GPU accelerated high-order discontinuous galerkin level set methods for incompressible multiphase flows

Download

index.pdf

Date

2015

Author

Karakuş, Ali

Metadata

Show full item record

Item Usage Stats

145
views

214
downloads

This thesis study focuses on the development of GPU accelerated, high-order discontinuous Galerkin methods for the solution of unsteady incompressible and immiscible multiphase flows with level set interface representation. Nodal discontinuous Galerkin framework is used for Navier-Stokes, level set evolution and reinitialization equations on unstructured elements. Computations are localized mostly near the interface location with an adaptive method to reduce computational cost without sacrificing the accuracy. An artificial diffusion approach with a modal decay rate based regularity estimator is used to damp out high frequency solution components near kinks. For the computation of interface equations, a multi-rate Adams-Bashforth time integrator is designed to avoid time step restrictions resulting from artificial diffusion stabilization and local mesh refinement. Implicit systems arising from the semi-explicit time discretization of the flow equations are solved with a matrix-free $ p $-multigrid preconditioned conjugate gradient method to minimize memory requirements and to increase overall runtime performance. The developed numerical scheme is accelerated using modern GPUs and many-core CPUs. Platform independence of the solver is achieved with an extensible multi-threading programming API as common kernel language that allows runtime selection of different computing devices and threading interfaces. Efficiency, scalability, local high-order accuracy and mass conservation of the method are confirmed through distinct numerical test cases.

Subject Keywords

Multiphase flow., Galerkin methods., Navier-Stokes equations.

URI

http://etd.lib.metu.edu.tr/upload/12619523/index.pdf
https://hdl.handle.net/11511/25264

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

BEM solution of MHD flow in a pipe coupled with magnetic induction of exterior region Tezer, Münevver (2013-05-01) In this paper, numerical solutions are presented for the MHD flow in a pipe surrounded by an electrically conducting medium, and under the influence of a transverse magnetic field. The boundary element method is used which discretizes only the pipe wall and is suitable for the infinite exterior region. Coupled MHD equations for the velocity and induced magnetic field inside the pipe, and the induced magnetic field equation for the outside medium are solved simultaneously taking into account also coupled bou...
Mol solution for transient turbulent flow in a heated pipe Uygur, AB; Tarhan, T; Selçuk, Nevin (2005-08-01) A computational fluid dynamics (CFD) code, based on direct numerical simulation (DNS) and method of lines (MOL) approach previously developed for the prediction of transient turbulent, incompressible, confined non-isothermal flows with constant wall temperature was applied to the prediction of turbulent flow and temperature fields in flows dominated by forced convection in circular tubes with strong heating. The code was parallelized in order to meet the high grid resolutions required by DNS of turbulent fl...
Development of electrochemical etch-stop techniques for integrated MEMS sensors Yaşınok, Gözde Ceren; Akın, Tayfun; Department of Electrical and Electronics Engineering (2006) This thesis presents the development of electrochemical etch-stop techniques (ECES) to achieve high precision 3-dimensional integrated MEMS sensors with wet anisotropic etching by applying proper voltages to various regions in silicon. The anisotropic etchant is selected as tetra methyl ammonium hydroxide, TMAH, considering its high silicon etch rate, selectivity towards SiO2, and CMOS compatibility, especially during front-side etching of the chip/wafer. A number of parameters affecting the etching are inv...
Distributed Optimal Control Problems Governed by Coupled Convection Dominated PDEs with Control Constraints Yücel, Hamdullah (2013-08-30) We study the numerical solution of control constrained optimal control problems governed by a system of convection diffusion equations with nonlinear reaction terms, arising from chemical processes. Control constraints are handled by using the primal-dual active set algorithm as a semi-smooth Newton method or by adding a Moreau-Yosida-type penalty function to the cost functional. An adaptive mesh refinement indicated by a posteriori error estimates is applied for both approaches.
Comparison of method of lines and finite difference solutions of 2-D Navier-Stokes equations for transient laminar pipe flow Selçuk, Nevin; Tanrikulu, S (2002-03-10) Performances of method of lines (MOL) and finite difference method (FDM) were tested from the viewpoints of solution accuracy and central processing unit (CPU) time by applying them to the solution of time-dependent 2-D Navier-Stokes equations for transient laminar flow without/with sudden expansion and comparing their results with steady-state numerical predictions and measurements previously reported in the literature. Predictions of both methods were obtained on the same computer by using the same order ...

Citation Formats

A. Karakuş, “GPU accelerated high-order discontinuous galerkin level set methods for incompressible multiphase flows,” Ph.D. - Doctoral Program, Middle East Technical University, 2015.