摘要
针对目前配电变压器长圆形绕组的故障诊断研究较少的问题,本文提出一种基于灰狼优化(GWO)算法的变压器长圆形绕组机械故障诊断的方法。首先,建立长圆形绕组的辐向和轴向振动数学模型;其次,建立变压器电磁—机械耦合有限元模型,计算长圆形绕组在正常、松动以及翘曲状态下的振动分布,选取R点作为特征值提取点;最后,对变压器样机进行正常、松动以及翘曲3种状态试验,获取R点的小波包能量特征,采用GWO算法优化支持向量机(SVM)参数对变压器长圆形绕组机械故障诊断,最终优化后的诊断综合准确率达到90%。
At present,aiming at the problem that there are few studies on fault diagnosis of distribution transformer oblong winding,a method for mechanical fault diagnosis based on grey wolf optimization(GWO)algorithm is proposed.Firstly,a mathematical model for the radial and axial vibration of the oblong winding is established.Secondly,an electromagnetic-mechanical coupling finite element model for transformer is established.The vibration distribution of the oblong winding in normal,loose and warped conditions is calculated,and the R point is selected as the feature value extraction point.Finally,the transformer prototype is tested in normal,loose and warped states to obtain the wavelet packet energy characteristics of the R point.The support vector machine(SVM)parameters are optimized using GWO algorithm to diagnose the mechanical faults of the transformer oblong winding.The final optimized diagnosis has a comprehensive accuracy of 90%.
作者
陈朝阳
杨文荣
张雨蒙
石小晖
CHEN Zhaoyang;YANG Wenrong;ZHANG Yumeng;SHI Xiaohui(State Key Laboratory of Reliability and Intelligence of Electrical Equipment,Hebei University of Technology,Tianjin 300130,China)
出处
《传感器与微系统》
CSCD
北大核心
2024年第3期63-66,75,共5页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(51877066)。
关键词
配电变压器
长圆形绕组
有限元仿真
灰狼优化算法
故障诊断
distribution transformer
oblong winding
finite element simulation
grey wolf optimization(GWO)algorithm
fault diagnosis
