Torna ai risultati della ricercaBando per assegno di ricerca
| Titolo del progetto di ricerca in italiano | Physics-informed data-driven dynamic modeling of satellites based on experimental data and generative adversial neural networks |
|---|---|
| Titolo del progetto di ricerca in inglese | Physics-informed data-driven dynamic modeling of satellites based on experimental data and generative adversial neural networks |
| Campo principale della ricerca | Engineering |
| Sottocampo della ricerca | Civil engineering |
| G.S.D. | 08/CEAR-06 - SCIENZA DELLE COSTRUZIONI |
| S.S.D | CEAR-06/A - Scienza delle costruzioni |
| Descrizione sintetica in italiano | Dynamic modeling of satellites is crucial for ensuring efficient operation, and enhancing mission success rates. Traditional methods of modeling making use of physical laws and first-principle equations can be computationally intensive. With the advent of machine learning and data-driven approaches, models can learn from experimental data while still incorporating fundamental physical principles. One promising approach is the integration of Generative Adversarial Networks (GANs) with physics-informed neural networks (PINNs). PINNs leverage neural networks to solve PDEs that govern physical systems. By embedding physical laws into the loss function of the neural network, PINNs can ensure that the learned models adhere to known physical constraints. This project aims to integrate GANs with PINNs to deliver powerful models of satellites and metamaterials. |
| Descrizione sintetica in inglese | Dynamic modeling of satellites is crucial for ensuring efficient operation, and enhancing mission success rates. Traditional methods of modeling making use of physical laws and first-principle equations can be computationally intensive. With the advent of machine learning and data-driven approaches, models can learn from experimental data while still incorporating fundamental physical principles. One promising approach is the integration of Generative Adversarial Networks (GANs) with physics-informed neural networks (PINNs). PINNs leverage neural networks to solve PDEs that govern physical systems. By embedding physical laws into the loss function of the neural network, PINNs can ensure that the learned models adhere to known physical constraints. This project aims to integrate GANs with PINNs to deliver powerful models of satellites and metamaterials. |
| Data del bando | 31/07/2024 |
| Numero di assegnazioni per anno | 1 |
| Paesi in cui può essere condotta la ricerca |
Italy |
| Paesi di residenza dei candidati |
EUROPE |
| Nazionalità dei candidati |
Italy |
| Sito web del bando | https://web.uniroma1.it/trasparenza/bando/221653_ar062024 |
| Destinatari dell'assegno di ricerca (of target group) |
Early stage researcher or 0-4 yrs (Post graduate) |
|---|
| Nome dell'Ente finanziatore | Dipartimento di Ingegneria Strutturale e Geotecnica |
|---|---|
| Tipologia dell'Ente | Public research |
| Paese dell'Ente | Italy |
| Città | Roma |
| Codice postale | 00187 |
| Indirizzo | Via Eudossiana 18 |
| Sito web | https://web.uniroma1.it/disg/ |
| stefania.pontecorvo@uniroma1.it |
| L'assegno finanziato/cofinanziato attraverso un EU Research Framework Programme? | No |
|---|
| Data di scadenza del bando | 30/08/2024 |
|---|---|
| Come candidarsi | Other |