CESS seminar series: Damage Identification in Structural Components using Vibrations aided with Physics-informed Neural Networks
Event details
| Date | 21.10.2022 |
| Hour | 12:15 › 13:00 |
| Speaker | Dr Saeid Hedayatrasa, Ghent University (UGent-MMS), Belgium |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
Abstract
High-performance composite materials have been increasingly used for manufacturing structural components in different industries e.g. aerospace, automotive and wind energy. Given their layered multi-material design, efficient inspection techniques are indispensable for reliable identification of production defects and in-service damage. In this presentation, performance of different techniques in inspection of aircraft composite panels will be compared. Furthermore, detection and quantification of damage through full-field scanning laser Doppler vibrometry and sparse array Guided Wave tomography will be presented. The limitations of the techniques and the potential of Physics-informed Neural Networks for more efficient identification of damage will be discussed and the approach will be examined based on simulation data with added noise.
Biography
Saeid Hedayatrasa is a senior postdoctoral research fellow at the Mechanics of Materials and Structures research group of Ghent University (UGent-MMS) in Belgium. After getting a Master’s degree in Mechanical Engineering and a few years of industrial and post-master research work in Iran, he completed his PhD in Mechanical and Manufacturing Engineering in October 2016 at the University of South Australia. He was employed by the University of South Australia until July 2017 as a research and teaching assistant, after which he joined Ghent University as a postdoctoral researcher. He has computational and experimental background in design optimization and characterization of acoustic meta-materials and composites. More specifically, his research focus has been on analysis of vibrations, elastic wave propagation and heat wave diffusion in advanced materials, and their synergic (vibro-thermal) interaction for non-destructive testing (NDT) and structural health monitoring (SHM). He is currently a visiting researcher at IMOS lab (EPFL) for a research collaboration on developing physics-informed deep learning algorithms for damage quantification through vibrational measurement data.
High-performance composite materials have been increasingly used for manufacturing structural components in different industries e.g. aerospace, automotive and wind energy. Given their layered multi-material design, efficient inspection techniques are indispensable for reliable identification of production defects and in-service damage. In this presentation, performance of different techniques in inspection of aircraft composite panels will be compared. Furthermore, detection and quantification of damage through full-field scanning laser Doppler vibrometry and sparse array Guided Wave tomography will be presented. The limitations of the techniques and the potential of Physics-informed Neural Networks for more efficient identification of damage will be discussed and the approach will be examined based on simulation data with added noise.
Biography
Saeid Hedayatrasa is a senior postdoctoral research fellow at the Mechanics of Materials and Structures research group of Ghent University (UGent-MMS) in Belgium. After getting a Master’s degree in Mechanical Engineering and a few years of industrial and post-master research work in Iran, he completed his PhD in Mechanical and Manufacturing Engineering in October 2016 at the University of South Australia. He was employed by the University of South Australia until July 2017 as a research and teaching assistant, after which he joined Ghent University as a postdoctoral researcher. He has computational and experimental background in design optimization and characterization of acoustic meta-materials and composites. More specifically, his research focus has been on analysis of vibrations, elastic wave propagation and heat wave diffusion in advanced materials, and their synergic (vibro-thermal) interaction for non-destructive testing (NDT) and structural health monitoring (SHM). He is currently a visiting researcher at IMOS lab (EPFL) for a research collaboration on developing physics-informed deep learning algorithms for damage quantification through vibrational measurement data.
Practical information
- Informed public
- Free
Organizer
- Prof. Olga Fink (IMOS), Prof. Alexandre Alahi (VITA), Prof . Dusan Licina (HOBEL) and Prof. Alain Nussbaumer (RESSLab)
Contact
- Prof. Olga Fink, IMOS (EPFL)