In order to improve the RUL prediction accuracy of aircraft engines

this paper proposes a stacked convolutional-long short-term memory network (SDCNN-LSTM) model to solve the mapping relationship of engine degradation characteristics. Firstly

the data monitored by multiple sensors is normalized to reduce the dimensionality of the data and make it fall within a small specific range

thus eliminating the influence of singular samples; Secondly

an SDCNN-LSTM neural network prediction model is constructed

wherein the dilation rate in the SDCNN is used to increase the receptive field of the convolutional layer while keeping the computational cost low. This enables the model to capture patterns that occur at various time scales; Then

the model uses a self-attention mechanism to encode features in order to extract time series features before adding LSTM layer to capture the temporal dependencies in the time series data and predict the next time step

thus effectively predicting the RUL of aircraft engines; Finally

the model is optimized using the Hyperband algorithm to improve its prediction accuracy

and the proposed method is validated using the C-MAPSS dataset of turbofan engine degradation processes. Experimental results show that the proposed method is superior to other models

verifying the feasibility and effectiveness of the method.