EKF-based fault detection and isolation for PMSM inverter

This paper deals with a new method PMSM inverter faults detection and isolation based on EKF (Extended Kalman Filter). The reliability of PMSM drivers has been dealt with an important factor in industrial applications requiring a precise operation and high performance. The fault of PMSM drivers may lead to the instability of the system, increase the cost and degrade the system's performance. Sometimes it can lead to the shut down or damage to the machine, even cause the safety problems for human. So the fault detection is necessary. Here, high risk switch open fault in PMSM inverter is detected. The resistance Ron in inverter switch FET is estimated by Extended Kalman Filter. If the open switch fault is happened, the estimation will be obviously higher than normal value. The fault can be detected by comparing the estimation with normal value of FET. Because the different switch is estimated in different rotator position field, the fault switch can be isolated. And the feasibility of the proposed scheme is proved by simulation and experimental results.