BiLSTM-Based Aircraft Engine RUL Prediction Method Integrated with Attention Mechanism

To address the issues of insufficient accuracy, poor model generalization ability, and difficulty in capturing longrange temporal dependencies in aircraft engine remaining useful life prediction, this study proposes a prediction model integrated with an attention mechanism. The model first achieves data augmentation through wavelet transform, then employs a bidirectional long short-term memory network for prediction, and introduces the attention mechanism to dynamically assign weights to each time step, before finally outputting the remaining useful life prediction value through a fully connected layer. On the Commercial Modular Aero-Propulsion System Simulation dataset generated by commercial modular aero-propulsion system simulations, compared with existing mainstream models, the proposed model achieves the minimum Root Mean Square Error values in two of the four data subsets and the second smallest Root Mean Square Error values in the other two subsets, which verifies that the model can effectively improve the prediction accuracy.