Artificial Intelligence-Driven Aircraft Maintenance with Real-Time Crack Detection and Predictive Analytics
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Utilizing cutting-edge deep learning methods in conjunction with machine learning strategies for aircraft systems, AI-based designs from the proposed research allow for optimized predictive maintenance. Three essential parts are included in the systemml: The system offers lifetime forecasts and real-time crack assessment for metal subassemblies like jet engines and aircraft batteries. YOLOv5 uses identification and annotation techniques to find structural flaws in photos of airplane fuselages, enabling precise yet quick inspections. Using battery life estimates for discharge cycles, voltage, and charging time, Random Forest Regressors forecast an asset’s Remaining Useful Life (RUL). In order to expedite maintenance procedures, sensor data from jet engines is fed into a neural network constructed with PyTorch, which determines their Remaining Useful Life. Test results show that YOLOv5 can identify flaws with 92% accuracy, and the battery .
Fields of Study & Subjects
Machine Fault Diagnosis Techniques
Infrastructure Maintenance and Monitoring
Industrial Vision Systems and Defect Detection
Publication details
- Journal / source
- CRC Press eBooks
- Pages
- 182–192
- Publication date
- 31 March 2026
- Language
- EN
- Document type
- Book chapter