Designing an Intelligent Decision Support System for Traction Motor Fault Detection using Artificial Intelligence

Abstract

Inverters and inductive instruments have been used with electric motors in recent years. They must be analyzed together to diagnose the inverter and induction machine properly. Inverter defects may affect application performance, increasing cost and diminishing efficiency. AI has improved defect detection in electric cars and industrial equipment traction motors. Safety and performance need efficient detection. Real-time monitoring and AI-powered decision assistance decrease operational disruptions. Analyzing massive sensor data reveals discrepancies for accurate predictions and preventive maintenance. Analysis of data and interactions notifies maintenance staff. This full strategy accelerates issue discovery, diagnosis, and maintenance. This work designs an artificial intelligence-based Multiple Electrical Units (MEU) for a fault detection system of traction motors to identify faults and ensure safe and stable operation reliably. This study first uses the system's optical fiber sensor, digital-to-analog converter module, system control unit, and phase-locked loop module to identify traction motor problems using artificial intelligence. Second, the optical fiber sensor collects the motor signal, and phase-locked loops monitor the traction motor's frequency. Third, improved cuckoo and difference analysis detects motor problems. The enhanced cuckoo algorithm can overcome the poor efficiency of motor fault identification, and the diverse analyses can minimize fundamental wave interference to a motor fault and combine their benefits to complete traction motor fault detection. The presented defect detection method performs well in experiments. The proposed approach improves motor defect detection accuracy, reduces error, and ensures MUE safety and stability.