MS47 Machine Learning Applications in Structural State Diagnosis and Prognosis based on Ultrasound and Vibration

Prof. Dimitrios Chronopoulos: dimitrios.chronopoulos@nottingham.ac.uk

Session Chairs:
Dimitrios Chronopoulos, Prof., University of Nottingham, UK, dimitrios.chronopoulos@nottingham.ac.uk
Mohamed Ichchou, Prof., Ecole Centrale de Lyon, France, mohamed.ichchou@ec-lyon.fr
WangJi Yan, Prof., University of Macao, China, wangjiyan@um.edu.mo

Abstract of the special session:
The session aims at bringing together a multidisciplinary team of academic and industrial researchers working on statistical learning techniques for health-state diagnostic and prognostic methodologies. Global industrial and academic stakeholders have long recognized the need for the development of robust platforms able to i) detect damage, preferably during operation of a structural product, ii) identify the type and size of the detected damage and quantify its criticality through estimating the remaining operational time of the structure. Focal point of these objectives is the accurate representation of damaged segments via data or fused data/physics driven approaches.
This session will form a platform for ideas exchange and knowledge dissemination concerning the latest developments in the field of wave based and vibration based defect identification with emphasis on predictive maintenance technologies for composite and complex structures. Topics relevant to the Session include, but are not limited to, implementations and algorithmic solutions for:

• Wave interaction modelling
• Damage detection and identification methods for real-life structures
• Machine learning methods for structural health monitoring
• Bayesian and AI diagnostic and prognostic approaches
• Vibration-based diagnostic methodologies

Contributions pertaining to the implementation of such methods on real-life applications are especially welcomed.