Aircraft hydraulic system priority valve fault diagnosis research based on WOA-CNN-BiLSTM

The priority valve, a critical component of the flight control system, is responsible for prioritizing oil supply to the main flight control system when hydraulic system pressure is low. Maintaining its stability is essential for ensuring flight safety. However, the aircraft's hydraulic system only has pressure sensors, making traditional diagnostic methods difficult to implement. To overcome this challenge, this study presents a fault diagnosis algorithm that combines the whale optimization algorithm, convolutional neural network, and bidirectional long and short-term memory network. By analyzing the differential pressure signal between the inlet and outlet of the priority valve, the algorithm uses a fusion of CNN and BiLSTM to identify fault features. The hyperparameters are optimised through the utilisation of the WOA algorithm to avoid local optima. Results from a multidisciplinary joint simulation using AMESim and Matlab demonstrate that the proposed method achieves an average diagnostic accuracy of over 95% in complex working conditions. Compared to existing methods, this approach demonstrates significant improvements in diagnostic accuracy and stability, providing a solution to the challenging diagnosis of aircraft hydraulic systems.