From text to tech: Shaping the future of physics-based simulations with AI-driven generative models

Alessio Alexiadis; Bahman Ghiassi

doi:10.1016/j.rineng.2023.101721

From text to tech: Shaping the future of physics-based simulations with AI-driven generative models

Alessio Alexiadis^*
, Bahman Ghiassi

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

137 Downloads (Pure)

Abstract

This micro-article introduces a method for integrating Large Language Models with geometry/mesh generation software and multiphysics solvers, aimed at streamlining physics-based simulations. Users provide simulation descriptions in natural language, which the language model processes for geometry/mesh generation and physical model definition. Initial results demonstrate the feasibility of this approach, suggesting a future where non-experts can conduct advanced multiphysics simulations by simply describing their needs in natural language, while the code autonomously handles complex tasks like geometry building, meshing, and setting boundary conditions.

Original language	English
Article number	101721
Journal	Results in Engineering
Volume	21
Early online date	30 Dec 2023
DOIs	https://doi.org/10.1016/j.rineng.2023.101721
Publication status	Published - 1 Mar 2024

Keywords

Multiphysics modelling
Physics-informed machine learning
computational modelling
large language models (LLMs)
Generative AI

Access to Document

10.1016/j.rineng.2023.101721Licence: Creative Commons: Attribution-NonCommercial-NoDerivs (CC BY-NC-ND)

AlexiadisA2024FromFinal published version, 4.83 MBLicence: Creative Commons: Attribution-NonCommercial-NoDerivs (CC BY-NC-ND)

https://linkinghub.elsevier.com/retrieve/pii/S2590123023008484

From Text to Tech notebooks
Alexiadis, A. (Creator), University of Birmingham, 8 Dec 2023
DOI: 10.25500/edata.bham.00001032
Dataset

Cite this

@article{d82148d296c24393b495be2bfcacc829,

title = "From text to tech: Shaping the future of physics-based simulations with AI-driven generative models",

abstract = "This micro-article introduces a method for integrating Large Language Models with geometry/mesh generation software and multiphysics solvers, aimed at streamlining physics-based simulations. Users provide simulation descriptions in natural language, which the language model processes for geometry/mesh generation and physical model definition. Initial results demonstrate the feasibility of this approach, suggesting a future where non-experts can conduct advanced multiphysics simulations by simply describing their needs in natural language, while the code autonomously handles complex tasks like geometry building, meshing, and setting boundary conditions.",

keywords = "Multiphysics modelling, Physics-informed machine learning, computational modelling, large language models (LLMs), Generative AI",

author = "Alessio Alexiadis and Bahman Ghiassi",

year = "2024",

month = mar,

day = "1",

doi = "10.1016/j.rineng.2023.101721",

language = "English",

volume = "21",

journal = "Results in Engineering",

issn = "2590-1230",

publisher = "Elsevier",

}

TY - JOUR

T1 - From text to tech: Shaping the future of physics-based simulations with AI-driven generative models

AU - Alexiadis, Alessio

AU - Ghiassi, Bahman

PY - 2024/3/1

Y1 - 2024/3/1

N2 - This micro-article introduces a method for integrating Large Language Models with geometry/mesh generation software and multiphysics solvers, aimed at streamlining physics-based simulations. Users provide simulation descriptions in natural language, which the language model processes for geometry/mesh generation and physical model definition. Initial results demonstrate the feasibility of this approach, suggesting a future where non-experts can conduct advanced multiphysics simulations by simply describing their needs in natural language, while the code autonomously handles complex tasks like geometry building, meshing, and setting boundary conditions.

AB - This micro-article introduces a method for integrating Large Language Models with geometry/mesh generation software and multiphysics solvers, aimed at streamlining physics-based simulations. Users provide simulation descriptions in natural language, which the language model processes for geometry/mesh generation and physical model definition. Initial results demonstrate the feasibility of this approach, suggesting a future where non-experts can conduct advanced multiphysics simulations by simply describing their needs in natural language, while the code autonomously handles complex tasks like geometry building, meshing, and setting boundary conditions.

KW - Multiphysics modelling

KW - Physics-informed machine learning

KW - computational modelling

KW - large language models (LLMs)

KW - Generative AI

U2 - 10.1016/j.rineng.2023.101721

DO - 10.1016/j.rineng.2023.101721

M3 - Article

SN - 2590-1230

VL - 21

JO - Results in Engineering

JF - Results in Engineering

M1 - 101721

ER -

From text to tech: Shaping the future of physics-based simulations with AI-driven generative models

Abstract

Keywords

Access to Document

Fingerprint

Datasets

From Text to Tech notebooks

Cite this