摘要
针对电磁式电流互感器故障诊断效率低、准确率不高问题,提出一种变分模态分解(VMD)和样本熵相结合的故障诊断方法。将原始故障信号通过VMD分解成一系列本征模函数(IMF)并进行优选,计算其样本熵作为新的互感器特征提取对象的特征值,与常见时频域特征指标组合成新的特征向量输入K近邻分类器进行训练。Matlab仿真实验结果表明,该方法中新的特征指标用于低压电流互感器故障诊断是有效可行的,可为电磁式电流互感器故障诊断提供参考。
Aiming at the problems of low efficiency and low accuracy of fault diagnosis of electromagnetic current transformers,a fault diagnosis method based on variational mode decomposition(VMD)and sample entropy is proposed.The original fault signal is decomposed into an Intrinsic Mode Function(IMF)series through VMD and optimized.The sample entropy is calculated as the feature value of the new transformer feature extraction object,which is combined with the common time-frequency domain feature index to input the K-nearest neighbor classifier for training.Matlab simulation experiments show that the new characteristic index of this method is effective and feasible for fault diagnosis of low-voltage current transformers,which can provide a reference for fault diagnosis of the electromagnetic current transformer.
作者
唐登平
蔡文嘉
周翔宇
李云峰
郭正
刘岑岑
TANG Dengping;CAI Wenjia;ZHOU Xiangyu;LI Yunfeng;GUO Zheng;LIU Cencen(Measurement Center of State Grid Hubei Electric Power Co.,Ltd.,Wuhan 430075,China;School of Mechanical and Electrical Engineering,Wuhan University of Technology,Wuhan 430075,China)
出处
《电力科学与技术学报》
CAS
北大核心
2021年第6期144-150,共7页
Journal of Electric Power Science And Technology
基金
湖北省技术创新专项(重大项目)(2018AAA049)。
关键词
互感器
故障诊断
变分模态分解
样本熵
K近邻分类器
current transformer
fault diagnosis
VMD
sample entropy
K-nearest neighbor classifier