Application of the EWA algorithm in electricity market simulation

作者：Jing, Zhaoxia^*; Yang, Ying

来源：Automation of Electric Power Systems, 2010, 34(24): 46-50.

• Agent-based computational economics
• Computational economics
• Electricity market simulation
• Intelligent agent model
• Learning models
• Market equilibria
• Market participants
• Market studies

Agent-based computational economics have become one of the most important electricity market study methods. Intelligent agent model is key part of such studies. Reinforcement learning and belief learning algorithms which are commonly used in the agent model all have their drawbacks. This paper introduces a combination of reinforcement learning and belief learning algorithm named experience-weighted attraction (EWA) algorithm to simulate the behavior of power plants in electricity market simulation. The simulation results based on pure-agent and mixed-agent system show that EWA learning model can describe the behavior of market participants better;and it is more advanced and intelligent than the Roth-Erev (RE) algorithm. Moreover, EWA algorithm also has a better ability to capture the market equilibrium.

• 出版日期2010-12-25
• 单位华南理工大学