Sensor fault diagnosis strategy based on rotor flux observers in three-phase induction motor drive

Abstract

Sensors are crucial in induction motor drives (IMDs) by providing real-time feedback essential for precise and stable speed control. Malfunctions in speed or current sensors can seriously impair system performance and reduce reliability. This study proposes a fault-tolerant control (FTC) strategy that combines a rotor flux observer with a signal similarity-based diagnosis algorithm to detect, isolate, and compensate for sensor faults within the field-oriented control (FOC) framework. Once a fault is detected, the control system is reconfigured instantly by replacing faulty measurements with estimated values, ensuring stability during operation. The effectiveness of the method is evaluated through both simulation and experimental tests on a laboratory platform using a TMS320F28335 digital signal controller. Simulation results confirm the approach's ability to identify different fault types with minimal delay accurately. Real-time experiments further validate the method's practical application under typical operating conditions. Across all test scenarios, the system successfully maintains performance by smoothly switching between sensor-based and sensorless control modes. These results highlight the robustness and dependability of the proposed FTC scheme for IMDs experiencing sensor faults.