摘要
配电台区中,用户窃电将增大台区线损电量,窃电电量直接对应非技术线损,并与台区线损电量间存在因果关系。分析指出,常见窃电手法中窃电用户的窃电量多与电表计量电量呈正相关关系,这将使得线损电量与窃电用户的计量电量之间存在隐含的因果关系。以此为基础,提出基于格兰杰归因分析的高损台区窃电检测方法。通过协整检验及格兰杰检验分析用户用电量与台区线损时间序列间的长期趋势及格兰杰因果关系,将对台区线损影响突出的用户识别为窃电用户。基于实际高损台区和高损线路用户数据的测试算例和现场稽查验证了所提方法的有效性。
In the distribution area,the line loss power increases with user electricity-theft.Electricity-theft power directly corresponds to the non-technical line loss,and has a causal relationship with the line loss of the distribution area.It is pointed out that there is a positive correlation between electricity-theft quantity of users and electricity quantity of electric meters,which leads to the implicit causal relationship between the line loss and electricity quantity of electricity-theft users.On this basis,a method for detecting electricity-theft in high-loss distribution area based on Granger causality analysis is proposed.The long-term trend and Granger causality relationship between electricity consumption quantity and line loss time series in the distribution area are analyzed through co-integration test and Granger causality test,and users who have significant impact on line loss are identified as electricitytheft users.The validity of the proposed method is proven by test experiments and field inspection based on the actual high-loss distribution area and high-loss line data.
作者
金晟
苏盛
曹一家
薛阳
杨艺宁
刘厦
JIN Sheng;SU Sheng;CAO Yijia;XUE Yang;YANG Yining;LIU Sha(Hunan Key L.aboratory of Smart Grid Operation and Control(Changsha University of Science and Technology),Changsha 410004,China;China Electric Power Research Institute,Beijing 100085,China)
出处
《电力系统自动化》
EI
CSCD
北大核心
2020年第23期82-89,共8页
Automation of Electric Power Systems
基金
国家自然科学基金资助项目(51777015)
国家自然科学基金委员会-国家电网公司智能电网联合基金资助项目(U19266207)
国家电网公司总部科技项目“反窃电及稽查监控关键技术研究”
湖南省自然科学基金课题(2020JJ4611)资助。
关键词
窃电
线损
归因分析
格兰杰检验
协整检验
electricity-theft
line loss
causality analysis
Granger test
co-integration test