Bearing defect detection and diagnosis using a time encoded signal processing and pattern recognition method

Abstract

Many new bearing monitoring and diagnosis methods have been explored in the last two decades to provide a technique that is capable of picking up an incipient bearing fault. Vibration analysis is a commonly used condition monitoring technique in world industry and has proved an effective method for rolling bearing monitoring systems. The focus of this paper is to combine two conventional methods: wavelet transform and envelope analysis with the Time Encoded Signal Processing and Recognition (TESPAR) to develop a better technique for detection of small bearing faults. Results show that TESPAR with these two combinations provides good fault discrimination in terms of location and severity for different bearing conditions.