Co-simulation of multibody dynamics and plenteous sphere of contacted particles using NVIDIA GPGPU

Ji Soo Park, Joon Shik Yoon, Jin Hwan Choi, Sung Soo Rhim

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

In this study, a dynamic simulation model that considers many spherical particles and multibody dynamics (MBD) entities is developed. Plenteous spherical particles are solved using the Discrete Element Method (DEM) technique and simulated on a GPU board in a PC. A fast algorithm is used to calculate the Hertzian contact forces between many spherical particles, and NVIDIA CUDA is used to increase the calculation speed. The explicit integration method is applied to solve the many spheres. MBD entities are simulated by recursive formulation. Constraints are reduced by recursive formulation, and the implicit generalized alpha method is applied to solve the dynamic model. A new algorithm is developed to simulate the DEM and MBD models simultaneously. As a numerical example, a truck car model and gear model are developed. The results show that the proposed algorithm using a general-purpose GPU in a PC has many advantages.

Original languageEnglish
Pages (from-to)465-474
Number of pages10
JournalTransactions of the Korean Society of Mechanical Engineers, A
Volume36
Issue number4
DOIs
Publication statusPublished - Apr 2012

Keywords

  • Co-simulation
  • DEM
  • GPGPU
  • MBD
  • Particles

Fingerprint

Dive into the research topics of 'Co-simulation of multibody dynamics and plenteous sphere of contacted particles using NVIDIA GPGPU'. Together they form a unique fingerprint.

Cite this