Numerical modelling of brittle fracture using lattice particle method with applications to fluid structure interaction problems via SPH coupling

K. C. Ng, Hailong Chen

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

This paper presents an improved failure model for simulating brittle fracture using the mesh-less Lattice Particle Method (LPM). By modelling the initial crack line using the Remove Bond (RB) approach as outlined in this paper, a new formulation is then developed for predicting the mode-I Stress Intensity Factor (SIF) near the crack tip. Compared to the conventional Remove Particle (RP) approach, it is found that the accuracy of the present SIF formulation based on the RB method is superior. A series of benchmark test cases are simulated to test the numerical accuracy and numerical convergence of the method. Finally, the LPM method is coupled with the Smoothed Particle Hydrodynamics (SPH) method for studying Fluid Structure Interaction (FSI) problems involving solid fracture and free surface. The coupled SPH-LPM method is implemented in the open-source code, DualSPHysics, which has been optimized for both CPU and GPU performances. Upon integrating LPM with SPH, the proposed FSI method is suitable for modelling fracture phenomena caused by natural hazards such as tsunami and flood.

Original languageEnglish
Article number109453
JournalEngineering Fracture Mechanics
Volume289
DOIs
StatePublished - Sep 1 2023

Bibliographical note

Publisher Copyright:
© 2023 The Author(s)

Keywords

  • Brittle fracture
  • DualSPHysics
  • Fluid structure interaction
  • Lattice particle method
  • Smoothed particle hydrodynamics
  • Volume-compensated particle method

ASJC Scopus subject areas

  • General Materials Science
  • Mechanics of Materials
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Numerical modelling of brittle fracture using lattice particle method with applications to fluid structure interaction problems via SPH coupling'. Together they form a unique fingerprint.

Cite this