A Three-Layer Collaborative Optimization Strategy for Virtual Power Plant Based on Multi-Agent Hybrid Game
摘要
为适应电力市场参与主体多元化发展态势,解决多虚拟电厂(virtual power plant,VPP)利益冲突问题,提出一种包含配电网运营商(distribution system operator,DSO)、虚拟电厂运营商(virtual power plant operator,VPPO)及用户聚合商(user aggregator,UA)多层架构的混合博弈策略,支撑多 VPP的协同优化.首先,通过聚合光伏产消者、电动汽车充电站、综合能源负荷等多种灵活性资源形成 UA集群;其次,面向 DSO、VPPO以及 UA的层级能源交易及 UA主体间的能源交易,构建了"主从博弈-主从博弈-合作博弈"的多层混合博弈模型;最后,利用二分法、Karush-Kuhn-Tucker条件结合交替方向乘子法对上述模型进行求解.算例研究表明:所提出的多层混合博弈策略使光伏产消者 UA与电动汽车充电站 UA的运行成本分别降低了 4.5%和 15.3%;同时,在 DSO动态定价机制的引导下,系统运营商总收益提升了 2.7%.该策略能够有效均衡多方利益诉求,优化各运营主体间的能源交易及效益分配机制.
To address the challenges of diversified participants in the electricity market and resolve conflicts of interest among multiple virtual power plants(VPPs),this paper proposes a three-layer hybrid game strategy involving the distribution system operator(DSO),virtual power plant operator(VPPO),and user aggregator(UA)to support collaborative optimization of multi-VPP systems.First,a UA coalition is formed by aggregating various flexible resources,including photovoltaic prosumers,electric vehicle charging stations,and integrated energy loads.Next,a three-layer hybrid game model is constructed,encompassing hierarchical energy transactions among DSO,VPPO,and UA,as well as peer-to-peer energy trading among UA entities,structured as a"Stackelberg game-Stackelberg game-cooperative game"framework.Finally,the proposed model is solved using the bisection method,Karush-Kuhn-Tucker conditions,and the alternating direction multiplier method.Case studies demonstrate that the proposed multi-layer hybrid game strategy reduces the operating costs of the photovoltaic prosumer UA and the electric vehicle charging station UA by 4.5%and 15.3%,respectively.Meanwhile,guided by the dynamic pricing mechanism of the DSO,the total profit of the system operators increases by 2.7%.This strategy can effectively balance the interests of multiple stakeholders,optimizing the energy trading and benefit allocation mechanisms among different operators.