摘要
为提高双Y型外熔丝电容器故障诊断精度,克服传统的中性线不平衡电流受初始不平衡量影响灵敏度较低的问题,提出了将母线电压值、三相电流有效值与中性线电流有效值信息融合,利用逻辑回归进行故障诊断的方法。该方法首先利用随机样本生成学习样本,利用仿真计算方法得到母线电压值、三相电流有效值与中性线电流的计算结果,并计算得到规则化特征指标;其次,基于上述学习样本利用逻辑回归学习得到决策函数;最后将现场实测数据代入决策函数,最终实现故障诊断判别。仿真结果表明,与传统的单纯依赖中性线电流进行故障判断相比,该方法有效提升了故障诊断灵敏度。同时该方法简单易实现,在工程上具有较强的推广性。
In order to increase the accuracy of fault diagnosis of external-fusing double-star power capacitor,a logical regression based fault diagnosis method is proposed.The RMS of voltage on the bus,the RMS of three-phase current and the RMS of current in the neutral line are all utilized in this method to overcome the problem that the initial unbalance rate may reduce the sensitivity of the diagnosis method only based on neutral line current.At first,the study set is generated by random sample method,and then the characteristic indexes could be calculated.Then according to the study set,determine function can be obtained by logical regression.Finally,based on the measure value of voltage and current,the fault probability of the capacitors can be obtained by the determine function.Simulation results show that the sensibility of the proposed method is much higher than the conventional one,which only uses natural line current.
作者
滕予非
唐嘉忆
段翔兮
宋梁
李世龙
Teng Yufei;Tang Jiayi;Duan Xiangxi;Song Liang;Li Shilong(State Grid Sichuan Electric Power Research Institute,Chengdu 610041,Sichuan,China;State Grid Sichuan Economic Research Institute,Chengdu 610041,Sichuan,China;State Grid Sichuan Maintenance Company,Chengdu 610041,Sichuan,China)
出处
《四川电力技术》
2019年第5期41-45,共5页
Sichuan Electric Power Technology
关键词
逻辑回归
双Y型外熔丝电容
故障诊断
机器学习
logical regression
external-fusing double-star power capacitor
fault diagnosis
machine learning
