接触网绝缘子污秽度预测模型的建立  

Establishment of Prediction Model for Contamination Degree of Catenary Insulators

作　　者：淮梦琪 张友鹏[1] 赵珊鹏 HUAI Mengqi;ZHANG Youpeng;ZHAO Shanpeng(School of Automation ＆ Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, Chin)

机构地区：[1]兰州交通大学自动化与电气工程学院,兰州730070

出　　处：《电瓷避雷器》2018年第3期155-160,共6页Insulators and Surge Arresters

基　　金：国家自然科学基金(编号:51567014);中国铁路总公司科技研究开发计划课题(编号:2016J010-C);甘肃省自然科学基金(编号:1606RJZA031);甘肃省青年科学基金(编号:1606RJYA216)

摘　　要：为探索污秽绝缘子泄漏电流与其表面污秽度之间的联系及它们之间具体的函数关系。首先,在人工污秽实验室多次试验,提取泄漏电流的特征量,筛选出3个有效特征量:泄漏电流有效值Ie、泄漏电流最大值Im及泄漏电流标准差σ,分别得到它们与绝缘子表面污秽度的关系;其次,利用特征量及环境湿度作为输入变量,搭建预测绝缘子污秽度的BP神经网络模型,并与试验结果对比。结果表明:提取的特征参量能够预测绝缘子污秽度,预测结果为工程人员清洁染污绝缘子工作提供参考。In order to explore relationship and function relations surface contamination of insulators. Firstly, characteristic quantities of between leakage current and the leakage current is extracted from the artificial pollution laboratory, and three effective characteristic quantities are selected： leakage cur- rent effective value Ie, maximum leakage current Im and leakage current standard deviation s, and their relationships with degrees of pollution are obtained. Then, characteristic quantities and humidity are used as input variables. BP neural network model which is used to predict the contamination degree of in- sulators is established, and the results are compared with the experimental results. The results show that： characteristic quantities are effective in predicting the degree of pollution. The prediction results provide a reference for the work of engineers to clean stained insulators.

关 键 词：绝缘子 泄漏电流 特征量 等值附盐密度 污秽度预测 BP神经网络 

分 类 号：TM216[一般工业技术—材料科学与工程] U226.8[电气工程—电工理论与新技术]