Identification Method of Motor Bearing Fault Location and Degree Based on Quadratic Discriminant Analysis

Abstract

The fault diagnosis of rolling bearings is of great significance to improve the efficiency of industrial production and ensure the stable and safe operation of industrial production. In order to improve the accuracy rate of rolling bearing fault location identification and analyze the fault degree, a fault diagnosis method of motor bearing based on the quadratic discriminant analysis method is proposed. The experimental results show that, compared with the traditional intelligent algorithm, the model takes the original vibration signal as the research object and can perform fault location diagnosis and fault degree analysis without going through complex feature extraction procedures. Moreover, with the increased number of vibration samples points, the accuracy of the diagnosis also continues to improve. Finally, comparing the quadratic discriminant analysis method with other intelligent algorithms, it is found that when the vibration sample length is 80, the quadratic discriminant analysis method improves the diagnostic accuracy by 44.2% on average and shortens the training time by 87.5% on average compared with other methods.