Abstract:Sensor health assessments are of great importance for accurately understanding the health of an aeroengine, supporting maintenance decisions, and ensuring flight safety. This study proposes an intelligent framework based on a physically guided neural network (PGNN) and convolutional neural network (CNN) to diagnose sensor faults under dynamic conditions. The strength of the approach is that it integrates information from physics-based performance models and deep learning models. In addition, it has the structure of prediction–residual–generation-fault classification that effectively decouples the interaction between sensor faults and system state changes. First, a PGNN generates the engine’s non-linear dynamic prediction output because the PGNN has the advantage of being able to handle temporal information from the long short-term memory (LSTM) network. We use a cross-physics–data fusion scheme as the prediction strategy to explore the hidden information of the physical model output and sensor measurement data. A novel loss function that considers physical discipline is also proposed to overcome the performance limitations of traditional data-driven models because of their physically inconsistent representations. Then, the predicted values of the PGNN are compared with the sensor measurements to obtain a residual signal. Finally, a convolutional neural network (CNN) is used to classify faults for residual signals and deliver diagnostic results. Furthermore, the feasibility of the proposed framework is demonstrated on an engine sensor fault dataset. The experimental results show that the proposed method outperforms the pure data-driven approach, with the predicted RMSE being reduced from 1.6731 to 0.9897 and the diagnostic accuracy reaching 95.9048%, thereby confirming its superior performance.

Fault recognition for engine system based on variable structure of process RBF neural network

Application of Neural Networks to Chemical Fault Diagnosis

Multiple Fault Diagnosis of Aeroengine Control System Based on Autoassociative Neural Network

Fault Diagnosis and Reconstruction for Sensor of Aeroengine Control System Based on AANN Network

Fault Diagnosis Method Based on Improved Genetic Algorithm and Neural Network

Study of Diesel Engine Fault Diagnosis Based on Bayesian Network

Research on Mechanical Fault Prediction Based on Improved Neural Network

Fault Diagnosis of Aeroengine Control System Sensor Based on Optimized and Fused Multidomain Feature.

Aeroengine Intelligent Fault Diagnosis System Based on Hidden Variable Mechanism

Fault Diagnosis in Flight Control System Based on Improved Algorithm of Neural Network

Data Refinement Approach to Diesel Engine Fault Diagnosis Based on Bayesian Networks

Fault Diagnosis of Aero-Engine Based on Self-Adaptive Neural Network

Physics-Guided Neural Network Model for Aeroengine Control System Sensor Fault Diagnosis under Dynamic Conditions

Fault Detection and Diagnosis for Nonlinear System Based on Neural Network On-Line Approximator

Fault Diagnosis Expert System of Automobile Engine Based on Neural Networks

Fault Diagnosis of Automobile Engine Based on Improved BP Neutral Network

Neural Network-based Online Fault-tolerance Design for Flight Control Systems

A Sensor Fault Diagnosis Method Based on Robust Auto-associative Neural Network

Neural Network-Based Sensor Online Fault Diagnosis And Reconfiguration For Flight Control Systems

Research On A Novel Rbf Neural Network And Its Application In Fault Diagnosis

Application of BP Neural Network to Fault Diagnosis of High-Pressure Common Rail Fuel System of Electronic Control Diesel Engines