The Influence of Participation in Ancillary Service ...€¦ · The Influence of Participation in Ancillary Service Markets on Optimal Energy Hub Operation Foteini Rafaela TSAOUSI

Donnerstag, 14. September 2017 1Foteini Rafaela Tsaousi - Screening of a Lifecycle Assessment (LCA) of and Advanced Adiabatic

Compressed Air Energy Storage (AA – CAES)

3 R D I N T E R N A T I O N A L C O N F E R E N C E O N

S M A R T E N E R G Y S Y S T E M S A N D 4 T H G E N E R A T I O N D I S T R I C T H E A T I N G

The Influence of Participation in Ancillary Service Markets on Optimal Energy Hub Operation

Foteini Rafaela TSAOUSI

Co-authors: Dr. Andrew BOLLINGER, Conor O‘ MALLEY, Prof. Gabriela HUG

Copenhagen, 12.09.2017

Outline

Donnerstag, 14. September 2017 2Foteini Rafaela Tsaousi - The influence of participation in ancillary service markets on optimal energy

hub operation

Introduction – Energy Hubs, Ancillary Services and Goal

Model – Assumptions and Objective Function

Evaluation – Assessment of Stochastic Modelling

Case Studies – Parameter Sensitivity Analysis

Conclusion

Outlook

Introduction – Energy Hubs (EH)


hub operation

Conceptual approach to describe District Multi Energy Systems (DMES)

Introduction – Ancillary Services (AS)


hub operation

Tertiary Control Reserve

Source: Survey on Ancillary Services Procurement, Balancing Market Design 2015, ENTSO-E, 2016

Introduction – Goal


hub operation

Model development to include AS in the EH modelling tool

Model development that determines the optimal bidding strategy

Investigation of important parameters that affect the EH operation

Investigation of possible profitable scenarios under AS participation

Outline


hub operation





Conclusion

Outlook

Model – Products and Time Frame


hub operation

Positive and Negative TRL

Week-ahead and Day-ahead products

Optimization horizon: 1 week

Simulation horizon: 1 year

Model – MILP


hub operation

Model – Two-Stage Stochastic MILP


hub operation

Model – Objective Function (Total Cost)


hub operation

NomenclatureConversion Technologies

Week or Day Ahead

Positive or Negative

Time

Energy Carrier

First Stage

Income for

Availability

Grid Input

Energy Cost

Price for Availability

Bidding Capacity

Input Energy

O&M Cost Coefficients

EH Conversion Coefficient

Model – Objective Function (Total Cost)


hub operation

Nomenclature

Conversion Technologies

Week or Day Ahead


Time

Energy Carrier

Scenarios

Second Stage

Cost of OperationIncome from

Activation

Price for Energy Binary Activation Variable

Outline


hub operation





Conclusion

Outlook

Evaluation – Base Case Comparison


hub operation

Definition of Base Case

Comparison of three different versions of Base Case:

Evaluation – Total Cost


hub operation

Total Cost [thous. CHF]

Perfect Forecast 1,612

Stochastic 1,627

No-Bidding 1,655

Outline


hub operation





Conclusion

Outlook



hub operation

Case Modified Parameter New Value

Base Case Value Unit

1 0.03

2 CHP Operating Cost 0.06 0.09 CHF/kWh

3 0.12

4 -0.08

5 TRE- Bid Price -0.02 0.08 CHF/kWh

6 0.02

Case Studies – Total Cost


hub operation

0.02

0.12

0.06

0.03

[CHF/kWh]-0.02

-0.08

[CHF/kWh]

No-BiddingBase Case

TRE- Bid Price

CHP Cost

Outline


hub operation





Conclusion

Outlook

Conclusion

Donnerstag, 14. September 2017 19Foteini Rafaela Tsaousi The influence of participation in ancillary service markets on optimal energy

hub operation

Model is capable of determining the optimal bidding strategy of the EH

Modeling evaluation shows that the uncertainty is captured well

Case Studies show Parameter Sensitivity

Small Margin for Profit

Outlook


hub operation

Model is extendable to include Primary, Secondary Control Reserve and

Control Pooling

Design Optimization

Capacity Market Uncertainty

Thank you for your attention!

21

3 R D I N T E R N A T I O N A L C O N F E R E N C E O N

S M A R T E N E R G Y S Y S T E M S A N D 4 T H G E N E R A T I O N D I S T R I C T H E A T I N G

Model (4/7) – Assumptions


hub operation

Capacity bid always accepted

All bidding decisions at the beginning of the week

No minimum bid size considered

Energy provision every hour

Perfect knowledge of EH demand and solar energy input

Model (7/7) – Uncertainty


hub operation

Nomenclature

,, TRE+ Market Scenario Price

,, TRE- Market Scenario Price

• Decorrelation technique

• ARIMA model

Time series of the TRE prices

Scenario generation

Evaluation – Day-Ahead, Average Bid Size


hub operation

Positive Control Negative Control

Case Studies – CHP Cost, Activation Income


hub operation

Positive Control Negative Control

CHP Cost [CHF/kWh]

Case Studies – TRE- Bid Price, Activation Income


hub operation

Negative Control

TRE- Bid Price [CHF/kWh]

Case Studies – Cost Deviation from No-Bidding


hub operation

37.6%

19.2%

-10.6% 1.5% 2% 2.3% 1.7%

Conclusion


hub operation

Model is capable of determining the income of TRL for EH

• Negative Control Bids are more frequent

• Day-Ahead Products are preferable

Case Studies show Parameter Sensitivity

• CHP cost has a great influence

• TRE- bid price is important for frequency of activation

Small Margin for Profit

Ancillary Services (AS) in Switzerland


hub operation

Frequency Control

Voltage Support

Compensation of active power losses

Black Start and island operation capability

System coordination

Operational measurement

Frequency Control


hub operation

CapacityPRL SRL TRL

Procurement Scheme Market Market Market

Minimum Bid size (MW) 1 5 5Product Resolution (time) Week(s) Week(s) Day(s)

Distance to Real Time of Auction Day(s) Day(s) Day(s)Settlement Rule Pay as Bid Pay as Bid Pay as BidCost Recovery Scheme Grid Users Grid Users Grid UsersMonitoring Real Time Real Time Real Time

Product Symmetric Symmetric Asymmetric

Tender Period Weekly Weekly Weekly or Daily

Capacity Provider Generators + Load + Pump Storage units pumping

Frequency Control


hub operation

EnergyPRL SRL TRL

Procurement Scheme - - Market

Activation Rule - Pro-Rata Merit Order

Minimum Bid size (MW) - - 5

Product Resolution (time) - - Hour or Blocks

Distance to Real Time of Auction - - Hour or BlocksSettlement Rule - Hybrid Pay as BidCost Recovery Scheme - BRP BRPMonitoring - Ex Post Check Ex Post Check

Product Symmetric SymmetricDaily Tenders for 4h-Blocks

Activation Time 30s 5m 15m

Partially Activated Product - - No

Volume of Control Power Required (MW) +/- 74 +/-400 -450,-300

Link Control Signal Control Signal Email, Call

Remuneration of Energy -According to Spot MarketPrice

For 4-h Block Offer and Energy Used

Model – Definition of Bidding Capacities


hub operation

, : Bidding Capacity Storage

, : Bidding Capacity from Generating Units


Week or Day Ahead


Time

Energy Carrier

Model – First Stage Constraints


hub operation


Week or Day Ahead


Time

Energy Carrier

Product Time Frame

Minimum Bid Size

Model – Second Stage Constraints


hub operation


Week or Day Ahead


Time

Energy Carrier

Positive Control

Negative Control

Model – Energy Market Constraints


hub operation

Positive Control

Negative Control

Evaluation - Results


hub operation

Day 310

CHP

Heat Pump

Day 359

Evaluation- Results


hub operation

Availability Income Activation Income

Evaluation- Results


hub operation

Availability Income Activation Income

Total Avail. Income [thous. CHF]

Perfect Forecast 17.8

Stochastic 5.9

No-Bidding 0

Total Activ. Income [thous. CHF]

Perfect Forecast -14.4

Stochastic -12.2

No-Bidding 0

Evaluation


hub operation

Scenario Forecast


hub operation



hub operation

Case Modified Parameter New Value

Base Case Value Unit

A 0.03

B CHP Operating Cost 0.06 0.09 CHF/kWh

C 0.12

D Heat Storage Capacity 0 1 MWh

E 2

F Heat Pump Capacity 0 2.2 MW

G 4

H -0.08

I TRE- Bid Price -0.02 0.08 CHF/kWh

J 0.02

Case Studies – Total Cost


hub operation

0.120.06

0.03

20

4

0.02-0.02-0.08

[CHF/kWh]

[MW]

[MW]

[CHF/kWh]

0

Case Studies – Availability Income


hub operation

CHP Cost Heat Storage

Bid PriceHeat Pump

Case Studies – Activation Income


hub operation

CHP Cost Heat Storage

Bid PriceHeat Pump

Case Studies – Cost Deviation from No-Bidding


hub operation

CaseCost Deviation to

No-Bidding(%)

A 37.6

B 19.2

C -10.6

D 1.1

E 1.9

F -3

G -22.5

H 1.5

I 2

J 2.3

Base Case 1.7

CHP Cost

Heat Storage Bid PriceHeat Pump

3 R D I N T E R N AT I O N A L C O N F E R E N C E O N

SMART ENERGY SYSTEMS AND 4TH GENERATION DISTRICT HEATING

C O P E N H A G E N , 1 2 – 1 3 S E P T E M B E R 2 0 1 7

The Influence of Participation in Ancillary Service Markets on Optimal Energy Hub Operation

Foteini Rafaela Tsaousi 12/09/2017

Co-authors: Dr. Andrew Bollinger, Conor O‘ Malley, Prof. Gabriela Hug

Documents

The Influence of Participation in Ancillary Service ...€¦ · The Influence of Participation in Ancillary Service Markets on Optimal Energy Hub Operation Foteini Rafaela TSAOUSI