12‐13 September 2017

Thermal storage technologies are a key component in the transition towards 100% renewable energy systems.

Increase integration of renewable and industrial waste energy sources into district heating networks.

Increase system flexibility through decoupling heat and electricity production/consumption.

Increase system stability.

• ObjectiveDevelopment of technical basis for increasing the share of decentralised alternativeheat sources, particularly industrial waste heat, heat pumps and solar thermalenergy in district heating networks.

• PeriodFrom 03.2015 to 02.2018

• FundsClimate and Energy Funds

• ProgrammeEnergieforschungsprogramm 2014

• FFG project number848849



Heat_portfolio

Background



GoalComparison of thermal storage sizing methods:

Simulation methods

Engineering methods Design rules from manufacturers

Mathematical methodsAssessment of consumers and producers profiles



GoalComparison of thermal storage sizing methods:

Engineering methods Design rules from manufacturers

Mathematical methodsAssessment of consumers and producers profiles

Simulation methods

Mathematical methods



• Rainflow counting algorithm determines the number of cycles present in a heat load profile.

• Fourier transformation decomposes a function of time (a signal) into frequencies.

• Minimum load boiler algorithmconsiders the heat load profile below theminimum load.

The following methodologies have been compared:

• Operation optimization model is developed.• The model is based on the mixed integer linear programming (MILP) method.• It provides the cost optimal storage volume and operation strategy.

7

Simulation method

3rd International Conference on Smart Energy Systems and 4th Generation District Heating Copenhagen, 12‐13 September 2017

Input Optimisation model Output

• Heat load• Power plant parameters• Storage parameters• Electricity prices• Fuel prices• Operation cost• CO2 factor

• Hydraulic constraints• Upper‐lower boundaries

• Minimized operational cost• Optimal storage volume

Plant 1Plant 2Plant 3Plant 4Plant 5Plant 6Plant 7

[V1,V2,V3,V4…]



A set of scenarios is developed based on representative Austrian district heatingnetworks.

Scenarios

FluctuationFluctuationShareShareFuture technologyFuture

technologyCurrent

technologyCurrent

technologyNetwork demandNetwork demand

SCENARIOS

Rural

Biomass boilerOil boiler

Gas boiler

Waste heat(high temp.) 5%

Low

Urban

CHPGas boilerOil boiler

Waste heat(low temp.)

25%

High

CHPIncineration

plantGas boiler

Solar thermal

50%

10

Scenarios – simulation results

0

500

1000

1500

2000

2500

Baseline constant 5% constant 25% constant 50% fluctuation 5% fluctuation25%

fluctuation50%An

nual heat p

rodu

ction

[MWh/year]

Biomass boiler [700 kW] Oil boiler [1500 kW] Waste heat

0.0

0.2

0.4

0.6

0.8

1.0

1.21

210

419

628

837

1046

1255

1464

1673

1882

2091

2300

2509

2718

2927

3136

3345

3554

3763

3972

4181

4390

4599

4808

5017

5226

5435

5644

5853

6062

6271

6480

6689

6898

7107

7316

7525

7734

7943

8152

8361

8570

Heat dem

and [M

W]

Time [h]

heat demand

constant 5%

constant 25%

constant 50%

fluctuation 5%

fluctuation 25%

fluctuation 50%

11

0.0

0.2

0.4

0.6

0.8

1.0

1.21

210

419

628

837

1046

1255

1464

1673

1882

2091

2300

2509

2718

2927

3136

3345

3554

3763

3972

4181

4390

4599

4808

5017

5226

5435

5644

5853

6062

6271

6480

6689

6898

7107

7316

7525

7734

7943

8152

8361

8570

Heat dem

and [M

W]

Time [h]

heat demand

constant 5%

constant 25%

constant 50%

fluctuation 5%

fluctuation 25%

fluctuation 50%

Scenarios – simulation results

0

20

40

60

80

100

0

500

1000

1500

2000

2500

Baseline constant 5% constant 25% constant 50% fluctuation 5% fluctuation25%

fluctuation50% He

atgene

ratio

ncost

[€/M

Wh]

Annu

al heat p

rodu

ction

[MWh/year]

Biomass boiler Oil boiler Waste heat Heat generation cost [€/Mwhyear]

4 m3

628 cycles4 m3

1369cycles6 m3

564 cycles4300 m3

1 cycle6 m3

884 cycles3800 m3

1 cycle15300 m3

1 cycle

12

Engineering methods

Source: [Wolff, 2011] D. Wolff und K. Jagnow, „Überlegungen zu Einsatzgrenzen und zur Gestaltung einer zukünftigen Fern‐ und Nahwärmeversorgung,“ delta‐q, Wolfenbüttel/Braunschweig, 2011.

1

10

100

1000

10000

1 10 100 1000 10000 100000 1000000

Stor

age

volu

me

[m³]

Total nominal thermal capacity [kWth] || Collector brutto area [m2]

Boilerv=80 [l/kWth]

Vacuum collector(CPC)v=60-70 l/m²

Boilerv=45 [l/kWth]

Flat collector (highperformance)v= 50 l/m²

Flat collector (standard)v= 40 l/m²

Boilerv=417 [l] + 26 [l/kWth]

CHPv=104 [l] + 10 [l/kWth]

1

10

100

1000

10000

1 10 100 1000 10000 100000 1000000

Stor

age

volu

me

[m³]


Boilerv=80 [l/kWth]

Vacuum collector (CPC)v=60-70 l/m²

Boilerv=45 [l/kWth]

Flat collector (highperformance)v= 50 l/m²

Flat collector (standard)v= 40 l/m²


CHPv=104 [l] + 10 [l/kWth]

Rural network (measureddata)

Sub-urban network(measured data)

Urban network(measured data)

13Source: [Wolff, 2011] D. Wolff und K. Jagnow, „Überlegungen zu Einsatzgrenzen und zur Gestaltung einer zukünftigen Fern‐ und Nahwärmeversorgung,“ delta‐q, Wolfenbüttel/Braunschweig, 2011.

Rural networks

Sub‐urban networks

Urban networks

Engineering methods – Austrian networks

1

10

100

1000

10000

1 10 100 1000 10000 100000 1000000

Stor

age

volu

me

[m³]


Boilerv=80 [l/kWth]

Vacuum collector (CPC)v=60-70 l/m²

Boilerv=45 [l/kWth]

Flat collector (highperformance)v= 50 l/m²Flat collector (standard)v= 40 l/m²


CHPv=104 [l] + 10 [l/kWth]

Rural network (measureddata)

Sub-urban network(measured data)

Urban network(measured data)

Simulation results

14Source: [Wolff, 2011] D. Wolff und K. Jagnow, „Überlegungen zu Einsatzgrenzen und zur Gestaltung einer zukünftigen Fern‐ und Nahwärmeversorgung,“ delta‐q, Wolfenbüttel/Braunschweig, 2011.

Rural networks

Sub‐urban networks

Urban networks

>25% renewableenergy integration

Engineering methods ‐ simulation methods

Conclusions



• Engineering methods

For baseline scenarios the calculated storage volumes are usually biggercompared to the results obtained from the simulation method.

Our study shows that V = 417[l] + 26 [l/kW] rule can be recommended for storage sizing in rural networks (peak storage 2‐3 hours).

Not suitable for storage sizing in scenarios with high integration of futurerewenable technologies.

• Simulation method

The results are validated with real scenarios and engineering methods. Holistic analysis of the network: consumers, power plants and storages. Further developments: yearly based optimisation would improve the

behaviour of seasonal storages.

3rd International Conference on Smart Energy Systems and 4th Generation District HeatingCopenhagen, 12‐13 September 2017

Thank you for your attention.

Olatz TerrerosAIT Austrian Institute of Technology GmbH

+43 664 [email protected]

16

comparison of methods for thermal storage sizing in ...€¦ · 3rdinternational conference on...

Documents