Research on Aero-Engine Sensor Failure Diagnosis with Dual Redundant Predictors Based on Neutral Network

Abstract

In consideration of the common sensor failures in aero-engine control system, a new approach is proposed using dual redundant predictors based on neutral network in this paper. The neutral network temporal redundant predictor and spatial redundant predictor are created over the time series redundant information of single sensor and the space redundant information of multi-sensor respectively. The threshold-value differentiate method is applied for real-time sensor failure detection. Isolate the fault sensor and use the outputs of spatial redundant predictor to accommodate it. Digital simulation results show that this method is effective and feasible.