CESS seminar series: From conventional aerospace structures to cyber-physical structural assets: Introducing the digital siblings concept
Event details
| Date | 28.10.2022 |
| Hour | 12:15 › 13:00 |
| Speaker | Prof. Dimitrios Zarouchas, TUDELFT |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
Abstract:
The progress in key-technologies such as Artificial Intelligence and Internet of Things opens new horizons for developing the lightweight structures of the future, where the structures will be able to perform self-diagnostic checks of their integrity, self-estimate their lifetime and communicate with each other valuable information. This lecture delivers the vision of the Pythia team on how to design cyber-physical systems from sensing to health indicators’ extraction and from diagnosis and prognosis to decision making protocols. Technical details on the concept of digital siblings, essential tool for physics-informed diagnosis, will be presented; the digital siblings are designed using stochastic finite elements, Artificial Neural Network and SHM data. A demonstration of the feasibility and efficiency of digital siblings will be given for the real-time strength estimation of composite structures while being tested under static loading.
Biography:
Dimitrios Zarouchas is an Associate professor and the Director of the Center of Excellence in Artificial Intelligence at the Aerospace Engineering Faculty of Delft University of Technology. He leads the Pythia team – Artificial Intelligence for Structures, Health Management and Prognostics, with his vision being the research, development and deployment of intelligent cyber-physical systems for enabling real-time data extraction for the purpose of health management of systems and structures used in aerospace, automotive, wind energy and maritime sectors. The team is developing machine learning algorithms for data-driven and physics-based diagnostics and prognostics and employs reinforcement learning for decision making protocols. Emphasis is given to self-learning (development of learning agents), learn-how-to learn prototypes and transfer learning. Dimitrios has been leading multi-millionaire research and innovation projects, funded by EU and National schemes and he has been the author of more than 70 journal publications.
The progress in key-technologies such as Artificial Intelligence and Internet of Things opens new horizons for developing the lightweight structures of the future, where the structures will be able to perform self-diagnostic checks of their integrity, self-estimate their lifetime and communicate with each other valuable information. This lecture delivers the vision of the Pythia team on how to design cyber-physical systems from sensing to health indicators’ extraction and from diagnosis and prognosis to decision making protocols. Technical details on the concept of digital siblings, essential tool for physics-informed diagnosis, will be presented; the digital siblings are designed using stochastic finite elements, Artificial Neural Network and SHM data. A demonstration of the feasibility and efficiency of digital siblings will be given for the real-time strength estimation of composite structures while being tested under static loading.
Biography:
Dimitrios Zarouchas is an Associate professor and the Director of the Center of Excellence in Artificial Intelligence at the Aerospace Engineering Faculty of Delft University of Technology. He leads the Pythia team – Artificial Intelligence for Structures, Health Management and Prognostics, with his vision being the research, development and deployment of intelligent cyber-physical systems for enabling real-time data extraction for the purpose of health management of systems and structures used in aerospace, automotive, wind energy and maritime sectors. The team is developing machine learning algorithms for data-driven and physics-based diagnostics and prognostics and employs reinforcement learning for decision making protocols. Emphasis is given to self-learning (development of learning agents), learn-how-to learn prototypes and transfer learning. Dimitrios has been leading multi-millionaire research and innovation projects, funded by EU and National schemes and he has been the author of more than 70 journal publications.
Practical information
- General public
- Free
Organizer
- CE