基于多智能体的电力市场竞价行为研究

发布时间：2018-11-18 17:16

【摘要】：20世纪宋期,全球范围内掀起了电力市场改革的浪潮,其内容是打破电力市场垄断,引入竞争机制,其目的是推动电力行业生产力提高和资源的合理配置。到目前为止,电力市场的改革已取得了很大成绩,然而对于一个新的制度风险也是难免的。为了使我国电力市场改革顺利进行,电力市场健康发展,起到资源优化配置的作用,对电力市场竞价行为的研究有很大的现实意义。本文对于电力市场竞价行为研究主要做了如下工作：通过阅读和分析大量国内外关于电力市场竞价行为研究的相关文献,对于国内外关于电力市场竞价行为研究的文献做出归纳总结,并选择用多智能体系统对电力市场竞价行为的仿真。研究了电力市场相关理论,主要包括电力市场的定义、要素、交易模式和竞价机制,分析了电力市场中发电商的成本和利润。研究了多智能体技术,将遗传算法作为多智能体的主要计算算法,并对算法进行研究、软件实现和仿真分析,确保算法准确可靠。选取两种经典市场模型(古诺模型和伯特兰德模型),基于多智能体技术建立了具有两家供应商的市场竞价模型,并用C++软件仿真此竞价行为,最后通过理论分析仿真结果,验证了本文所设计的多智能体系统的有效性和实用性。在此基础上,将两家供应商扩展到多家供应商,基于多智能体技术建立了多家发电商的电力市场模型。分别在PAB竞价机制下和MCP竞价机制下,对不同电力市场需求量(1000MW、3150MW、3200MW、5700MW)的电力市场竞价行为进行仿真、对比分析,得出结论。同时,在相同电力市场需求量下,对PAB竞价机制下和MCP竞机制下的电力市场竞价行为进行对比分析。
[Abstract]:In the Song Dynasty of the 20th century, the reform of the electric power market was carried out in the whole world. Its content was to break the monopoly of the electricity market and introduce the competitive mechanism, the purpose of which was to promote the improvement of the power industry productivity and the rational allocation of resources. Up to now, great achievements have been made in the reform of electricity market, but it is inevitable for a new system risk. In order to make the reform of electricity market in our country go on smoothly, the electricity market develops healthily and plays the role of optimizing the allocation of resources, it is of great practical significance to study the bidding behavior of electricity market. The main work of this paper is as follows: through reading and analyzing a large number of domestic and foreign research on electricity market bidding behavior, This paper summarizes the literatures about the bidding behavior of electricity market at home and abroad, and chooses to use multi-agent system to simulate the bidding behavior of electricity market. This paper studies the relevant theories of electricity market, including the definition, elements, trading mode and bidding mechanism of electricity market, and analyzes the cost and profit of the generator in the electricity market. The multi-agent technology is studied, and the genetic algorithm is regarded as the main computing algorithm of multi-agent. The algorithm is studied, software realization and simulation analysis are carried out to ensure the accuracy and reliability of the algorithm. Two kinds of classical market models (Cournot model and Bertrand model) are selected. Based on the multi-agent technology, the market bidding model with two suppliers is established, and the bidding behavior is simulated by C software. Finally, the simulation results are analyzed theoretically. The effectiveness and practicability of the multi-agent system designed in this paper are verified. On this basis, two suppliers are extended to multiple suppliers, and the electricity market model of multiple generators is established based on multi-agent technology. Under the PAB bidding mechanism and the MCP bidding mechanism, the electricity market bidding behavior of different electricity market demand (1000MW / 3150MW / 3200MW / 5700MW) is simulated and analyzed, and the conclusion is drawn. At the same time, under the same demand of electricity market, the bidding behavior of electricity market under PAB bidding mechanism and MCP competitive mechanism is compared and analyzed.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F426.61;TP18

【参考文献】