Survey of Gear Fault Feature Extraction Methods Based on Signal Processing

Abstract

Gear fault diagnosis technology is significant in reducing casualties and economic losses caused by industrial accidents. Signal processing is an important step in the diagnosis of gear faults, which affects the accuracy of fault recognition seriously. Traditional signal processing method can be divided into three categories: time domain, frequency domain and time-frequency domain. For stationary signals, feature extraction methods can be divided into two categories: time domain and frequency domain. The time-frequency analysis method is more suitable for dealing with non-stationary signals, which can effectively reflect the distribution of non-stationary signals in the time domain and frequency domain. This paper focuses on elaborating various signal processing methods, analyzing their advantages and disadvantages, summarizing existing research results and problems, and looks forward to future research directions.