Prediction Method of Aeroengine Residual Life Based on Stacked Sparse Automatic Encoder

Abstract

In order to ensure the safe and reliable operation of the aircraft, improve the efficiency of aviation engine maintenance and improve the prediction accuracy of the remaining life of the aeroengine, a prediction method of the remaining life of the aeroengine based on the stacked sparse self- coding neural network is proposed. The method firstly constructs a plurality of self-encoding networks to form a deep stack self- encoding network, selects the state data of the engine as the training input of the network, and enables the network to extract the distributed rules between the data layer by layer intelligently, thereby constructing the engine degraded stack self-encoding learning. model. The BP residual neural network is used to classify the remaining life of the engine as a result of engine residual life prediction. Finally, the algorithm is validated by the PHM2008 aeroengine degradation data. The results show that the method can effectively predict the remaining life of the aeroengine.