Algorithm refinement for fluctuating hydrodynamics
This paper introduces an adaptive mesh and algorithmrefinement method for fluctuating hydrodynamics. This particle-continuumhybrid simulates the dynamics of a compressible fluid with thermalfluctuations. The particle algorithm is direct simulation Monte Carlo(DSMC), a molecular-level scheme based on the Boltzmann equation. Thecontinuum algorithm is based on the Landau-Lifshitz Navier-Stokes (LLNS)equations, which incorporate thermal fluctuations into macroscopichydrodynamics by using stochastic fluxes. It uses a recently-developedsolver for LLNS, based on third-order Runge-Kutta. We present numericaltests of systems in and out of equilibrium, including time-dependentsystems, and demonstrate dynamic adaptive refinement by the computationof a moving shock wave. Mean system behavior and second moment statisticsof our simulations match theoretical values and benchmarks well. We findthat particular attention should be paid to the spectrum of the flux atthe interface between the particle and continuum methods, specificallyfor the non-hydrodynamic (kinetic) time scales.
- Research Organization:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- USDOE Director. Office of Science. Advanced ScientificComputing Research
- DOE Contract Number:
- DE-AC02-05CH11231
- OSTI ID:
- 929326
- Report Number(s):
- LBNL-63064; R&D Project: K11004; BnR: KJ0101010; TRN: US200813%%190
- Journal Information:
- Multiscale Modeling and Simulation, Vol. 6, Issue 4; Related Information: Journal Publication Date: 2008
- Country of Publication:
- United States
- Language:
- English
Similar Records
Numerical methods for the weakly compressible Generalized Langevin Model in Eulerian reference frame
Smoothed particle hydrodynamics model for Landau-Lifshitz-Navier-Stokes and advection-diffusion equations