Multiple Fault Diagnosis of Aeroengine Control System Based on Autoassociative Neural Network

Aeroengine is a kind of complicated thermal machinery which works under high speed, high load and high temperature for a long time. In order to ensure the high reliability and stability of the engine, accurate and effective fault diagnosis is essential. The traditional model-based fault diagnosis method is difficult to achieve satisfactory results. The emergence of neural network intelligent algorithm provides a new idea. In order to obtain a fault diagnosis system with strong robustness and high detection rate, we design a the Autoassociative Neural Network (AANN) group to complete the detection and isolation of engine sensor faults and component faults, as well as the reconstruction of sensor faults. Firstly, the signal of the sensor of the aeroengine control system was preprocessed, and then a group of AANN network was designed according to the fault parameters for multiple fault detection and isolation of aeroengine. Finally, it was verified based on the MATLAB/Simulink platform. It is worth mentioning that this method does not require a model. It can be seen from simulation results that the proposed method can effectively reduce the noise of measurement data. Moreover, it has the advantages of fast diagnosis speed, strong robustness and synchronous detection and isolation. And it can effectively detect, isolate and reconstruct the faults of aeroengine.