摘要
机电故障是真空断路器开断异常、拒动或误动的主要原因,准确有效地掌握真空断路器机电状态对保障电网安全至关重要。该文以合闸线圈电流和行程信号为主要在线监测量,通过故障模拟试验,系统地分析了监测量在不同故障类型及严重程度下特征量的演变规律,提出一种基于三元图谱的实用化故障诊断模型。借助主成分分析(principal componentanalysis,PCA)方法,提取出具有正交贡献的三类主成分特征,并以此构建三元特征图谱,直观展示了不同故障下曲线的演变路径;在此基础上,采用局部加权回归给出故障边界方程,实现了故障的类型识别;最后,通过k-means聚类中心和欧式距离计算,提出了故障发展程度量化分析方法。试验结果表明,对于同型号设备,该方法在缺失刚分、刚合点的条件下,判断准确率达到97%以上,为真空断路器的在线诊断和故障预警提供了一种更为高效、直观和实用化的方法。
Mechanical and electrical failure is the main reason for the abnormal breaking,rejection or malfunction of vacuum circuit breaker.It is crucial for the safety of the power grid to grasp the mechanical and electrical state of vacuum circuit breaker accurately and effectively.This paper took the closing coil current and stroke signals as the main online monitoring variables.Through the fault simulation test,the evolution of the characteristic variables under different fault types and severity levels was systematically analyzed,and a practical fault diagnosis model based on ternary plot was proposed.With the application of principal component analysis(PCA),three types of principal component features with orthogonal contributions were extracted,and a ternary plot which visually showed the evolution path of the curve under different failures was constructed.On this basis,locally weighted regression was utilized for giving the fault boundary equation,which realized the type identification of the fault.Finally,through the calculation of k-means clustering center and Euclidean distance,a quantitative analysis method of the development degree of the fault was proposed.The test results show that for the same type of equipment,this method can achieve a judgment accuracy of more than 97%under the condition of missing the opening and closing points,which provides a more effective,visualized and practical way for the online diagnosis and fault warning of vacuum circuit breakers.
作者
任明
陈荣发
夏昌杰
余家赫
李琛
董明
张乐萌
潘卫国
REN Ming;CHEN Rongfa;XIA Changjie;YU Jiahe;LI Chen;DONG Ming;ZHANG Lemeng;PAN Weiguo(State Key Laboratory of Electrical Insulation and Power Equipment(Xi’an Jiaotong University),Xi’an 710049,Shaanxi Province,China;China Railway Signal&Communication(Changsha)Metro-Transportation Control Technology Co.,Ltd.,Changsha 410000,Hunan Province,China)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2022年第S01期332-342,共11页
Proceedings of the CSEE
基金
国家自然科学基金项目(51877171)
关键词
真空断路器
三元图谱
主成分分析
局部加权回归
K-MEANS聚类
vacuum circuit breaker
ternary plot
principal component analysis
locally weighted regression
k-means clustering
作者简介
任明(1987),男,工学博士,副教授,主要从事高压试验技术、状态监测智能传感器、图像融合与故障可视化等方面的技术研究方面的研究工作,renming@xjtu.edu.cn。
