Hybrid Diagnosis System for Aeroengine Sensor and Actuator Faults

Abstract

This paper presents a hybrid model-based fault diagnosis system to distinguish different types of faults and provide accurate fault information under complex fault situations. In the proposed system, a bank of hybrid fault observers (HFOs) is designed for the actuator fault detection and sensor fault location, and a nonlinear fault-tolerance estimator (NFTE) is built for the actuator fault estimation. In addition, a switching structure is designed to vary the working process between banks of HFOs and the NFTE. To prove the effectiveness of the proposed approach, a series of simulations was conducted based on a nonlinear aeroengine model. The results verified that the proposed hybrid diagnosis system is capable of locating the sensor fault, detecting the actuator faults, and estimating the actuator faults effectively.