Intelligent fault monitoring and diagnosis in electrical machines

The aim of this paper is to develop an intelligent diagnosis method for fault detection and isolation in induction motors. We consider failures in three components of induction motor: bearing, stator winding and rotor winding. Firstly, a model-based nonlinear observer in the proposed method is designed based on available information. The fault detection decision is carried out by comparing the model-based observer speed with their signatures. Secondly, multiple state observers are constructed based on possible fault function set. The fault isolation decision is made by checking each residual generated by observer state estimation. Finally, simulation tests are given to verify the effectiveness of the proposed fault diagnosis scheme.