Fault detection in bearing using the Wavelet transform

Abstract

Faults in bearing are very common in various industrial segments. Monitoring its operating status through predictive techniques is essential to prevent unexpected failures. Thus, it is possible to increase the availability of the equipment inside the plant. Vibration Analysis is one of the most relevant monitoring parameters to assess the working condition of the equipment. However, the vibration signals from defects in bearings are of transient nature, therefore not being well analyzed by conventional analysis techniques. The purpose of this research is to present a study of the Wavelet Transform, a recent promising technique to detect bearing faults demonstrating its advantages and limitations using Matlab software. The defects were inserted in three different bearings mounted on an experimental bench. Results show the capability and feasibility of the Wavelet Transform, and its potential to be included in predictive maintenance programs.

Keywords

• Manutenção preditiva. Detecção de falhas. Rolamentos. Transformada de Wavelet.