Physics-informed machine learning for numerical modelling in engineering and science

Yuantong Gu; Charith Rathnayaka; Jinshuai Bai

Back

Conference presentation

Physics-informed machine learning for numerical modelling in engineering and science

Yuantong Gu, Charith Rathnayaka and Jinshuai Bai

World Congress on Computational Mechanics, 16th (Vancouver, Canada, 21-Jul-2024–26-Jul-2024)

2024

Files and links (2)

url

https://storage.googleapis.com/usacm_static_shared/wccm2024/MS_0834.pdfView

Published Version

url

https://www.wccm2024.org/View

Event Website

Abstract

Numerical modelling and mechanical characterisation

Machine learning

Artificial intelligence

Expanding knowledge in the information and computing sciences

Expanding knowledge in engineering

physics informed machine learning

Computational Mechanics

Nonlinear and complex problems

MINISYMPOSIUM: Machine learning has gained increasing attention in the field of numerical modelling. Fuelled by data, it provides researchers with powerful computing tools and has already led to significant innovations. However, in many real-world engineering and science applications, data scarcity can pose significant challenges for machine-learning-driven numerical modelling, hindering its practical implementation. Recent advancements in 'physics-informed machine learning' have enabled incorporation of guidance from 'physics', such as governing equations and boundary conditions, into machine learning inspiring a transition away from sole reliance on data. Physics-informed machine learning methods have demonstrated the ability to use 'physics' as a remedy to insufficient data, resulting in superior performances in terms of accuracy and robustness, specifically for applications with increased complexity and non-linearity. In such a context, this mini-symposium aims to foster a rich and comprehensive dialogue at WCCM 2024 about latest advancements in physics-informed machine learning for numerical methods in engineering and science.

Details

Title: Physics-informed machine learning for numerical modelling in engineering and science
Authors: Yuantong Gu (Corresponding Author) - Queensland University of Technology
Charith Rathnayaka (Author) - University of the Sunshine Coast, Queensland, School of Science, Technology and Engineering
Jinshuai Bai (Author) - Queensland University of Technology
Conference details: World Congress on Computational Mechanics, 16th (Vancouver, Canada, 21-Jul-2024–26-Jul-2024)
Date published: 2024
Organisation Unit: School of Science, Technology and Engineering
Language: English
Record Identifier: 991064598902621
Output Type: Conference presentation

Metrics

38 Record Views