摘要
检测异常用电的目的是打击异常用电,并减少电能的非技术性损失。文中提出了一种基于逻辑回归算法的异常用电辨识方法,主要包括特征提取、算法构建以及检验模型等模块。提取电网用电量等相关数据,并从数据集中提取出电量下降趋势指标、线损指标和告警类指标用作异常用电评判体系。进行电量下降趋势指标、线损指标和告警类指标的归一化处理,再进行离群邻近度的计算,初步筛选异常用电用户。对初步筛选的结果进行逻辑回归算法的再次筛选,进一步提高识别准确率。经过电网部分用电数据的检验后,该算法相较于逻辑回归算法,识别率更高,识别效果更好。
The purpose of detecting abnormal power consumption is to strike abnormal power consumption and reduce non-technical loss of power.In this paper,an identification method of abnormal power consumption based on logistic regression algorithm is proposed,which includes feature extraction,algorithm construction and test model.Firstly,the related data such as power consumption of power grid are extracted,and the index of power decline trend,line loss index and warning index are extracted from the data set as the evaluation system of abnormal power consumption.Then,the trend index,line loss index and alarm index are normalized and used to calculate the outlier proximity so that part normal users can be distinguished.And then,the results of the preliminary screening are screened again by logistic regression algorithm to further improve the recognition accuracy.Compared with the logic regression algorithm,this algorithm has higher recognition rate and better recognition effect after testing some power consumption data of the power grid.
作者
袁翔宇
张蓬鹤
熊素琴
赵波
成达
Yuan Xiangyu;Zhang Penghe;Xiong Suqin;Zhao Bo;Cheng Da(China Electric Power Research Institute Co.,Ltd.,Beijing 100192,China;Beijing Information Science and Technology University,Beijing 100192,China)
出处
《电测与仪表》
北大核心
2021年第12期81-87,共7页
Electrical Measurement & Instrumentation
基金
国家电网有限公司总部科技项目(5442JL170007)。
关键词
异常用电
离群算法
离群邻近度
逻辑回归
学习速率
abnormal power consumption
outlier algorithm
outlier proximity
logic regression
learning rate
