Inner and Outer Bearing Fault Diagnosis of electrical Motors Using a Proposed Algorithm and Vibration Signals

This paper presents an approach for detecting bearing faults by using vibration signals. The proposed approach encompasses a classification algorithm incorporating preprocessing, window selection, feature selection and extraction, classification methods, and validation. This algorithm regards the Statistical features and harmonics of fault-related frequencies obtained from vibration signals as indicators of faults. The application of this approach is demonstrated in three cases: inner-race, outer-race, and inner and outer-race bearing fault detection. Inner and outer bearing faults have been investigated in three different cases of 0.3, 1, and 3 millimeters, and the proposed algorithm has detected each of these cases separately. The impact of window length on classification accuracy is investigated in this study, considering various feature sets. Additionally, a new feature set is introduced to enhance accuracy. Furthermore, a comparative assessment of different classification algorithms is conducted. The experimental findings indicate that the proposed method is capable of effectively identifying bearing faults and their status, thereby enhancing the accuracy of fault diagnosis.