引用本文:陆双,彭曙蓉,杨云皓,苏盛,刘登港,张恒,王书龙.基于平均影响值-启发式前向搜索的异常光伏用户识别方法[J].电力自动化设备,2022,42(2):
LU Shuang,PENG Shurong,YANG Yunhao,SU Sheng,LIU Denggang,ZHANG Heng,WANG Shulong.Identification method of abnormal photovoltaic users based on mean impact value and heuristic forward searching[J].Electric Power Automation Equipment,2022,42(2):
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 13636次   下载 1746  
基于平均影响值-启发式前向搜索的异常光伏用户识别方法
陆双1, 彭曙蓉1, 杨云皓2, 苏盛1, 刘登港1, 张恒1, 王书龙3
1.长沙理工大学 电气与信息工程学院,湖南 长沙 410114;2.浙江大学 计算机科学与技术学院,浙江 杭州 310027;3.武汉大学 电气与自动化学院,湖北 武汉 430072
摘要:
随着国家对光伏产业的大力推进与扶持,以及国家补贴政策具有长期性,出现了很多不法用户以虚假记录发电量骗取国家补贴的行为。针对现有的分布式光伏防窃电技术,提出了一种基于平均影响值(MIV)-启发式前向搜索的异常光伏用户识别方法。通过获取同一地区的标杆光伏用户及其他光伏用户在同一时段的发电数据,利用原始数据训练BP神经网络,再根据MIV的计算原理构造2组新的训练样本,用新样本的仿真结果计算各光伏用户的MIV,结合启发式前向搜索算法筛选得到与标杆光伏用户发电数据关联性大的用户,未被筛选的用户就是异常光伏用户。仿真结果验证了所提方法对异常光伏用户识别的有效性。
关键词:  数据相关性  标杆光伏用户  异常光伏用户识别  MIV  启发式前向搜索  BP神经网络
DOI:10.16081/j.epae.202112019
分类号:TM615
基金项目:国家自然科学基金资助项目(51777015);湖南省教育厅重点项目(20A021)
Identification method of abnormal photovoltaic users based on mean impact value and heuristic forward searching
LU Shuang1, PENG Shurong1, YANG Yunhao2, SU Sheng1, LIU Denggang1, ZHANG Heng1, WANG Shulong3
1.School of Electrical & Information Engineering, Changsha University of Science & Technology, Changsha 410114, China;2.College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China;3.School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, China
Abstract:
With the country’s vigorous promotion and support of the photovoltaic industry and the long-term nature of the national subsidy policy, many illegal users cheat the state subsidy by falsely recording the electricity generation. Aiming at the existing distributed photovoltaic anti-theft technologies, an identification method of abnormal photovoltaic users based on MIV(Mean Impact Value) and heuristic forward searching is proposed. By obtaining the power generation data of benchmark photovoltaic users and other photovoltaic users in the same time segment and the same area, the original data are used to train the BP neural network. And then according to the MIV calculation principle, two sets of new training samples are constructed, and the simulative results of new samples are used to calculate the MIV of each photovoltaic user. Combining heuristic forward searching algorithm, the users with high power generation data correlation with the benchmark photovoltaic users are filtered, and the unfiltered users are abnormal photovoltaic users. The simulative results verify the effectiveness of the proposed method for identifying abnormal photovoltaic users.
Key words:  data correlation  benchmark photovoltaic users  identification of abnormal photovoltaic users  mean impact value  heuristic forward searching  BP neural network

用微信扫一扫

用微信扫一扫