标题:Method for diagnosis of on-load tap changer based on wavelet theory and support vector machine
作者:Duan, Xiaomu ;Zhao, Tong ;Li, Tan ;Liu, Jinxin ;Zou, Liang ;Li, Zhang
作者机构:[Duan, Xiaomu ;Zhao, Tong ;Li, Tan ;Liu, Jinxin ;Zou, Liang ;Li, Zhang ] School of Electrical Engineering, Shandong University, Jinan; 250061, China
来源:Journal of Engineering
出版年:2017
卷:2017
期:13
页码:2193-2197
DOI:10.1049/joe.2017.0719
摘要:In order to extract the vibration signal features of the on-load tap changer (OLTC) mechanical fault accurately, and enhance the quality of OLTC online monitoring, this study proposed a new method based on multi-resolution feature extraction algorithm and genetic optimisation support vector machine (GA-SVM). Firstly, the mechanical vibration signal of OLTC is decomposed by wavelet packet transform. Then, the information entropy of the wavelet packet coefficients of the effective frequency band is calculated. Based on above steps, multi-resolution feature parameters of the mechanical vibration signal of OLTC are obtained. Based on the multi-resolution feature parameters, fault classification model is established. Also, this model is on the basis of SVM which is optimised by genetic algorithm (GA). The results of simulation and analysis show that the integrated model proposed here can effectively realise the OLTC fault diagnosis, which can provide reference for the operation and maintenance of OLTC.
© 2017 Institution of Engineering and Technology. All rights reserved.
收录类别:EI
资源类型:会议论文
TOP