含V2G的虚拟电厂双层逆鲁棒优化调度策略  

Strategy of Bilevel Inverse Robust Optimization Dispatch of Virtual Power Plant Containing V2G

作　　者：卢志刚[1] 王荟敬[1] 赵号 冯慧波[2] 

机构地区：[1]电力电子节能与传动控制河北省重点实验室(燕山大学),河北省秦皇岛市066004 [2]国网河北省电力公司,河北省石家庄市050021

出　　处：《电网技术》2017年第4期1245-1252,共8页Power System Technology

基　　金：国家自然科学基金项目(61374098;61473246);教育部高等学校博士学科点专项科研基金(20131333110017)~~

摘　　要：采用虚拟电厂(virtual power plant,VPP)实现对电动汽车入网(vehicle to grid,V2G)智能充放电和风力发电的打包管理。考虑风电入网和V2G的不确定性影响,建立含VPP的双层逆鲁棒优化调度模型。该模型内层以发电总利润最大为目标;外层考虑内层优化结果的鲁棒性和决策者要求,引入风电出力和V2G充放电功率的最优逆鲁棒指标(optimal inverse robust index,OIRI),以此来分析风电消纳水平和V2G充放电功率之间的极限制约关系。采用网格细菌群体趋药性(grid multi-objective bacterial colony chemotaxis,GMOBCC)算法、拓扑映射和二分法相结合的嵌套优化方法对模型进行求解。最后应用仿真算例系统验证算法的有效性。In this paper, virtual power plant（VPP） was adopted to achieve package management of intelligent charge-discharge of vehicle to grid（V2G） and wind power generation. Based on uncertainty of wind power and V2G, a bi-level inverse robust optimization model containing VPP was established. Inner layer of the model aimed at optimum gross profit of generation. In consideration of optimization results＇ robustness of inner layer and requirements of decision maker, outer layer introduced optimal inverse robust index（OIRI） for wind power output and V2G charge-discharge. On this account, restrictive relation between wind power utilization capacity and V2G charge-discharge power is analysed. Nested optimization method fusing grid multi-objective bacterial colony chemotaxis（GMOBCC） algorithm, topology mapping and dichotomy model were applied for solving the model. Finally, case study simulation was performed to verify rationality and validity of the method.

关 键 词：风电 虚拟电厂 电动汽车入网 最优逆鲁棒指标 总利润 

分 类 号：TM721[电气工程—电力系统及自动化]