Multitask Learning-Based Self-Attention Encoding Atrous Convolutional Neural Network for Remaining Useful Life Prediction

Abstract

Any failure of the turbofan engine, as one of the key components of space shuttle, can lead to serious accidents. Therefore, it is necessary to predict the remaining useful life (RUL) to guarantee its reliability and safety. This paper proposes a multi-task learning-based self-attention encoding atrous convolutional neural network called MSA-CNN to effectively realizes RUL prediction. Specifically, in order to extract fault feature information, an atrous convolutional neural network (ACNN) is used as the auxiliary task network, which is more efficient than the traditional CNN in the process of down sampling. Moreover, a model with ACNN and self-attention encoder (SAE) is used as the main task network to capture short-long term dependencies in a time sequence and thus realize RUL prediction. Compared with other recurrent neural networks, SAE proposed in this paper has the advantage of parallel computation. Besides, a novel multi-tasking loss function is also proposed to realize the interaction among multiple tasks. After MSA-CNN experiments on four subsets of C-MAPSS dataset, the RMSE average between the predicted RUL and the real value is about 14.66, which is better than the existing methods. Several other comparative experiments were conducted to verify the benefits of each submodule.