Load Fault Diagnosis in Induction Motor using Artificial Intelligence Algorithm

Industries require early diagnosis of induction motor faults to avoid complete failure. The use of machine learning and condition monitoring to detect faults has huge promise. Machine learning can be used to detect motor faults. To avoid losses, induction motor faults must be rectified promptly. Fault detection using machine learning algorithms is an excellent method for preventive maintenance. If any fault arises, the motor may continue to process and produce failure in windings core etc. Hence, the faults can be prevented by the monitoring the motor output values and cutoff the power before the motor gets damaged. This research develops a machine learning strategy based on algorithms in order to learn the characteristics from vibration signal’s frequency distribution. This is mainly to characterize the operational status of induction motors. The operational status includes the parameters such as temperature, voltage and current. It automates and intelligently diagnoses faults by combining feature extraction and categorization.