MODELING BLOOD FLOW IN AN ECCENTRIC STENOSED ARTERY USING LARGE EDDY SIMULATION AND PARALLEL COMPUTING

Abstract

Computational fluid dynamics (CFD) is an excellent computational tool to assess the hemodynamics and detailed blood-flow structure for cardiovascular applications. Modeling turbulence for cardiovascular applications can be achieved (to some extent) using available numerical models such as Reynolds average Navier–Stokes (RANS), the large eddy simulation (LES) and the direct numerical solution (DNS). In order to develop an efficient model which is as accurate as DNS and as quick as RANS, our laboratory's focus is on LES. In this study, we develop an efficient numerical model which is based on LES and structured but non-orthogonal finite volumes. Using the proposed model, the detailed flow structure and turbulent features of the blood stream in a complicated geometry is captured. The aim of this study is to model blood-flow through an eccentric stenosis accurately and quickly. The results are similar to those obtained using DNS but in a fraction of the CPU time. The computational tools implemented in this study are based on a FORTRAN based in-house code coupled with parallel computing using SHARCNET. The developed model is a significant computational tool which can be used to assess the hemodynamic properties for cardiovascular applications, e.g., prosthetic heart valves and atherosclerosis.