基于主从博弈的虚拟电厂双层竞标策略

虚拟电厂是一种集分布式能源和负荷为一体的综合能源服务商,也是一种新型的上网交易模式。由于虚拟电厂内部主体存在产权独立性,且电价竞标与电量竞标存在先后行动次序,应用Stackelberg动态博弈理论,建立虚拟电厂竞标问题的动态博弈模型。考虑风、光及负荷的出力波动和预测误差,分别建立虚拟电厂的电价竞标模型和电量竞标模型。基于两个子模型的主从递阶关系,利用最优性一阶条件和粒子群智能优化算法,求出模型的均衡解,确定虚拟电厂内多个分布式电源的交易电价与调度计划。最后对模型进行仿真,讨论不同场景下虚拟电厂的收益、电价竞标和电量竞标结果,验证模型的有效性。

Stackelberg Game Based Bi-level Bidding Strategy for Virtual Power Plant

Virtual power plant(VPP)is an energy service provider integrating load and distributed energy, which is also a new type of online market transaction mode. On one hand, internal members of VPP have independent property and on the other, electricity price and electricity quantity bidding act in order. After taking into account the above-mentioned essentials, a bidding model for VPP based on the dynamic game theory is developed using the Stackelberg game theory. Considering the output fluctuation and forecasting error of wind turbine, photovoltaic and load, the electricity price bidding model and electricity quantity bidding model are built, respectively. Based on the principal and subordinate hierarchical relation between the two submodels, the equilibrium solution of the model is obtained using the first-order optimality condition and the particle swarm intelligence optimization algorithm. Thus the transaction price and dispatching scheme of the VPP are determined. Finally, the model is simulated, the revenues from electricity price and electricity quantity bidding results in different scenarios are discussed to verify the validity of the model.