A neural network approach to instrument fault detection and isolation

An instrument fault detection and isolation (IFDI) technique based on the use of an artificial neural network (ANN) is proposed. The ANN input layer is fed by instrument outputs, and its output layer gives information for instrument diagnosis. The method adopted is described in detail and tested on a complex automatic measurement station for induction motor testing. The performance of the proposed IFDI scheme is experimentally evaluated mainly in terms of correct diagnosis, incorrect fault isolation, missed fault detection, and false alarms. The proposed diagnostic scheme performs well also out of the domain in which it was trained.<>