Back to Journal

SM Journal of Biomedical Engineering

Multi-field Semi-Lagrangian Reproducing Kernel Particle Method for Landslide Simulation

[ ISSN : 2573-3702 ]

Abstract
Details

Received: 06-Feb-2017

Accepted: 20-Feb-2017

Published: 24-Feb-2017

Thanakorn Siriaksorn¹ and Sheng-Wei Chi²*

¹Ph.D. Candidate, Department of Civil and Materials Engineering, University of Illinois at Chicago, Chicago, IL 60607
²Assistant Professor, Department of Civil and Materials Engineering, University of Illinois at Chicago, Chicago, IL 60607

Corresponding Author:

Sheng-Wei Chi, Assistant Professor, Department of Civil and Materials Engineering, University of Illinois at Chicago, Chicago, IL 60607, Email: swchi@uic.edu

Abstract

This paper presents a multi-field meshfree method based on a semi-Lagrangian reproducing kernel particle method for simulation of landslide processes. In the proposed method, the approximation functions of field variables are constructed based on particles, without the need of grids or structured mesh, and the discrete state variables follow the particles, thus Lagrangian description; while the interpolation function of each particle is continually updated during deformation. The formulation naturally handles extremely large deformation and material separation, thus the method is well-suited to capture the deformation of geomaterials transitioning from elastic to an almost fluid-like deformation mode during a landslide event. The general Biot theory is incorporated in the multi-field semi-Lagrangian formulation to describe the behaviors of porous geomaterials. Several numerical examples are presented for verification of the proposed method and a landslide simulation is validated against experimental results.

Citation

Siriaksorn T and Chi S. Multi-field Semi-Lagrangian Reproducing Kernel Particle Method for Landslide Simulation. SM J Biomed Eng. 2017; 3(2): 1014s.