Fault Detection for Brushless Direct-Current Motor Using Descriptor System-Based Set-Membership Estimation

Abstract

Brushless direct-current (BLdc) motors are pivotal in electric vehicles, drones, and industrial systems due to their high efficiency and reliability. However, faults in stators, rotors, or inverters may degrade performance. In this article, we focus on the problem of model-based fault detection of BLdc motors. First, a high-fidelity model of the BLdc motor is developed, explicitly incorporating inverter switching behaviors, winding, back EMF, rotor inertia, and Hall sensors, which is formulated as a discrete-time-varying descriptor system. Based on this model, a fault detection method is proposed using a set-membership estimation theory. The proposed BLdc motor model has higher fidelity, and the fault detection method has more relaxed design conditions. Finally, a hardware-in-the-loop (HIL) platform, including a BLdc motor, is established. After that, the platform is used to validate the fidelity of the proposed BLdc motor model and the effectiveness of the fault detection method.