Physics-informed neural networks (PINNs) have demonstrated promise in solving forward and inverse problems involving partial differential equations. Despite recent progress on expanding the class of problems that can be tackled by PINNs, most of existing use-cases involve simple geometric domains. To date, there is no clear way to inform PINNs about the topology of the domain where the problem is being solved. In this work, we propose a novel positional encoding mechanism for PINNs based on the eigenfunctions of the Laplace-Beltrami operator. This technique allows to create an input space for the neural network that represents the geometry of a given object. We approximate the eigenfunctions as well as the operators involved in the partial differential equations with finite elements. We extensively test and compare the proposed methodology against different types of PINNs in complex shapes, such as a coil, a heat sink and the Stanford bunny, with different physics, such as the Eikonal equation and heat transfer. We also study the sensitivity of our method to the number of eigenfunctions used, as well as the discretization used for the eigenfunctions and the underlying operators. Our results show excellent agreement with the ground truth data in cases where traditional PINNs fail to produce a meaningful solution. We envision this new technique will expand the effectiveness of PINNs to more realistic applications. Code available at: https://github.com/fsahli/Delta-PINNs.

Δ-PINNs: Physics-informed neural networks on complex geometries / Sahli Costabal, Francisco; Pezzuto, Simone; Perdikaris, Paris. - In: ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE. - ISSN 0952-1976. - 127, part B:(2024), pp. 10732401-10732416. [10.1016/j.engappai.2023.107324]

Δ-PINNs: Physics-informed neural networks on complex geometries

Pezzuto, Simone;
2024-01-01

Abstract

Physics-informed neural networks (PINNs) have demonstrated promise in solving forward and inverse problems involving partial differential equations. Despite recent progress on expanding the class of problems that can be tackled by PINNs, most of existing use-cases involve simple geometric domains. To date, there is no clear way to inform PINNs about the topology of the domain where the problem is being solved. In this work, we propose a novel positional encoding mechanism for PINNs based on the eigenfunctions of the Laplace-Beltrami operator. This technique allows to create an input space for the neural network that represents the geometry of a given object. We approximate the eigenfunctions as well as the operators involved in the partial differential equations with finite elements. We extensively test and compare the proposed methodology against different types of PINNs in complex shapes, such as a coil, a heat sink and the Stanford bunny, with different physics, such as the Eikonal equation and heat transfer. We also study the sensitivity of our method to the number of eigenfunctions used, as well as the discretization used for the eigenfunctions and the underlying operators. Our results show excellent agreement with the ground truth data in cases where traditional PINNs fail to produce a meaningful solution. We envision this new technique will expand the effectiveness of PINNs to more realistic applications. Code available at: https://github.com/fsahli/Delta-PINNs.
2024
Sahli Costabal, Francisco; Pezzuto, Simone; Perdikaris, Paris
Δ-PINNs: Physics-informed neural networks on complex geometries / Sahli Costabal, Francisco; Pezzuto, Simone; Perdikaris, Paris. - In: ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE. - ISSN 0952-1976. - 127, part B:(2024), pp. 10732401-10732416. [10.1016/j.engappai.2023.107324]
File in questo prodotto:
File Dimensione Formato  
Sahli Costabal (2024) Delta-PINN.pdf

Solo gestori archivio

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 3.82 MB
Formato Adobe PDF
3.82 MB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/399170
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact