A Stabilised Semi-Implicit Double-Point Material Point Method for Soil-Water Coupled Problems

TL;DR

This paper introduces a semi-implicit double-point Material Point Method for soil-water problems, improving stability and reliability in large deformation geotechnical simulations by modeling soil and water separately with a new stabilisation technique.

Contribution

It develops a novel semi-implicit MPM with two material point sets for soil and water, enhancing stability and accuracy over existing single-point models.

Findings

The method stabilizes stress oscillations in soil-water simulations.

Two-point approach yields more reliable landslide analysis results.

Computational cost remains acceptable for practical applications.

Abstract

A semi-implicit two-phase double-point Material Point Method (MPM) formulation, based on the incremental fractional-step method to model large deformation geotechnical problems has been derived. The semi-implicit formulation has two advantages compared with the explicit approach: the time step is independent of the water phase, and the pore pressure field is more stable. The semi-implicit MPM models based on the incremental fractional-step method available in the literature consist of modelling the soil and water mixture using a single set of material points only, in order to save computational time. In this study, we further derive this formulation with two sets of material points to represent the soil and water phases separately. The stress oscillations that are frequently found in the water and soil phases are stabilised with this approach. A new stabilisation method is developed based on the modified F-bar method. The proposed method is validated with two numerical examples under small and large deformations, respectively. After that, Nor-Sand constitutive soil model is used to simulate landslides. Numerical examples show an excellent performance of the proposed coupled MPM and the stabilisation method. The formulation with two sets of material points yields significantly different but more reliable results in the landslides analysis, compared with the single-point approach. Additionally, this research shows that the additional computational cost caused by the additional water material points is acceptable. Therefore, it is recommended to use two sets of material points for certain large deformation geotechnical problems.