Induction motors modeling and fuzzy logic based turn-to-turn fault detection and localization

The development of a turn-to-turn fault diagnosis system in induction motors stator windings is presented. The system consists in a digital faults simulator based on state variables and in an Adaptive Fuzzy Inference System (AFIS). The developed prototype was validated through comparisons between the digital simulator's results and experimental results in a three-phase, 5Hp, 4 poles, 60Hz motor. The fuzzy system uses as fault indicators the rotor slip, the negative-sequence current and the positive-sequence current angle. The proposed system is based on a sensorless data acquisition method (voltages and currents) and it is immune to variations of the load operating point. Another advantage is the faulty phase localization.