基于社交网上演化博弈的光伏台区用户需求响应特性研究

“双碳”背景下,通过实施需求侧响应实现新型电力系统的协调运行是解决新能源出力不确定性的主要手段。然而,现实中的用户具有有限理性,其对电价或激励措施的响应呈现出异质性和不确定性。针对此,文章计及有限理性用户对电价或激励措施响应过程中的信息交互以及策略学习特征,提出了一种基于社交网上演化博弈模型的用户需求响应分析模型。首先,考虑到现实经济人在决策中存在信息交互以及策略学习与更新,且该信息交互背后为复杂社群系统,通过构建社交网刻画了用户之间的信息交互关系;其次,根据电价理论和电力供需平衡关系,单个用户的响应决策将会影响其他用户利益,采用博弈模型描述了群体用户的决策过程;最后,考虑到用户对所获取信息处理的有限理性特征,基于社交网上演化博弈模型来描述用户对电价和激励措施的响应过程。仿真分析了不同社交网络结构以及电价或激励措施对用户响应的影响。结果表明,用户社交网小世界属性的增强将会提高用户的响应,价格系数的提高会降低用户的用电意愿。

Study on Demand Response of Users in Photovoltaic Station Area Applying Evolutionary Game Model in Social Network

In the "double carbon target" background, the coordinated operation of the new power system by implementing demand-side response is the main means to solve the uncertainty of new energy output. However, in reality, customers have bounded rationality, and their response to tariffs or incentives is heterogeneous and uncertain. In view of this, considering the information interaction and strategy learning characteristics of bounded rational users in the response process of electricity price or incentive measures, this paper proposes a user-side demand response analysis model based on evolutionary game model in social network. Firstly, considering that the real economic man has information interaction and strategy learning and updating in decision-making, and that the information interaction is a complex community system, the information interaction relationship between users is described by constructing a social network. Secondly, according to the electricity price theory and the balance between power supply and demand, the response decision of a single user will affect the interests of other users. The game model is used to describe the decision-making process of group users. Finally, considering the bounded rationality of users’ information processing, the evolutionary game model in social network is used to describe the response process of users to electricity price or incentive measures. The effects of different social network structures and electricity prices or incentives on user response are simulated and analyzed. The results show that the enhancement of the small world attribute of users’ social network will improve users’ response, and the increase of price coefficient will reduce users’ willingness to use electricity.