Switch fault diagnosis of PM brushless DC motor drive using adaptive fuzzy techniques

An adaptive neuro-fuzzy inference system (ANFIS) is developed to diagnose open switch faults of PM brushless dc motor drives. Features extracted under healthy and faulty operations using wavelet transform are used to train ANFIS. Testing of the proposed diagnostic system shows it could not only diagnose the fault but identify the faulty switch as well. Good agreement between experimentation and simulation is obtained.