多虚拟电厂日前随机博弈与实时变时间尺度优化方法

Day-ahead stochastic game and real-time adaptive time scale optimization method for multiple virtual power plants

针对虚拟电厂(VPP)的运营性能存在多元信息风险、调度灵活性受限等问题,提出计及网络约束的多VPP日前随机博弈与实时变时间尺度优化方法。为应对VPP日前运行中面临的多种风险,综合考虑VPP内随机预测信息的概率分布及调节水平,建立风险效用模型,定量刻画各时段各元件及VPP整体的风险水平。考虑VPP的跨地域特性,构建网络相关约束及过网费与报价的动态耦合约束,建立更具可行性的日前电能交易模型。针对实时预测信息的更新和波动,考虑调度偏差减少率与综合成本增加率的博弈,建立实时变时间尺度优化模型。最后,仿真验证了所提方法能有效适应多种不确定性运行场景,在保证经济性的同时提升了功率曲线跟踪能力。

In order to solve the problems of multiple information risks and limited scheduling flexibility in the operation performance of virtual power plant(VPP),a day-ahead stochastic game and the real-time adaptive time scale optimization method for multiple VPPs considering network constraints is proposed.In order to cope with various risks faced by the VPP in the day-ahead operation,the risk utility model is established by comprehensively considering the probability distribution and the adjustment level of the random prediction information in VPPs,and the risk level of all components in the VPP and the whole VPP are quantitatively characterized during all time periods.Considering the cross-regional characteristics,the network-related constraints and the dynamic coupling constraints between the transmission fee and the bidding price are constructed,more feasible day-ahead VPP power energy transaction model is established.In view of the update and fluctuation of real-time forecast information,considering the game between the scheduling deviation reduction rate and the comprehensive cost increase rate,a real-time adaptive time scale optimization model is established.The simulative results show that the proposed model can effectively adapt to various uncertain operation scenarios,and improve the power curve tracking ability while ensuring economy.