Abstract:The wetting deformation of coarse-grained materials can seriously affect the safety of earth and rock dams during initial water storage. The wetting model formulas are expressed in various forms and have complex parameters. Only a small amount of test data is fitted by mathematical statistics, and the universality of the obtained wetting model is unknown. Duncan-Chang E-B constitutive model cannot accurately reflect the wetting deformation characteristics of coarse-grained materials. Through the double-line wetting test of coarse-grained materials, the wetting model proposed by predecessors was verified and analyzed. Based on the indoor wetting test data, the parameters of each wetting model were fitted to analyze the accuracy of each wetting model in describing the wetting deformation characteristics. According to the P-Z model in the elastic-plastic theory and the wetting model formula, the P-Z wetting model is established, and the BP artificial neural network is introduced to establish the artificial neural network wetting deformation prediction model based on the P-Z model. The results show that the relationship between wetting axial strain and wetting stress level is best expressed by the exponential function. The relationship between wetting volumetric strain and wetting stress level is best described by Cheng's linear function. The relative errors between the predicted and experimental values of the proposed neural network prediction model are all within 6%. The relationship between wetting axial strain and wetting stress level is exponential function, and the relationship between wetting volumetric strain and wetting stress level is linear function. The P-Z wetting model proposed in this research can better reflect the wetting deformation characteristics of coarse-grained materials under complex stress paths. The artificial neural network prediction model based on P-Z wetting model is more reliable and accurate, which can meet the prediction requirements of actual engineering for wetting deformation of coarse-grained materials.

A PINN-based modelling approach for hydromechanical behaviour of unsaturated expansive soils

A Multifidelity Neural Network (MFNN) for Constitutive Modeling of Complex Soil Behaviors

Hierarchical Multiscale Modeling of Fluid-Saturated Soils

Constitutive relations and deformation calculation for unsaturated soils

Modeling water flow in unsaturated soils through physics-informed neural network with principled loss function

An effective stress-based approach to modeling the hydro-mechanical behavior of unsaturated soils

A Deep Learning Approach for Predicting Two-Dimensional Soil Consolidation Using Physics-Informed Neural Networks (PINN)

A constitutive model considering the interaction between evolution of microstructure and hydro-mechanical behaviour of unsaturated soils

Simulating Field Soil Temperature Variations with Physics-Informed Neural Networks

Modelling the mechanical behaviour of soils using machine learning algorithms with explicit formulations

Reconstructing unsaturated infiltration behavior with sparse data via physics-informed deep learning

A Unified Thermo-Mechanical Bounding Surface Model for Saturated Clay and Sand

A micro-mechanical model for unsaturated soils based on DEM

Study on wetting deformation model of coarse-grained materials based on P-Z model and BP neural network

Physics‐informed neural networks with monotonicity constraints for Richardson‐Richards equation: Estimation of constitutive relationships and soil water flux density from volumetric water content measurements

A constitutive model for structured soils under saturated and unsaturated conditions

An elastic-viscoplastic model for time-dependent behavior of unsaturated soils

Application of improved physics-informed neural networks for nonlinear consolidation problems with continuous drainage boundary conditions

Data-driven forward and inverse analysis of two-dimensional soil consolidation using physics-informed neural network

New Unsaturated Dynamic Porosity Hydromechanical Coupled Model and Experimental Validation

Machine learning-based soil–structure interaction analysis of laterally loaded piles through physics-informed neural networks