Research On Virtual Power Plant Formation And Bidding Strategy Oriented To Frequency Regulation Mechanism

The virtual power plants,which are composed of intermittent distributed power generations,controllable distributed power generations,and energy storage devices,aggregate various distributed resources through advanced control and communication technologies to participate in power system operation and market transactions as a whole.It can solve the challenges brought by large-scale distributed generations to the safe and stable operation of the power system,and can also increase the opportunities for distributed generations to participate in the power market and increase its own revenue.With the continuous deepening of power reform in China,the participation of virtual power plants in the energy and frequency regulation market is gradually becoming possible.This paper studies the formation,bidding,clearing and control strategies of virtual power producers participating in the spot market considering the frequency regulation auxiliary service market mechanism.The details are as follows:(1)Propose a virtual power producers formation strategy with multiple resources.The randomness of intermittent power supply such as wind power and photovoltaic contained in the virtual power plant is characterized by probability form.The correlation of the output of intermittent power supply is analyzed and its Copula joint probability distribution model is constructed.Combined with the market access mechanism,the aggregation strategy is designed for the performance requirements of the frequency regulation market.Distributed resources are aggregated to form virtual power producers to participate in the energy and frequency regulation market through multi-agent polymerization.(2)Proposed stochastic joint bidding strategy for virtual power producers in the spot market.Based on the characteristics of multi-agent hierarchical control presented by virtual power producers,a hierarchical bidding model for virtual power generators participating in the energy and frequency regulation joint market is established.Taking into account the randomness of market price of electricity and the intermittent power output,the Latin Hypercube Sampling and scene reduction techniques are used for processing.The upper layer of the model is aimed at the maximum total bidding revenue of the virtual power producers.The lower layer is aimed at minimizing the power generation cost of each agent,and fully considers the correlation of intermittent power supply output.The upper and lower layers are jointly optimized to obtain the optimal bidding strategy of the virtual power producer.The simulation of the example verifies that the bidding model can provide a strong support for the virtual power producers to participate in the electricity market transactions,and analyzes the correlation of intermittent power supply.(3)Construct a spot market clearing mechanism that considers the risk of frequencyregulation performance.Combined with the day before market organization process,the trading center organizes the clearing after the bidding of each market entity.In order to ensure the quality of the system frequency regulation,the pCVaR index which measures the risk of frequency regulation is designed and the market clearing model considering the frequency regulation risk is established.The risk is avoided for the virtual power producer with the uncertainty of output.In order to refine the frequency regulation contribution provided by the frequency regulation supplier,and to more accurately measure the frequency regulation revenue,the concept of effective mileage is proposed.At the same time,an insurance mechanism that can realize risk transfer and encourage virtual power producers to participate in market enthusiasm is designed.The simulation of the example proves that the model can effectively avoid the frequency regulation risk of virtual power producers.(4)Construct a real-time optimal control model of virtual power producer based on model predictive control.Considering that the virtual power producer's bidding strategy is based on the previous forecast output of the internal power generation units,the prediction error cannot be ignored.In order to avoid punishment,the virtual power producers adjust the internal resources in real time according to the clearing information.The real-time rolling optimization control model of virtual power producers based on model predictive control is established and the rolling feedback optimization is optimized with the minimum deviation from the scheduling instruction.The simulation results show that the model can reduce the intermittent resource uncertainty in a timely and effective manner and meet the actual market operation needs of virtual power producers.