Jet Engine Health Monitoring using Hyperspectral Imaging and Artificial Intelligence
Overview
- Project Code: JEMHy
- Start Date: July 1, 2023
- End Date: December 31, 2027
- Reference: DAP 23/07
- Funding: Defense Funded Research (DFR)
- Royal Military Academy Involvement: Director
- Quad Chart
Team
Staff
Partners
- Coordinator: Rob Haelterman
- Coordinator Affiliation: Royal Military Academy, Department of Mathematics (MWMW)
- Project Co-promotor: Romuald Van Riet (CHCH)
- RMA Researchers: Skralan Hosteaux
- Project Partners (RMA): Department of Chemistry (CHCH)
- Project Partners (BE-DEF): N/A
- Project Partners (Other): N/A
Context
The current method of jet engine health monitoring at Belgian air bases is considered to be time- and cost innefficient as it requires to open up the engine to extract oil samples and analyze them in the lab. A method to facilitate this process by assessing jet engine health and the need to conduct preventive maintenance in a non-intrusive manner would be of great interest.
Objectives
“- Asses the spectral and spatial features in jet engine exhaust plumes indicative of engine degradation and engine component failure
- Develop the sensor suite capable of meeting the requirements of detecting early degradation in jet engines
- Develop ML algorithms capable of detecting early degradation in jet engines based on hyperspectral observations of echaust plumes”
Methodology
Through real observations and intricate modelling of jet engine exhaust plumes, develop a large database of HSI measurements that allows training ML models to be able to detect the necessity of preventive mainenance of the engines.