An Efficient and Robust Open-Phase Fault Diagnosis Scheme for Dual Three-Phase PMSMs Drive Using Axis Transformation

Abstract

Efficient diagnostic strategies that can be integrated into microcontroller are still important to the industrial equipment, before implementing artificial intelligence (AI) models with edge computing. At present, the signal-based open-phase fault (OPF) diagnosis strategy can meet this goal, but it has problems such as multithreshold and nonstationary process sensitivity. The current distortion of multiphase drive after OPFs is more serious, and it is almost impossible to provide intuitive threshold and diagnostic indicators. In this article, a simple OPF diagnosis method for dual three-phase permanent magnet motors is presented. The axis transformation (AT) is creatively used for OPF diagnosis. Specifically, the six-phase currents of the motor are transformed into six reconstructed axis frames, and then, the ratio combinations of the axis currents can be employed to detect and localize OPFs. Importantly, these ratio combinations of axis currents are independent of motor parameters, speed, and load; they can achieve OPF diagnosis directly, without being affected by thresholds and operating conditions. The experimental results demonstrate that the proposed strategy can be executed online efficiently and achieve high-precision OPF diagnosis within one-fourth of the current cycle.