A Fault Classification Method for Rolling Bearing Based on Multisynchrosqueezing Transform and WOA-SMM

Abstract

To solve the problem that the limited time-frequency features cannot fully represent the deep-seated state information of rolling bearing, the time-frequency analysis method, whale optimization algorithm (WOA) and support matrix machine (SMM) are combined, and a fault diagnosis model based on multisynchrosqueezing transform (MSST) and WOA-SMM is proposed. First, the time-frequency trait of the original signal is extracted by MSST. Then, using the time-frequency spectrum processed by MSST as the input of SMM, MSST solves the problem of state information loss when constructing a characteristic matrix. Finally, the parameters of the SMM are optimized by WOA and the ideal parameters can be obtained adaptively, and the problem of setting parameters subjectively is solved. The experimental analysis of the two datasets shows that WOA-SMM is superior not only to other classifiers in classification performance, but also has higher in convergence accuracy and speed for rolling bearing fault diagnosis. INDEX TERMS WOA-SMM; Whale optimization algorithm; Fault diagnosis; Rolling bearing; Multisynchrosqueezing transform I. INTRODUCTION

Rolling bearing fault diagnosis based on fine-grained multi-scale Kolmogorov entropy and WOA-MSVM

“…Whale Optimization Algorithm (abbreviated as WOA) was proposed by Mirjalili in 2016, which mainly stems from the blister net feeding mechanism of humpback whales, and this algorithm has some advantages such as high optimization seeking accuracy, few parameters, and simple structure. WOA can be divided into three stages including enclosure hunting, bubble net predation and spiral updating, and searching for prey [ 28 ].…”

Rolling Bearing Fault Diagnosis Based on Time-Frequency Compression Fusion and Residual Time-Frequency Mixed Attention Network

“…They proved that the error between the time-frequency representation obtained and the ideal case becomes smaller with the increase of iterations. In other words, this method can theoretically approach the ideal time-frequency representation infinitely, which makes it widely used in the field of bearing fault diagnosis [13][14][15][16][17][18].…”