基于两阶段随机规划的虚拟电厂优化交易策略  

Optimal Bidding Strategy Based on Two-Stage Stochastic Programming for Virtual Power Plant

作　　者：周博[1] 吕林[1] 高红均[1] 谭心怡 吴泓灏 ZHOU Bo;LU Lin;GAO Hongjun;TAN Xinyi;WU Honghao(College of Electrical Engineering and Information Technology,Sichuan University,Chengdu 610065,China)

机构地区：[1]四川大学电气信息学院,成都市610065

出　　处：《电力建设》2018年第9期70-77,共8页Electric Power Construction

基　　金：国家自然科学基金项目(51377111);中央高校基本科研业务费专项资金资助项目(YJ201750)~~

摘　　要：针对市场电价以及可再生能源(renewable energy sources,RES)出力的不确定性,文章以虚拟电厂(virtual pow er plant,VPP)模式聚合分布式能源(电动汽车、需求响应等)参与电力市场交易,通过分布式能源间的协调优化互补,提升虚拟电厂出力稳定性和市场竞争力。采用多场景法模拟日前市场出清电价和风电的不确定性,以虚拟电厂运行效益最大化为目标,构建基于两阶段随机规划的虚拟电厂最优交易策略模型,其中第一阶段日前市场考虑出清电价的不确定性,第二阶段平衡市场考虑风电出力的随机性。此外,利用条件风险价值(conditional value at risk,CVaR)来度量交易策略风险,从而实现经济性和风险的偏好选择。最后,通过算例分析了不确定性和风险偏好对虚拟电厂收益以及风险损失的影响,为不同风险偏好主体提供参考。Considering the uncertainty of the market price and the power of renew able energy,this paper adopts virtual power plant（ VPP） model to aggregate distributed energy（ electric vehicles,demand response,etc.） to participate in electricity market transactions,to improve the stability and market competitiveness of VPP by optimizing and coordinating distributed energy sources. Using multi-scenario method to simulate uncertainty of day-ahead market clearing price and the output of wind power plant,with the goal of maximizing the operating efficiency,the optimal trading strategy model of VPP based on two-stage stochastic programming is formulated. In the first-stage,day-ahead market consider the uncertainty of the price; in the second-stage,balance market consider the uncertainty of the wind power plant production. In addition,the risk associated with the VPP profit is explicitly taken into account through the incorporation of the conditional value-at-risk metric,so as to realize the preference of economy and risk. Finally,the influence of risk appetite and uncertainty on the profit and risk loss of VPP are analyzed,which can provide valuable reference for different risk appetite of VPP.

关 键 词：虚拟电厂(VPP) 随机规划 电动汽车 需求响应 条件风险价值(CVaR) 

分 类 号：TM61[电气工程—电力系统及自动化] F426[经济管理—产业经济]