Bridging experiments and models, towards a new paradigm : DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems
Résumé
A fast and robust PINN-based algorithm is introduced to model turbulent breakage
• DROP-PINN infers breakage frequencies directly from image sequences of droplets.
• The methodology, based on various simulated configurations, is caseindependent
• The DROP-PINN performance is demonstrated by dedicated stirred tank experiments
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |