Session: 05-05 Data-Driven Methods and AI for Diagnostics

Paper Number: 127081

127081 - Implementation of Artificial Intelligence for Aircraft Engine Health Monitoring and Prognostics 

Improving the availability and reliability of aircraft engines is of paramount importance in managing the aircraft fleet’s efficiency. While previous efforts have primarily focused on condition-based monitoring and Remaining Useful Life (RUL) prediction based on physics-based models, this paper introduces a novel approach to Engine Health Monitoring (EHM) using deep learning models. In particular, this work leverages critical engine parameters such as surge margin, exhaust gas temperature margin for interpretable EHM and prognostics. We present three deep learning models, namely, the Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and Long-Short Term Memory (LSTM), optimized for these tasks.  The models were trained using synthetic data of 100 CFM 56 5B Turbofan engine-inspired model, simulating various flight cycles at steady-state cruise conditions using TURBOMATCH software (Cranfield University in-house aircraft engine performance simulation tool). The degradation in each engine was based on mass flow capacity and efficiency variation in fan, compressor, and turbine which is the effect of fouling, erosion, corrosions etc. Unlike the existing datasets, this study deployed full factorial degradation of engine components along with a wide range of degradation scenarios.

Results demonstrate the competitiveness of the proposed models, as evidenced by low Root Mean Square Error (RMSE) values. The CNN model excels in health monitoring, achieving an RMSE of 0.0148 health margin prediction, while the LSTM model proves most effective in predicting Remaining Useful Life, with an RMSE of 53.64 flight cycles. In conclusion, a combination of deep CNN and LSTM models showed a promising method for accurate engine condition monitoring and RUL predictions.

Keywords: Deep Learning, Condition Monitoring, Predictive Maintenance, Artificial Intelligence.

Presenting Author: Theoklis Nikolaidis Cranfield University

Presenting Author Biography: Dr Nikolaidis is a Reader in gas turbine engine performance running the 'Gas Turbine Performance Simulation & Diagnostics' course in Thermal Power MSc programme. He is the Course Director of Thermal Power MSc. Before this, he studied Aeronautical Engineering (BSc) and received his MSc in Thermal Power at Cranfield University in 2003. He followed this with a PhD in 2008, also at Cranfield, conducting a research on the effects of rain ingestion on gas turbine aero-engine performance. Before his appointment, Dr Nikolaidis had been working as an aircraft engineer, gaining experience of aero-engines operation, maintenance, malfunction investigation and troubleshooting.
His research activity focuses on the area of gas turbine performance analysis and modelling-simulation advanced methods. It includes the modelling and simulation of steady state/transient performance and engine's control system. Dr Nikolaidis carries out research on aircraft & engine thermal management, on variable and novel cycles, engine degradation and health monitoring advanced methods, the effects of particulate/multiphase flows on engine's performance and the use of alternative fuels.

Furthermore, he has contributed on the development of Cranfield's software for Gas Turbine Performance Simulation (known as 'Turbomatch') and he is the custodian of the code.

Authors:

Aditya Aditya Cranfield University
Theoklis Nikolaidis Cranfield University
Arias Chao Manuel Zurich University of Applied Sciences
Simone Togni Cranfield University