Remaining Useful Life Prediction Based on Improved Transformer Network for Pneumatic Control Valve

Pneumatic control valve plays a crucial role in the industrial control system. This paper proposes a hybrid neural network model for predicting the remaining useful life (RUL) of pneumatic control valve system. Firstly, highly related features are selected by the pearson correlation coefficient (PCC) to minimize the computational burden of the model. Then, convolutional neural networks (CNN) is employed to extract local features of sequence data, and Transformer is used to capture global representations. Finally, all the features are fed into the CNN-Transformer model. The dataset generated by development and application of methods for actuator diagnosis in industrial control systems (DAMADICS) platform is verified. The experimental result shows that the proposed hybrid model has high prediction accuracy and stability.