Samila Bandara obtained her Bachelor's degree in Civil Engineering at University of Moratuwa in 2007 with a First Class Honours and was awarded the medal for the Best Civil Engineering student. After graduation, she had been working as an assistant lecturer in the Department of Civil Engineering at the University of Moratuwa. She joined the Geotechnical & Environmental Engineering Research Group at the University of Cambridge in October 2008 as a PhD student under the supervision of Professor Kenichi Soga. Samila was offered the 'Jafar Studentship in Petroleum Engineering at Churchill College', to carry out her research work. She is a member of Churchill College.

Samila’s research topic is ‘Meshfree methods to simulate fluidisation problems’.

Introduction to the research

Fluidisation problems in fluid-saturated granular soils are of great interest within the field of geotechnology. Such problems generally involve large deformations and different phases, and hence conventional mesh-based numerical methods such as finite element method (FEM) and finite difference method (FDM) are very difficult to apply.
Large deformation problems can be solved using so-called meshfree methods that are particle based and trace the history of the state variables at particle locations. Material point method (MPM) and smoothed particle hydrodynamics (SPH) method are two outstanding methods in recent years due to their capability to study complex problems.
This research is directed to develop a numerical code (using C++ programming language) based on meshfree methods (SPH, MPM) that is capable of modelling fluidisation problems. The developed code will be used to analyse granular flow problems and coupled consolidation problems.

Coupled consolidation analysis using MPM

A new coupling algorithm using explicit time integration is proposed to study problems in saturated soil which consists of two layers of particles (Soil & water). Both 2D plane strain numerical code and an axisymmetric numerical code has been developed in this research to study the problems in saturated soil domain.

Implicit time integration analysis for MPM

Computational time can be reduced in implicit analysis due to the applicability of larger time steps and the use of unconditionally stable time integration method. GPGPU (General-purpose computing on graphics processing units) will be used to reduced the computational time.