Detection of faults in gearboxes using acoustic emission signal

Abstract

Vibration analysis is widely used in machinery diagnosis, and wavelet transform and envelope analysis have also been implemented in many applications to monitor machinery condition. Envelope analysis is well known as a useful tool for the detection of rolling element bearing faults, and wavelet transform is used in research to detect faults in gearboxes. These are applied for the development of the condition monitoring system for early detection of the faults generated in several key components of machinery. Early detection of the faults is a very important factor for condition monitoring and a basic component to extend CBM (Condition-Based Maintenance) to PM (Prediction Maintenance). The AE (acoustic emission) sensor has a specific characteristic on the high sensitivity of the signal, high frequency and low energy. Recently, AE technique has been applied in some studies for the early detection of machine fault. In this paper, a signal processing method for AE signal by envelope analysis with discrete wavelet transforms is proposed. Through the 15 days test using AE sensor, misalignment and bearing faults were observed and early fault stage was detected. Also, in order to find the advantage of the proposed signal processing method, the result was compared to that of the traditional envelope analysis and the accelerometer signal.

Development and trend of condition monitoring and fault diagnosis of multi-sensors information fusion for rolling bearings: a review

“…Gu Et al. [61] used the envelope analysis and discrete wavelet analysis to deal with the acoustic emission signal of the gearbox fault. Tests results showed it has a higher accuracy compared to the traditional envelope analysis.…”

The Use of the Acoustic Emission Method to Identify Crack Growth in 40CrMo Steel

“…A more practical crack identification and detection method is still lacking. Aiming to improve recognition accuracy and generalization, in papers [29,30,31] a novel crack identification method based on acoustic emission and pattern recognition is proposed. In these methods, the sound pulses from cracks are firstly separated from the original signal by pre-processing.…”

A Method to Decompose the Streamed Acoustic Emission Signals for Detecting Embedded Fatigue Crack Signals

“…The common sources that generate AE in gearbox include plastic deformation, microfracture, wear, bubbles, friction and impact [12]. Once transient signals are collected, signal processing methods, such as wavelet decomposition [13], empirical mode decomposition [6], and multivariate pattern recognition [14] are applied. Typical parameters extracted from the transient signals are root mean square value, frequency domain characteristics, energy, spectral kurtosis, and peak-to-peak vibration level.…”