Extendable outflow boundary conditions for dissipative particle dynamics simulation

Z. G. Huang, Tai Man Yue, Y. Deng, Z. N. Guo

Research output: Journal article publicationJournal articleAcademic researchpeer-review


This paper presents a systematic study on a basic kind of outflow boundary condition for Dissipative Particle Dynamics (DPD) simulation, in which the flow near the outlet can be assumed as fully developed. The boundary conditions are expressed in Neumann-type constraints due to the zero gradients of hydrodynamic quantities on the boundary. The flow rate is included as an additional constraint to ensure the stability of a simulation. The boundary conditions are imposed through a reservoir, which is a replica of the simulation domain in the vicinity of the outlet. All the particle configurations are conserved in the reservoir. A velocity adjusting scheme is developed for the reservoir in order to control the deviation of the outflow rate and to remove the pressure discontinuity across the boundary. The effectiveness of the boundary method is examined through the simulations of a uniform flow and a laminar flow in a bifurcated tunnel. The outflow rates, as well as the velocity profile and pressure distribution, all agree well with the Navier-Stokes solution by comparison. The assumption of the fully developed outflow is validated by measuring the gradients of the flow rates. It is expected that this boundary method can be used in the simulation of flows with heterogeneous particle distribution, without having to evaluate the distribution function in advance.
Original languageEnglish
Article number1750071
JournalInternational Journal of Modern Physics C
Issue number6
Publication statusPublished - 1 Jun 2017


  • Boundary conditions
  • dissipative particle dynamics
  • outflow boundary
  • steady flow

ASJC Scopus subject areas

  • Statistical and Nonlinear Physics
  • Mathematical Physics
  • Physics and Astronomy(all)
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this