Application of Zhao-Atlas-Marks Transforms in Non-stationary Bearing Fault Diagnosis

Abstract

Rolling element bearings (Ball Bearings) are the main rotating element in mechanical engineering applications such as Thermal Power plants, Nuclear power plants, Aviation and Chemical industries. The defects in the rolling element bearings may arise due to reasons such as overloading, fatigue, improper design and manufacturing of the bearing, misalignment of bearing races, etc. Depending on the application, the speed and load conditions of shaft may cause some failures which leads to non-stationary operating conditions. Since early fault detection can save emergency maintenance cost, the bearing fault diagnosis is important in monitoring applications. This paper is attempt to analyze the effectiveness of the new time-frequency distributions called the Zhao-Atlas-Marks (ZAM) distribution to enhance non stationary vibration signal analysis for fault diagnosis in bearings. Also the performance of ZAM with Short Term Fourier Transform (STFT) is discussed in this paper.