基于朴素贝叶斯算法的避雷器缺陷识别方法研究  

Research on Classification of Arrester Defect Diagnosis Based on Naive Bayes Algorithm Inference

作　　者：李亚锦[1] 刘英男 张婉莹 于大洋[1] 张国新 苏宁 LI Ya-jin;LIU Ying-nan;ZHANG Wan-ying;YU Da-yang;ZHANG Guo-xin;SU Ning(School of Electrical Engineering,Shandong University,Jinan 250061,China;Qionghai Power Supply Bureau of Hainan Electric Power Co.,Ltd,Qionghai 571400,China)

机构地区：[1]山东大学电气工程学院,山东济南250061 [2]海南电网有限责任公司琼海供电局,海南琼海571400

出　　处：《电工电气》2022年第1期20-23,共4页Electrotechnics Electric

摘　　要：高温、高湿、高盐特殊环境下,加速了避雷器劣化或潜伏性缺陷的发展。仅依靠避雷器监测指标判断缺陷,难以识别特殊环境下避雷器的异常状态。提出一种基于朴素贝叶斯算法的避雷器缺陷识别技术,提取特殊环境下影响避雷器运行状态的关键特征量,通过朴素贝叶斯算法计算训练样本的先验概率和测试样本的后验概率,从而识别避雷器缺陷类型。利用实际监测和检测数据进行分析,验证了所提方法的可行性和正确性。The special environment of high temperature,high humidity,and high salt could accelerate the development of the deterioration or latent defects of the arrester.It is difficult to identify the abnormal state of the arrester in the special environment just relying on the monitoring index of the arrester.This paper proposed arrester defect diagnosis technology based on Naive Bayes,which extracted the key features that affect the operation state of the arrester in a special environment.It calculated the prior probability of training samples and the posterior probability of test samples through a Naive Bayes algorithm to identify the type of arrester defects.The feasibility and correctness of the proposed method are analyzed and it verified by the actual monitoring and detection data.

关 键 词：朴素贝叶斯算法 特殊环境 带电检测 避雷器 缺陷识别 

分 类 号：TM862[电气工程—高电压与绝缘技术]