摘要
针对当前窃电识别和线损评估方面存在的数据分析方法单一、精确度不足的问题,首先对营配数据进行融合分析,提取出台区的窃电特征和线损特征;然后提出了基于自适应遗传算法优化的SVM窃电识别方法和基于LM算法改进的WNN线损评估方法,并通过仿真分析了方法的性能;最后结合实际算例展开分析,验证了所提方法应用于台区窃电识别和线损评估的有效性和准确性。
At present,there are problems of single data analysis method and insufficient accuracy in the current electricity theft identification and line loss assessment.In order to solve this problem,the fusion analysis of the distribution data is carried out,and the characteristics of power theft and line loss in the station area are extracted.Then,an SVM steal recognition method based on adaptive genetic algorithm optimization and a WNN line loss evaluation method improved based on LM algorithm are proposed,and the performance of the method is analyzed by simulation.Finally,combined with the actual examples,the effectiveness and accuracy of the proposed method applied to the identification of electric theft and line loss assessment in station area are verified.
作者
孙文川
SUN Wenchuan(Linyi Power Supply Company,State Grid Shandong Electric Power Company,Linyi 276002,China)
出处
《电工技术》
2023年第7期67-71,共5页
Electric Engineering
关键词
营配数据
特征提取
窃电识别
线损评估
marketing and distribution data
feature extraction
electricity theft detection
line loss evaluation
