Jiajia Waters

A parallel Large Eddy Simulation in a finite element projection method for all flow regimes

ABSTRACT A new Finite Element Method (FEM) system has been developed for engine and combustion modeling, KIVA-hpFE. This new FEM uses a projection method for the solution of the momentum equations and runs on parallel computer systems using the Message Passing Interface (MPI). The modeled fluid is multispecies and employs either Reynolds Averaged Navier–Stokes (RANS) (k-ϖ) or a dynamic Large Eddy Simulation (LES) model for the solution of turbulent reactive flow. This FEM-based code provides an excellent platform for developing better in-cylinder fuel and species evolution. A dynamic LES method applicable through transitional to fully turbulent flow is described in this paper. LES uses a fine grid resolution to provide the accurate subgrid-scale (SGS) modeling of turbulence and to capture the majority of the turbulent energy in the resolved regions (the grid scale). To this end, a parallel method is required, particularly for engineering solutions of 3D domains, and is described along with the linear algebraic processes.

Global Versus Localized RBF Meshless Methods for Solving Incompressible Fluid Flow with Heat Transfer

Global and localized radial basis function (RBF) meshless methods are compared for solving viscous incompressible fluid flow with heat transfer using structured multiquadratic RBFs. In the global approach, the collocation is made globally over the whole domain, so the size of the discretization matrices scales as the number of the nodes in the domain. The localized meshless method uses a local collocation defined over a set of overlapping domains of influence. Only small systems of linear equations need to be solved for each node. The computational effort thus grows linearly with the number of nodes—the localized approach is slightly more expensive on serial processors, but is highly parallelizable. Numerical results are presented for three benchmark problems—the lid-driven cavity, natural convection within an enclosure, and forced convective flow over a backward-facing step—and results are compared with the finite-element method (FEM) and experimental data.

Modeling multiphase flow: Spray breakup using volume of fluids in a dynamics LES FEM method

ABSTRACT Currently, all commercial software for engine modeling investigates the dispersed droplet phase of the injection process. Understanding the effect of geometry of the injector nozzle, initial jet conditions, fluid properties in the liquid film, breakup, resulting droplet sizes, and distribution are of primary importance to improve fuel efficiency and lower gas emissions. We have developed an innovative computational method and models to make this atomization process more predictive: a multiscale, multiphase fluid simulation, using a volume-of-fluid method implemented in a large eddy simulation algorithm found in the new KIVA-hpFE, a finite element method flow solver for all flow regimes.

Sensitivity analysis of the grad-div stabilization parameter in finite element simulations of incompressible flow

Abstract We present a numerical study of the sensitivity of the grad-div stabilization parameter for mixed finite element discretizations of incompressible flow problems. For incompressible isothermal and non-isothermal Stokes equations and Navier-Stokes equations, we develop the associated sensitivity equations for changes in the grad-div parameter. Finite element schemes are devised for computing solutions to the sensitivity systems, analyzed for stability and accuracy, and finally tested on several benchmark problems. Our results reveal that solutions are most sensitive for small values of the parameter, near obstacles and corners, when the pressure is large, and when the viscosity is small.

A Local Meshless Method for Approximating 3D Wind Fields

AbstractAn efficient, mesh-free numerical method has been developed for creating 3D wind fields using data from meteorological towers. Node points are placed within a region of interest, generally based upon topological features. Since meshless methods do not require connective mesh generation, storage is greatly reduced, permitting implementation of the code using MATLAB on a personal computer. Utilizing locally collocated nodes and radial basis functions, a 3D wind can be quickly created that satisfies mass consistency. The meshless method yields close approximations to results obtained with mesh-dependent finite-difference, finite-volume, and finite-element techniques.

A Dynamic LES Model for Turbulent Reactive Flow with Parallel Adaptive Finite Elements

An adaptive finite element method (FEM) is used for the solution of turbulent reactive flows in 3-D utilizing parallel methods for fluid dynamic and combustion modeling associated with engines. A dynamic LES method permits transition from laminar to turbulent flow without the assumptions usually required for turbulent sublayers near wall area. This capability is ideal for engine configurations where there is no equilibrium in the turbulent wall layers and the flow is not always turbulent and often in transition. The developed adaptive FEM flow solver uses “h” adaptation to provide for grid refinement. The FEM solver has been optimized for parallel processing employing the message passing interface (MPI) for clusters and high-performance computers.