摘要
当前配电网存在信息采集不全、在线获取电网精确模型困难的问题,导致对分布式光伏的调控存在误差,难以满足配电网安全运行的要求,因此提出了一种计及模型误差的分布式光伏配电网调控方法。基于近似灵敏度建立了光伏调控量粗略计算模型;采用极限学习机(ELM)方法建立人工智能辅助决策模型,作为光伏调控量粗略计算模型的修正;进一步地,基于上述2个模型,设计了计及模型误差的分布式光伏优化调控策略;最后进行仿真分析,结果表明提出的调控方法弥补了仅依赖电网模型进行优化带来的误差,提高了优化调控的精度。
At present,the information collection is incomplete and the on-line accurate grid model is inac?cessibility in distribution network,which leads to error of distributed PV(PhotoVoltaic)control and makes it difficult to meet the requirement of the safe operation for distribution network.Thus,an optimal control method of distributed PV considering model errors in distribution network is proposed.A rough calculation model of PV control based on approximate sensitivity is built.Meanwhile,the artificial intelligence assistant decision model is established adopting ELM(Extreme Learning Machine)method as a modification of the rough calculation model for PV control.Based on the above two models,the optimal control strategy of dis?tributed PV considering model errors in distribution network is designed.Finally,the simulative results show that the proposed control method makes up the errors caused by the optimization only relying on grid model,and improves the accuracy of the optimal control.
作者
窦晓波
蔡超
段向梅
韩俊
刘之涵
陈曦
DOU Xiaobo;CAI Chao;DUAN Xiangmei;HAN Jun;LIU Zhihan;CHEN Xi(School of Electrical Engineering,Southeast University,Nanjing 210096,China;State Grid Economic Research Institute of China Electric Power Research Institute,Nanjing 210009,China)
出处
《电力自动化设备》
EI
CSCD
北大核心
2019年第12期41-48,共8页
Electric Power Automation Equipment
基金
国家自然科学基金资助项目(51777031)
国网江苏省电力有限公司科技项目(J2018058)~~
关键词
分布式光伏
配电网
优化调控
近似灵敏度
极限学习机
辅助决策
distributed PV
distribution network
optimal control
approximate sensitivity
ELM
assistant decision
