Aero-engine Sensor Fault Diagnosis Based on Convolutional Neural Network

A fault diagnosis method based on convolutional neural network (CNN) is proposed for aero-engine sensors. Convolutional layers and pooling layers in CNN are used to extract correlation features among sensed signals, based on which fully connected layers are used to diagnose sensor faults. The inception module method is used to extract features on different data sizes among sensors, making the method capable of detecting multi-sensor faults. The simulation results based on a turbofan engine model demonstrate the feasibility and effectiveness of this method.