A novel nonlinear observer for fault diagnosis of induction motor. (May 2020)

• Record Type: Journal Article
• Authors: Yi, Lingzhi; Liu, Yue; Yu, Wenxin; Zhao, Jian
• Is Part Of: Journal of algorithms & computational technology. Volume 14(2020)
• Publication Date: 2020-05
• Subjects: BP neural network -- Cuckoo Search algorithm -- fault diagnosis -- induction motor -- nonlinear observer; Computer algorithms -- Periodicals; Numerical calculations -- Periodicals; Computer algorithms; Numerical calculations; Periodicals; 518.1
• DOI: 10.1177/1748302620922723
• Languages: English
• ISSNs: 1748-3018

Abstract

In order to accurately diagnose the fault of induction motor, a fault diagnosis of nonlinear observer method based on BP neural network and Cuckoo Search algorithm is proposed. It is a new method which mixes analytical model and artificial neural network; firstly, the induction motor model is divided into linear and nonlinear parts, and BP neural network is used to approximate the nonlinear part. Then an adaptive observer is established, in which a simple and effective method for selecting the feedback gain matrix is offered. Cuckoo Search algorithm is utilized to improve the convergence speed and approximation accuracy in BP Neural Network. Compared with some other algorithms, the simulation results show that the proposed method has higher prediction accuracy. The designed nonlinear observer can estimate the current and speed accurately. Finally, the experiment of winding fault is implemented, and the online fault detection of induction motor is realized by analyzing the current residual errors.