EU CRISP project/ PowerMatcher Active demand response and production control for power system balance René Kamphuis, ECN/Intelligent Grids

Transcription

1 EU CRISP project/ PowerMatcher Active demand response and production control for power system balance René Kamphuis, ECN/Intelligent Grids

2 Dutch MP Paul de Krom Wind energy: what is possible? EWEA Chairman Arthouros Zervos An electricity system based on wind energy is impossible. If there is no wind, I wouldn t be able to watch TV and I couldn t drink a cold beer

3 It is not wind or solar, but it is the system The system, it is changing So, the way the system is controlled needs to be changed too

4 The Challenge of the Changing System Traditional situation: Demand Behaves stochastically Partially predictable Supply Large scale production Fully controllable Control Centralised Done by the supply side Future situation: Demand Behaves stochastically Partially predictable Supply Behaves stochastically Partially predictable Control Challenges How to maintain the supply & demand balance? Who delivers system support services?

5 The characteristics of the future Smart Grid All connections actively participate in managing the electricity grid New (real time) transparent markets for Ancillary Services (voltage stabilisation, power quality, reactive power, blackstart, etc.) exist in addition to existing markets for real power. These markets are accessible for everybody and every grid connection Technology required: real time metering and distributed control. Allows for real time pricing Parts of the grid can be decoupled from main grid intentionally, functioning autonomously and able to reconnect later. (Network of networks)

6 ECN s contribution to a new control approach for a new system Automatic matching of demand and supply in a system Based on novel ICT technologies Developed in the CRISP project: Further developed in other projects Can be used for different applications Reduction of imbalance (topic of this presentation) Optimal Dispatch of demand and supply of energy trader Creation of Virtual Power Plant Reduction of peak load in the system Reduction of stress on local MV/LV transformer stations

7 PowerMatcher overview

8 microeconomics, control theory & agents p r r x x set price goods inputs output Microeconomics Control Theory price resources goal state Multi Agent Theory

9 PowerMatcher structure

10 PVP 2.5 MW Wind Turbine Park I Local CRISP Node Central CRISP Node Cold Store 1,5 MW 8 MW Wind Turbine Park II Local CRISP Node Data Communications Network Local CRISP Node Local CRISP Node Emergency Generator 200 kw 6 MW Local CRISP Node Local CRISP Node 0.8 kw Residential Heat Production (CHP) ECN Test Dwelling Scaled up by simulation to 800 kw

11 The nodes

12 Balancing Mechanism in the Dutch liberalised market: Electricity Market Parties: daily plans for production, transport and consumption ( Program Responsibility ) The Transmission System Operator: Buys real time power from Market Parties to balance the system. larger market parties are obliged to offer a part of their production to the TSO on this market. price per MWh set every 15 minutes (imbalance price). Equal for every offer, set by highest offer taken Prices are announced few days afterwards Each party having a deviation in the wrong direction pays the imbalance price for every MWh to TSO. Each party having a deviation in right direction receives imbalance price for every MWh to TSO Balancing Responsible Parties: wind energy is faced with extra imbalance costs due to partly predictable wind energy production.

13 Kreileroord (windfarm) outcome

14 IRS Market schedules

15 Market outcome one turbine 43.5 % imbalance reduction

16 Wind imbalance vs. marketprice

17 Test dwelling outcome

18 Imbalance Reduction System results: Lessons learned Principle is feasible: shown by real time operation of heat pump 40 50% imbalance reduction shown (100% possible with better cluster, improved flexibility of devices and improved agent algoritms) Cluster composition stays important: Needs of cluster composition might differ over the year during different seasons. Local behavior stays within process constraints

19 More general conclusions PowerMatcher principle works in practice (Proof of Feasibility) An important electricity market barrier (imbalance by imperfect prediction) for Wind is solved, in principle The PowerMatcher can be used for other barriers to Wind and Solar as well Commercialization of this technology is considered Of course there is still research work to be done, you will hear from us To showcase the working of the principle we have developed the Elektra Game If there is no wind, make sure you don t need it!

20 µchp DSO Virtual Power Plant... Local VPP Node Local VPP Node Local VPP Node Local VPP Node GPRS Wireless Communication PowerMatcher VPP Controller

21

22 Demand, price forming and VPP supply (first 5 micro CHP connected)

23 Concluding remarks PowerMatcher Principle works in real life situations,also for micro CHP Local conditions stay within local boundaries There is no complaint from residents If communication fails, devices just go on as stand alone micro CHPs PowerMatcher will be refined and further developed in on going and coming projects (national as well as EU funded) These kind of solutions are essential for integrating large scale micro CHP and renewables in the electricity system

24 Concluding remarks PowerMatcher Principle works in real life situations,also for micro CHP Local conditions stay within local boundaries There is no complaint from residents If communication fails, devices just go on as stand alone micro CHPs PowerMatcher will be refined and further developed in on going and coming projects (national as well as EU funded) These kind of solutions are essential for integrating large scale micro CHP and renewables in the electricity system

25 End of show Questions Discussion