Research on Remaining Useful Life Prediction Method Based on Multi-Channel Deep Neural Network

Remaining Useful Life prediction is crucial for ensuring operational safety and reducing maintenance costs. This paper leverages data from a NASA engine model and proposes a novel multi-channel deep learning framework that integrates Convolutional Neural Networks, Bidirectional Long Short-Term Memory, and attention mechanisms, while incorporating Bayesian Neural Networks for uncertainty quantification. Experiments on the dataset demonstrate that the proposed method outperforms existing state-of-the-art approaches in key metrics, including Root Mean Squared Error, Mean Absolute Error, and the PHM08 score..