计及风光预测误差的微电网日前多目标优化控制策略  

Day-ahead Multi-objective Optimization Control Strategy of Micro-grid Considering Forecast Error of Wind/Photovoltaic

作　　者：孙丛丛[1] 王致杰[1] 江秀臣[2] 刘三明[1] 盛戈皞[2] 

机构地区：[1]上海电机学院电气学院,上海201306 [2]上海交通大学电气工程系,上海200240

出　　处：《高压电器》2018年第1期73-81,共9页High Voltage Apparatus

基　　金：国家自然科学基金资助项目(51477099);上海市自然科学基金资助项目(15ZR1417300;14ZR1417200);上海市教委创新基金项目(14YZ157;15ZZ106)~~

摘　　要：微电网风光资源出力预测的随机性导致优化决策方案存在偏差,为了减轻预测误差随机性对微电网运行的不利影响,针对风/光/柴/储并网型微电网系统,提出了计及风光预测误差的日前多目标优化控制策略。在建立微电网各单元模型的基础上,分析了影响风光资源的随机性因素;考虑系统供需平衡以及微电网各单元运行特性等主要约束,建立了同时优化经济效益和环境效益的多目标优化模型。建模过程中采用情景生成与削减方法处理风光资源预测误差带来的随机性。采用差异进化算法求解模型,通过算例仿真,比较不同模式下微电网的优化结果,验证了文中策略的正确性和有效性。The uncertainty of wind/photovohaic output power forecast of micro-grid leads to the large deviation of op- timization decision. In order to mitigate the adverse effects of the randomness of forecast error on the micro-grid, the day-ahead multi-objective optional control strategy of micro-grid considering forecast error of wind/photovoltaic is put forward in a grid-connected micro-grid consisting of wind power, photovoltaic system, diesel generator and bat- tery storage system（BESS）. The influence factors of wind and solar resources are analyzed on the basis of establishing the model of every unit; a multi-objective optimization model is proposed to optimize the economic and environmen- tal benefits simultaneously by considering the main constraints of supply and demand balance and the operation char- acteristics. In the process of modeling, scene generation and reduction method is adopted to deal with the uncertainty caused by the forecast error of the wind/photovoltaic. The differential evolution algorithm （DE） is employed to solve the model, and the experimental results under different modes verify the correctness and effectiveness of the pro- posed strategy.

关 键 词：微电网 风光预测误差 情景生成与削减方法 多目标优化 差异进化算法 

分 类 号：TM727[电气工程—电力系统及自动化]