基于社团聚类算法的低压电能表故障状态分类研究

Research on fault state classification of low voltage watt hourmeters based on community clustering algorithm

目的:研究低压故障电能表的故障状态分类问题。方法:通过对智能电能表故障类型数据进行社团聚类与分析,初步确定故障电能表的故障类型。结果:在实际确定故障原因的32块电能表中,用该方法正确分析出故障原因的电能表有26块,精确率达81.25%。结论:所提出的分析方法不仅能够较为准确地确定故障电能表主要故障原因,具有较高的精确率与现场实际应用性,而且对后续电力公司进行电能表更换工作具有重要意义。

Aims:The paper studies the fault state classification of low voltage fault watt hour meters.Methods:Through the community clustering and analysis of the fault type data of intelligent watt hour meters,the fault type of the fault watt hour meter was preliminarily determined.Results:Among the 32 watt hour meters that determined the cause of the fault,26 of them were correctly analyzed by this method;and the accuracy rate was 81.25%.Conclusions:The analysis results show that the analysis method proposed in this paper can not only predict the main fault causes of the watt-hour meter,but also play an important role in the replacement of the watt-hour meter in the following power companies.The proposed analysis method can not only accurately determine the main fault causes of the faulty watt hour meter,but also has a high accuracy and practical application in the field.It is of great significance for the subsequent power companies to replace the energy meters.