Thermodynamically consistent modeling of granular soils using physics-informed neural networks

Abstract In recent years, data-driven approaches have gained considerable momentum in the scientific and engineering communities, owing to their capacity to extract complex patterns from high-dimensional data. Despite their potential, these approaches often require extensive high-quality datasets, m...

Full description

Saved in:
Bibliographic Details
Main Authors: Nazanin Irani, Mohammad Salimi, Torsten Wichtmann
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-12844-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333088501366784
author Nazanin Irani
Mohammad Salimi
Torsten Wichtmann
author_facet Nazanin Irani
Mohammad Salimi
Torsten Wichtmann
author_sort Nazanin Irani
collection DOAJ
description Abstract In recent years, data-driven approaches have gained considerable momentum in the scientific and engineering communities, owing to their capacity to extract complex patterns from high-dimensional data. Despite their potential, these approaches often require extensive high-quality datasets, may exhibit limited extrapolation capability beyond the training domain, and lack a rigorous foundation grounded in physical and thermodynamic principles. To overcome these limitations, physics-informed neural networks have been introduced, embedding governing equations directly into the learning process. Building upon this paradigm, this study presents a novel thermodynamically consistent constitutive model for granular soils, developed within the framework of geotechnically- and physics-informed neural networks (GINN). The model integrates physical laws with data-driven learning via a composite loss function. These include: (i) strictly non-negative material dissipation rate to ensure thermodynamic admissibility, (ii) an admissible range for the predicted stress state, and (iii) bounds on the predicted void ratio. The material dissipation rate is calculated using the total work input and a free energy potential expressed in terms of stress invariants. The model is validated against monotonic drained triaxial test data for specimens prepared with diverse initial void ratios and stress states. The model accurately simulates both the shear strength and dilative response of granular soil samples. Its predictive performance is further benchmarked against two widely adopted constitutive models from the literature, demonstrating comparable accuracy while maintaining consistency with thermodynamic laws.
format Article
id doaj-art-d566438fd38c4009a01eb8ce94e02d7b
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-d566438fd38c4009a01eb8ce94e02d7b2025-08-20T03:45:59ZengNature PortfolioScientific Reports2045-23222025-07-0115111210.1038/s41598-025-12844-4Thermodynamically consistent modeling of granular soils using physics-informed neural networksNazanin Irani0Mohammad Salimi1Torsten Wichtmann2Chair of Soil Mechanics, Foundation Engineering, and Environmental Geotechnics, Ruhr-University BochumChair of Soil Mechanics, Foundation Engineering, and Environmental Geotechnics, Ruhr-University BochumChair of Soil Mechanics, Foundation Engineering, and Environmental Geotechnics, Ruhr-University BochumAbstract In recent years, data-driven approaches have gained considerable momentum in the scientific and engineering communities, owing to their capacity to extract complex patterns from high-dimensional data. Despite their potential, these approaches often require extensive high-quality datasets, may exhibit limited extrapolation capability beyond the training domain, and lack a rigorous foundation grounded in physical and thermodynamic principles. To overcome these limitations, physics-informed neural networks have been introduced, embedding governing equations directly into the learning process. Building upon this paradigm, this study presents a novel thermodynamically consistent constitutive model for granular soils, developed within the framework of geotechnically- and physics-informed neural networks (GINN). The model integrates physical laws with data-driven learning via a composite loss function. These include: (i) strictly non-negative material dissipation rate to ensure thermodynamic admissibility, (ii) an admissible range for the predicted stress state, and (iii) bounds on the predicted void ratio. The material dissipation rate is calculated using the total work input and a free energy potential expressed in terms of stress invariants. The model is validated against monotonic drained triaxial test data for specimens prepared with diverse initial void ratios and stress states. The model accurately simulates both the shear strength and dilative response of granular soil samples. Its predictive performance is further benchmarked against two widely adopted constitutive models from the literature, demonstrating comparable accuracy while maintaining consistency with thermodynamic laws.https://doi.org/10.1038/s41598-025-12844-4Physics-informed neural networksGINNThermodynamics lawsConstitutive modellingEnergy conservation
spellingShingle Nazanin Irani
Mohammad Salimi
Torsten Wichtmann
Thermodynamically consistent modeling of granular soils using physics-informed neural networks
Scientific Reports
Physics-informed neural networks
GINN
Thermodynamics laws
Constitutive modelling
Energy conservation
title Thermodynamically consistent modeling of granular soils using physics-informed neural networks
title_full Thermodynamically consistent modeling of granular soils using physics-informed neural networks
title_fullStr Thermodynamically consistent modeling of granular soils using physics-informed neural networks
title_full_unstemmed Thermodynamically consistent modeling of granular soils using physics-informed neural networks
title_short Thermodynamically consistent modeling of granular soils using physics-informed neural networks
title_sort thermodynamically consistent modeling of granular soils using physics informed neural networks
topic Physics-informed neural networks
GINN
Thermodynamics laws
Constitutive modelling
Energy conservation
url https://doi.org/10.1038/s41598-025-12844-4
work_keys_str_mv AT nazaninirani thermodynamicallyconsistentmodelingofgranularsoilsusingphysicsinformedneuralnetworks
AT mohammadsalimi thermodynamicallyconsistentmodelingofgranularsoilsusingphysicsinformedneuralnetworks
AT torstenwichtmann thermodynamicallyconsistentmodelingofgranularsoilsusingphysicsinformedneuralnetworks