ItemArticle

A physics-informed neural network framework for consolidation parameter prediction using controlled clay-sand mixtures

Title
A physics-informed neural network framework for consolidation parameter prediction using controlled clay-sand mixtures
Creator
Urs C, Yogeshraj; Prasanna, H S; Unnam, Anil
Description
This paper introduces a novel Physics-Informed Neural Network (PINN) model for predicting the coefficient of consolidation (Cv) in high plasticity clays. The model was trained from experimental data obtained from controlled clay-sand mixtures. The input parameters include clay content, Atterberg limits, initial void ratio, compaction energy, applied pressure and consolidation characteristics like compression index (Cc) and volumetric compressibility (mv). Additional parameters like plasticity index, porosity, activity-clay interaction and compaction efficiency were derived from feature engineering. The proposed PINN model combines data-driven loss and physics-based loss into a total loss function. The physics loss includes three constraints derived from modified Kozeny-Carman equations, activity-based mineralogical relations, and compression-volume consistency. Hyperparameter optimization identified the optimal configuration: 800 epochs, learning rate 0.001, architecture [128, 64, 32], and physics loss weights distributed as 0.7, 0.25, and 0.05. Five-fold cross-validation demonstrated robust performance (R2 = 0.9903 0.0026), significantly outperforming baseline neural networks (R2 = 0.9682 0.0126, p = 0.0116) with 73.9% reduction in Root Mean Square Error (RMSE = 6.37 10-11 m/s) and 5.71% improvement in Mean Absolute Percentage Error (MAPE = 4.48%). External validation showed the PINN (R = 0.9968) substantially outperformed empirical correlations (best R2 = 0.1636) and conventional machine learning models (best R2 = 0.9878). SHapley Additive exPlanations (SHAP) interpretability analysis validated physically meaningful decision-making, with plastic limit and activity emerging as primary drivers. This framework provides a transferable, physics-consistent solution applicable across diverse clay types for foundation design and site characterization. Copyright 2026. Published by Elsevier B.V.
Source
Results in Engineering;Volume;29;Issue;;Article No.;109255;
Date
01-01-2026
Publisher
Elsevier B.V.
Subject
Activity-clay interaction; Coefficient of consolidation; Explainable AI; Geotechnical engineering; High plasticity clays; Machine learning; Physics-informed neural networks
Coverage
Urs C Y., Department of Civil Engineering, The National Institute of Engineering, Mysuru, Visvesvaraya Technological University, Karnataka, Belgaum, India, Department of Civil Engineering, SOET, CHRIST (Deemed to be University), Karnataka, Bengaluru, India; Prasanna H.S., Department of Civil Engineering, The National Institute of Engineering, Mysuru, Visvesvaraya Technological University, Karnataka, Belgaum, India; Unnam A., Department of Civil Engineering, The National Institute of Engineering, Mysuru, Visvesvaraya Technological University, Karnataka, Belgaum, India
Rights
All Open Access; Gold Open Access
Relation
ISSN: 25901230;
Format
online
Language
English
Type
Article

Citation

Urs C, Yogeshraj; Prasanna, H S; Unnam, Anil, “A physics-informed neural network framework for consolidation parameter prediction using controlled clay-sand mixtures,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 18, 2026, https://archives.christuniversity.in/items/show/22444.

Output Formats