关键词: 需求响应, 电动汽车充电, 动态电价, 主从博弈, 纳什均衡

Abstract:

Smart communities utilize demand response to achieve integrated management of electric vehicles (EVs) and photovoltaic-storage systems. However, traditional static pricing, which disregards actual load responses, tends to induce new load peaks and increases the volatility of the equivalent load. To address these challenges, this paper proposes a dynamic pricing strategy, wherein the price not only varies over time but is also correlated with the internal net load of the community. Firstly, the operator forecasts the photovoltaic generation and the baseline load within the scheduling horizon, and establishes the initial model of dynamic pricing. Secondly, a stackelberg game framework is constructed: the operator, as the leader in the upper-level model, aims to minimize the volatility of the equivalent load by formulating and publishing the dynamic prices as well as scheduling the energy storage output; EVs, as followers in the lower-level model, respond to the dynamic prices and optimize their charging strategies with the objective of minimizing charging costs. Furthermore, in the lower-level model, the introduction of dynamic pricing makes the charging decision of EVs mutually dependent, thereby forming an aggregative game structure, in which the optimal charging load of each EV is determined through the Nash equilibrium. Finally, the optimal dynamic pricing is determined through a genetic algorithm. Simulation results demonstrate that the proposed model effectively reduces load volatility, avoids new load peaks, and achieves the management objectives of the community operator while safeguarding the economic interests of EV users.

Key words: demand response, electric vehicles charging, dynamic prices, stackelberg game, Nash equilibrium