Motor Bearing Fault Prediction Using Artificial Intelligence Techniques

Early detection and diagnosis of faults that are only starting to appear are essential for the examination, reliability, trust, and greater operational efficiency of induction motors driven by mains electricity. A model-based fault detection method for induction motors is created to identify the state of the motor. In this work, both the inner race fault and the outer race fault in the bearings are considered. The achieved accuracy % serves as evidence of the diagnosing system's efficacy. Based on comparing different algorithms, the highest accuracy achieved is 85% for Random Forest Classifier under different health conditions. Implementation of the machine learning model on the given dataset is done with the help of Python. The suggested approach can accurately and quickly identify the defect.