基于自加强Watkinss Q 算法的发电商报价策略行为.pdfVIP

26 卷 17 期 中 国 电 机 工 程 学 报 Vol.26 No.17 Sep. 2006 2006 年9 月 Proceedings of the CSEE 2006 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2006) 17-0012-06 中图分类号：TM74 文献标识码：A 学科分类号：47040 基于自加强学习算法的发电商报价策略研究 1 1 1 2 1 马豫超 ，蒋传文 , 候志俭 ，Ettore Bompard ，王承民 ( 1. 海交通大学电气工程系, 海市 徐汇区 200240；2. 意大利都灵理工大学电气工程系，意大利 都灵） Strategic Bidding of the Electricity Producers Based on the Reinforcement Learning 1 1 1 2 1 MA Yu-chao , JIANG Chuan-wen , HOU Zhi-ji an , Ettore Bompard , WANG Cheng-min ( 1. Department of Electric Power Engineering, Shanghai Jiao Tong University, Xuhui District, Shanghai 200240, China ； 2. Politecnico Di Torino Electrical Engineering Department I - 10129,Torino,Italy) ABSTRACT: The decision making process of the electricity 关键词：电力市场；最优报价策略；多代理； 加强Watkins’s producers and their interactions in the market are a typical q(l)学习算法 complex problem that is difficult to be modeled explicitly, and 0 引言 can be studied with a multi agents approach. This paper proposes a model being able to capture the decision making 研究发电商最优报价策略在国内外的电力市场 process of the producers in submitting strategic biddings to the [1-3] 文献中已成为热点问题 。依据传统最优化理论， market and simulate the market outcomes resulting from those 文献[1]提出了两层优化(bi-level) 问题，其中发电商 interactions. The model is based on the Watkins’s Q(l) 的生产效益优化问题是基于市场最优化调度(最优 Reinforcement Learning and takes into account the network