基于主从博弈的需求响应遗传算法研究  

Research on demand response genetic algorithm based on master-slave game theory

作　　者：杨欣 施天成 王绪利 周帆 龙雪 YANG Xin;SHI Tiancheng;WANG Xuli;ZHOU Fan;LONG Xue(State Grid Anhui Electric Power Co.,Ltd.,Economic and Technological Research Institute,Hefei 230000,Chin;Zhongneng Bowang(Beijing)Technology Co.,Ltd.,Beijing 102400,China)

机构地区：[1]国网安徽省电力有限公司经济技术研究院,合肥230000 [2]中能博望(北京)科技有限公司,北京102400

出　　处：《自动化与仪器仪表》2024年第10期98-100,105,共4页Automation & Instrumentation

基　　金：考虑需求侧多元柔性资源梯级响应的多层次配电网规划优化方法研究(B6120923000D)。

摘　　要：需求响应(DR)在电力系统中是一种有效的方式来平衡供需关系、减轻负荷压力,并提高能源利用效率。而博弈论已被证明是DR管理中一种非常有效的工具。将DR在电网公司和负载聚合器(LA)之间建模为主从博弈(Stackelberg博弈),其中电网公司扮演领导者的角色,而LA则是追随者。针对集中式求解方法的限制,提出了一种基于遗传算法的分散求解方法。为验证该方法的有效性,设计并优化了一个具有单个电网公司和三个LA的实时定价市场的日前优化案例。研究结果表明,所提出的方法能够调整需求峰值,并证明在电网公司和多个LA之间能够充分利用DR潜力。Demand response(DR)is seen in the power system as an effective way to balance supply and demand,reduce load pressure,and improve energy efficiency.And game theory has proved to be a very effective tool in DR management.In this paper,DR is modeled as a Stackelberg game(master and slave game)between grid companies and load aggregators(LA),where state grid corporation plays the role of the leader and LA is the follower.Given the limitations of the centralized solution method,this paper proposes a decentralized solution based on a genetic algorithm.To verify the effectiveness of this method,this paper designs and optimizes a day-ahead optimization case for a real-time pricing market with a single state grid corporation and three LA.The results show that the method proposed herein is able to adjust peak demand and demonstrate the full potential of DR between grid companies and multiple LAs.

关 键 词：需求响应 主从博弈 电网公司 负载聚合器 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]