Sensor Fault Diagnosis and Reconstruction of Engine Control System Based on Autoassociative Neural Network

Abstract

The topology and property of Autoassociative Neural Networks (AANN) and the AANN's application to sensor fault diagnosis and reconstruction of engine control system are studied. The key feature of AANN is feature extract and noise filtering. Sensor fault detection is accomplished by integrating the optimal estimation and fault detection logic. Digital simulation shows that the scheme can detect hard and soft failures of sensors at the absence of models for engines which have performance deteriorate in the service life, and can provide good analytical redundancy.