基于LLE降维和BP_Adaboost分类器的GIS局部放电模式识别  

Pattern Recognition of GIS Discharge Types Based on LLE Dimensionality Reduction and BP_Adaboost Classifier

作　　者：律方成[1] 张波[1] 

机构地区：[1]华北电力大学河北省输变电设备安全防御重点实验室,河北保定071003

出　　处：《电测与仪表》2014年第15期37-41,共5页Electrical Measurement & Instrumentation

基　　金：国家高技术研究发展计划项目(863计划)(2011AA05A121);中央高校基本科研业务费专项资金资助项目(13ZD14)

摘　　要：在气体绝缘组合电器(gas insulated switchgear,GIS)实体模型中分别放置了针-板、悬浮金属颗粒和绝缘子表面固定金属颗粒放电模型,用超声波传感器采集到其放电波形。对放电波形提取的特征向量进行局部线性嵌入(local linear embedding,LLE)算法降维处理,用降维后的向量作为输入对BP_Adaboost分类器进行训练和测试类型识别。识别结果表明,用这样方法进行GIS绝缘缺陷类型识别可以在减少计算量的同时保持较高的识别率,说明了其在局部放电模式识别应用中的有效性。Needle -plate, suspended metal particles and metal particles fixed on insulating surface discharge models are placed in gas insulated switchgear （GIS） entity model.The corresponding discharge waveforms are collected by the ultrasonic sensor.The eigenvectors of the discharge waveforms has been processed by local linear embedding algorithm to reduce the dimension.The processed vectors are used as the input to train and test BP _ Adaboost classifi-er.Recognition results show that GIS insulating defects recognition with this method will reduce the calculation amount and keep a high recognition rate at the same time , which will illustrate the effectiveness in the application of partial discharge pattern recognition.

关 键 词：GIS LLE算法 BP_Adaboost分类器 放电类型识别 

分 类 号：TM855[电气工程—高电压与绝缘技术]