Abstract
An observer-based fault detection scheme utilizing the interacting multiple model (IMM) approach is presented for an industrial actuator benchmark problem. The requirement is to detect current and position faults in a brushless D.C. motor which is subject to both actuator and sensor faults in the presence of unknown load torques. Fault detection is carried out by threshold checking on the model probabilties calculated by the algorithm. The proposed scheme has been applied to the given benchmark data with current and position fault detection achieved in a good detection time and with a high detection probabilty, regardless of the load torque.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have