基于随机森林算法的断路器分合闸线圈故障电流曲线识别  

Fault Current Curves Identification of Circuit Breaker Opening/Closing Coil Based on Random Forest Algorithm

作　　者：刘芹 彭在兴 王颂 易林 陈曦 褚飞航 骆挺 梁梦婕 羿敏 刘定新[3] LIU Qin;PENG Zaixing;WANG Song;YI Lin;CHEN Xi;CHU Feihang;LUO Ting;LIANG Mengjie;YI Min;LIU Dingxin(Electrical Power Research Institute of CSG,Guangzhou 510000,China;China Southern Power Grid Co.,Ltd.,Guangzhou 510000,China;College of Electrical Engineering,Xi'an Jiaotong University,Xi'an 710049,China)

机构地区：[1]南方电网科学研究院有限责任公司,广州510000 [2]中国南方电网有限责任公司,广州510000 [3]西安交通大学电气工程学院,西安710049

出　　处：《高压电器》2019年第7期93-100,共8页High Voltage Apparatus

摘　　要：高压断路器是电力系统的核心设备,其稳定可靠运行具有重要意义。高压断路器分合闸电流曲线能够很大程度上反映高压断路器分合闸机构的运行状态,因此文中研究了在几种典型故障条件下,断路器分合闸线圈电流的变化。随机森林算法由于算法实现过程中的两次随机过程,使得算法过拟合风险大大降低,模型泛化能力大大增强。提出了使用随机森林算法识别故障电流曲线,取得了很好的效果。To identify the fault types of high-voltage circuit breaker accurately and reliably,a fault diagnosis method for high voltage circuit breaker is formed.In this paper,the reclosing coil current of HV circuit breaker under different conditions is analyzed,in order to extract time and current as characteristic parameters to form training set.Combining random forest algorithm(RF)with training set,random forest classifier is established to distinguish fault types of high voltage circuit breakers efficiently.Experiments show that the algorithm can accurately distinguish the fault of high voltage circuit breaker,which greatly improves the utilization rate and efficiency of HV circuit breaker.

关 键 词：高压断路器 分合闸线圈电流 随机森林算法 故障诊断 

分 类 号：TM561[电气工程—电器]