电力市场模拟中基于信念的发电商智能个体决策模型.pdfVIP

29 卷 14 期 电 网 技 术 Vol. 29 No. 14 2005 年7 月 Power System Technology Jul. 2005 文章编号：1000-3673 （2005 ）14-0004-06 中图分类号：TM734 文献标识码：A 学科代码：470 4054 电力市场模拟中基于信念的 发电商智能个体决策模型 江健健，康重庆，夏 清 （清华大学电机工程与应用电子技术系，北京市 海淀区 100084 ） A NEW BELIEF-BASED BID DECISION MODEL OF GENCO AGENT IN ELECTRICITY MARKET SIMULATION JIANG Jian-ji an，KANG Chong-qing，XIA Qing （Department of Electrical Engineering，Tsinghua University，Haidian District，Beijing 100084 ，China ） ABSTRACT: To simulate the bid decision-making behavior of 该决策模型构建的Multi-Agent 电力市场模拟方法，可以 GenCo, on the basis of Bid Acceptance Probability (BAP), the 对电力市场进行定量和定性的分析。算例结果表明：市场成 authors build up a new bid decision-making model of GenCo in 员的学习能力和发电成本分布是影响市场演化过程的关键 electricity market. The new model can optimize the bidding 因素。 本文所提出的思路同样可以建立购电者个体的决策 curve of GenCo agent to maximize its expected profit. The 模型。 features of the proposed model is analyzed in detail and 关键词：电力市场模拟；信念学习；中标概率；中标决策； theoretical criterion of optimized bidding price is given, therefore the physical characteristics of optimized bidding 最优判据 point are revealed and a directviewing explanation is given 1 引言 according to the bidding curve ’s geometrical shape. The proposed bid decision-making model can be applied to 在基于Multi-Agent 结构的电力市场模拟研究方 construct a Multi-Agent based simulation method to, 法中，如何设计高效、合理的智能个体行为模型， qualitatively and quantitatively analyze electricity markets. [1] 是整个模拟方法的关键 。在建立市场成员智能个体 Numerical examples show that the learning ability of mark