基于自加强学习算法的发电商报价策略研究  

Strategic Bidding of the Electricity Producers Based on the Reinforcement Learning

作　　者：马豫超[1] 蒋传文[1] 候志俭[1] Ettore Bompard 王承民[1] 

机构地区：[1]上海交通大学电气工程系,上海市徐汇区200240 [2]意大利都灵理工大学电气工程系

出　　处：《中国电机工程学报》2006年第17期12-17,共6页Proceedings of the CSEE

基　　金：国家自然科学基金重点项目(50539140)~~

摘　　要：电力市场中发电商的决策过程和多发电商的相互作用过程是个复杂动态问题,很难用传统的解析方法进行分析计算,这在考虑中长时间段交易时尤为突出,且多代理作用机制是个很好的补充。文中提出了一个能够模拟发电商在市场中进行策略性报价的中长期交易时间段决策过程模型,对多发电商交互作用导致的市场行为进行了仿真试验。该模型基于自加强Watkins’sQ(λ)学习算法并包含了可能对电力市场运行产生重要影响的网络阻塞因素。该模型可以形成发电商的最优策略以最大化中长期生产效益和可以找到中长期市场平衡点并据此评估市场中长期运行情况。通过在标准IEEE-14节点系统中进行仿真计算表明该模型的有效性和新颖性。The decision making process of the electricity producers and their interactions in the market are a typical complex problem that is difficult to be modeled explicitly, and can be studied with a multi agents approach. This paper proposes a model being able to capture the decision making process of the producers in submitting strategic biddings to the market and simulate the market outcomes resulting from those interactions. The model is based on the Watkins＇s Q（λ） Reinforcement Learning and takes into account the network constraints that may pose considerable limitations to the electricity markets. The model can be used to define the optimal bidding strategy for each producer and, as well, to find the market equilibrium and assessing the market performances. The model proposed is applied to a standard IEEE 14-bus test system to illustrate its effectiveness.

关 键 词：电力市场 最优报价策略 多代理 自加强Watkins's q(λ)学习算法 

分 类 号：TM74[电气工程—电力系统及自动化]