Bidding Behavior Simulation For Electricity Market Participants Based On Intelligent Agent

The construction and operation of electricity markets is a complicated system project.At the same time,the operation of electricity market is closely related with power system operation,especially it is inseparable with power dispatching and operation.Electricity markets plays an important role in safe operation of power grid.The design of different electricity market modes has a decisive influence on the efficiency of market operation,revenue of the market participants,and stable operation of power system.So the designers of electricity markets always desires to carry out a forward-looking study in order to reduce the cost of trial and error before putting the markets into operation.In simulation of electricity markets,not only needs to implement market rules and market models,but also needs to simulate the decision-making behavior of market participants.But traditional modeling method is not applicable to current decision-making behavior simulation of electricity market participants.At the same time,the decision-making behavior of the market participants will change dramatically from planning to market competition.The decision-making behavior will be influenced by factors such as cost and risk preference of market participants,and supply and demand relationship of electricity market and so on,so we need to carry out research on the bidding behavior of electricity market participants.In this paper,the agent technology is introduced into bidding behavior simulation model of market participants.Based on the analysis of related factors that affect the bidding decision of market participants,the paper refines the key influencing factors and builds a model that can simulate bidding behavior of market participants.Multi-decision factors are considered in the bidding model,which include comprehensive cost and decision objects of market participants,and market clearing rules,offer and risk preference of generators.Validity of the model is verified by an example.Finally,this paper use the key indexes of electricity markets to evaluate the bidding behavior of market participants.The main work and innovations are as follows:(1)Through analysis of comprehensive cost of market participants,a comprehensive cost model adapted to the electricity markets experiment in day ahead market is established,which provide basis for bidding behavior simulation of market participants.(2)The key factors are extracted from factors that affect bidding behavior of generators,and bidding model,which consider multi-input factors,of generators is build.Then intelligent agent technology and reinforcement learning algorithm are introduced into the model,and multi-input decision factors algorithm based on intelligent agent is used to simulate bidding behavior.Bidding behavior and dynamic behavior evolution process of generators under changing condition are simulated,and validity of the model is verified by an example.(3)This paper verifies validity of the model by simulating a single-day electricity markets operation.The intelligent algorithm can simulate the bidding behavior of generators under continuous market operation.Then,the influence of forward contracts on bidding behavior of generators in day ahead electricity market is compared and analyzed,and this paper use key indexes of electricity markets to evaluate bidding behavior of electricity market participants.