Motor bearing fault diagnosis based on local characteristic-scale decomposition and support vector optimized by sine cosine algorithm

Abstract

When the motor bearing fails, the nonlinear and unstable characteristics of the vibration signal will lead to a low fault identification accuracy. This paper proposes a fault diagnosis method for motor bearings based on local characteristic-scale decomposition (LCD) and sine-cosine algorithm to optimize support vector machine. In this paper, the LCD algorithm is used to decompose the vibration signal, and screened out the effective intrinsic scale components (ISC) component. Then, Sample Entropy and Permutation Entropy are extracted from the filtered components. For the selection of parameters c and g of SVM, this paper introduces the sine and cosine optimization algorithm (SCA) for position update optimization, and builds an SVM model to obtain a classification model for motor bearing faults after training on the data set. After importing the remaining test data set into the classification model for testing, the fault classification accuracy results of the motor bearing can be obtained. The results of the simulation experiments show that the recognition rate of motor bearing faults based on LCD and SCA-SVM is 98.75%.