The TZ-Multi-Level Inverter Fed Multi-Phase Induction Motor Drive Systems Fault Diagnostics

Abstract

More and more high-power, safety-critical industries including electric vehicles, aircraft, and renewable energy are using multiphase induction motor drive systems (MPIMDS) because they are more fault-tolerant and have better torque density. The Trans-Z-Source Multi-Level Inverter (TZ-MLI) topology has been added to boost performance even more. It has lower switching losses, larger voltage gain, and better harmonic performance. But the reliability of these systems is greatly affected by possible problems, such as inverter switch failures and stator inter-turn short circuits, which cause current distortion, torque pulsations, and lower efficiency. This study describes a complete fault diagnosis framework for a five-phase induction motor drive system powered by a TZ-MLI. A comprehensive
mathematical model of the TZ-MLI and MPIMDS is developed, succeeded by fault analysis under various operating scenarios. Researchers are looking into new diagnostic methods, such as negative-sequence current monitoring and classification based on convolutional neural networks (CNNs). A MATLAB/Simulink model is created to mimic both normal and abnormal situations. The results show that the suggested system keeps total harmonic distortion (THD) below 3 perentage during normal operation. However, it does show clear current and torque problems when there is a failure. The CNN-based diagnostic method gets more than 95 percntgae of the classifications right, which is better than traditional Fast Fourier Transform (FFT) and wavelet-based methods in noisy settings. The suggested method is very reliable, can find faults quickly, and is very strong, which makes it perfect for modern fault-tolerant motor drive applications using a five phase induction motor drive system.