Point-Cloud-based Deep Learning Models for Finite Element Analysis

Meduri Venkata Shivaditya (1) , Francesca Bugiotti (1, 2, 3) , Frederic Magoules (4)

1 CentraleSupélec
2 LISN - Laboratoire Interdisciplinaire des Sciences du Numérique
3 LaHDAK - Données et Connaissances Massives et Hétérogènes - LISN
4 MICS - Mathématiques et Informatique pour la Complexité et les Systèmes

Current simulation of metal forging processes use advanced finite element methods. Such methods consist of solving mathematical equations, which takes a significant amount of time for the simulation to complete. Computational time can be prohibitive for parametric response surface exploration tasks. In this paper, we propose as an alternative, a Graph Neural Networkbased graph prediction model to act as a surrogate model for parameters search space exploration and which exhibits a time cost reduced by an order of magnitude. Numerical experiments show that this new model outperforms the Point-Net model and the Dynamic Graph Convolutional Neural Net model.

Mots clés: Deep Learning Graph Neural Network Point-Net Dynamic Graph Convolutional Neural Net Finite Element Analysis GPU; Deep Learning; Graph Neural Network; Point-Net; Dynamic Graph Convolutional Neural Net; Finite Element Analysis; GPU

Domaines: Informatique [cs]

Meduri Venkata Shivaditya, Francesca Bugiotti, Frederic Magoules. Point-Cloud-based Deep Learning Models for Finite Element Analysis. 2022 21st International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES), Oct 2022, Chizhou, China. pp.50-53, ⟨10.1109/DCABES57229.2022.00049⟩. ⟨hal-04142075⟩