Differential Mathematical Morphological-Based Online Diagnosis of Stator Interturn Failures in Direct Torque Control Drive Systems

Abstract

This article presents an in-service noninvasive fault diagnosis (FD) routine to detect incipient stator's interturn failure (ITF) in the direct torque control (DTC) driven induction motors. The developed FD routine is based on the application of the mathematical morphological gradient (MMG) technique on the stator currents. Being a time-domain technique, it is less computational burdensome than the frequency-domain-based FD. To detect the faulty phase, we introduced and implemented the differential MMG (ΔMMG) algorithm concept. A comprehensive electromagnetic investigation of harmonics and interharmonics contents due to the ITF, the DTC controller's reaction, the inverter's switching activity is presented. A design guideline of selecting the structuring element length in the FD routine was introduced. A physics-based cosimulation platform was built to investigate the feasibility of the proposed FD algorithm. The simulation reinforced by the experimental results depicted the robustness of the FD method. The comparative analysis is presented between the developed ΔMMG and the conventional motor current signature analysis (MCSA). The experimental results proved that developed FD is faster and more robust than the MCSA in detecting the ITFs at their embryonic stages.