Fault detection and isolation based on nonlinear analytical redundancy applied to an induction machine

Abstract

This paper deals with the design of a fault detection and isolation (FDI) of the sensors for an induction machine using nonlinear analytical redundancy (NLAR). This paper investigates the detection and isolation of faults using elimination of unknown variables of the system and in particularly the unknown system states. The induction machine (IM), it is highly nonlinear, multivariable, time-varying system and particularly when subject to the faults, it is difficult to detected them by linear approaches. The nonlinear parity space algorithm is able to detect and isolate sensor faults such IM speed and stator currents or actuator faults (stator voltage). In order to prove the accuracy of approaches studying in this work, simulation results will be given.