Title: Intelligent fault diagnosis of three-phase asynchronous motor based on PCA-SVCNN

Authors: Lingzhi Yi; Xiu Xu; Jian Zhao; Wang Li; Junyong Sun; Yue Liu

Addresses: College of Automation and Electronic Information, Xiangtan University, Xiangtan, Hunan, China ' Hunan Province Multi-Energy Cooperative Control Technology Engineering Research Center, College of Automation and Electronic Information, Xiangtan, Hunan, China ' Hunan Province Multi-Energy Cooperative Control Technology Engineering Research Center, College of Automation and Electronic Information, Xiangtan, Hunan, China ' State Key Laboratory of High Power AC Drive Electric Locomotive System Integration, CRRC Zhuzhou Locomotive CO., LTD, Zhuzhou, Hunan, China ' State Key Laboratory of High Power AC Drive Electric Locomotive System Integration, CRRC Zhuzhou Locomotive CO., LTD, Zhuzhou, Hunan, China ' Key Laboratory of Intelligent Computing and Information Processing, Ministry of Education, Xiangtan, Hunan, China

Abstract: In order to solve the problems caused by the complex motor structure signals and big data of non-stationary machinery in the traditional asynchronous motor fault diagnosis method, the speed and accuracy of three-phase asynchronous motor fault diagnosis are improved. In this paper, a new fault diagnosis method of three-phase asynchronous motor is proposed. Firstly, the principal component analysis (PCA) method is used to reduce the dimension of the collected current data, and then the support vector machine (SVM) is used to realise the two classification of the data. Finally, the two types of data are classified by convolutional neural network (CNN), and the accurate diagnosis of three-phase asynchronous motor fault can be realised. The simulation results show that the proposed algorithm can improve the accuracy of fault classification quickly and effectively, which is of great significance to the accurate diagnosis of motor faults.

Keywords: principal component analysis; PCA; support vector machine; SVM; neural network; fault diagnosis.

DOI: 10.1504/IJAMECHS.2021.116458

International Journal of Advanced Mechatronic Systems, 2021 Vol.9 No.2, pp.66 - 76

Received: 27 May 2020
Accepted: 12 Oct 2020
Published online: 26 Jul 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article