Performance of k-ω and k-ε Model for Blood Flow Simulation in Stenosed Artery

Abstract

Blood flow through arterial stenosis can play a crucial role at the post stenotic flow regions. This produces a disturbance in the normal flow path. The intensity of the flow disturbance (i.e. laminar, transitional and turbulent flow characteristics) depends not only on the severity of the stenosis but also on the pattern of the geometrical model. In that case, the turbulence model plays vital role to measure these flow disturbances. However, it is very important to choose a proper flow simulation model that can predict the flow behavior of fluid accurately and efficiently with less computational cost and time. Thus, this study aims to analyze the results of two turbulence models i.e. k-ω and k-ε for blood flow simulation to compare their performance for the prediction of the flow behavior. Simulations have been performed with 75% area reductions in the arteries. The results of simulation show that, the flow parameters obtained from the k-ε model exhibits lack of fits with the experimental data. On the other hand, k-ω model can capture large scale gradient in the different parameters of blood flow and has a good agreement with the experimental data. This study suggests that, k-ω model has the better performance comparing to k-ε model to predict the behavior of blood flow in stenosed artery. GANIT J. Bangladesh Math. Soc. 40.2 (2020) 111-125

CFD Study of the Effect of the Angle Pattern on Iliac Vein Compression Syndrome

“…This result suggests that the k-ε model used in the present investigation is a suitable Reynolds-averaged Navier–Stokes turbulence model that captures the turbulent transition and vessel obstruction found in compressed iliac veins. However, this finding is inconsistent with research into CFD analysis in stenosed vessels by Kabir et al [ 38 ] which concluded that simulations using k-ω models are more realistic than those using the k-ε models, perhaps because their simulation was performed in blood vessels that exhibited non-Newtonian properties, rather than in the Newtonian fluid setting of the present study. In addition, studies have shown that the k-ω model is overly sensitive to turbulent fluids, and thus more suited to analyzing shear stress in near-wall regions [ 42 ], which falls beyond the scope of the present study.…”

“…Blood plasma, which is an incompressible, Newtonian-homogeneous fluid that makes up over 50% of blood [ 33 ], exhibits Newtonian behavior [ 25 ]. In microcirculatory systems, however, such as small vessels and capillaries with diameters under 1 mm, blood can be treated as a non-Newtonian fluid [ 38 ] whose changing viscosity can be determined using a modified Quemada model [ 30 ]. Previous reports analyzing stenosis conditions have stated that assuming a non-Newtonian state may lead to better simulation results compared to a Newtonian model [ 25 ], because a non-Newtonian model is more appropriate for analyzing wall shear stress [ 40 ] and arterial stenosis in regions with Reynolds numbers below 100 [ 27 , 41 ].…”

“…Since the diameter of blood vessels in the human body varies from small capillaries to large arteries, the Reynolds number varies from 1 in small arterioles to 4000 in the largest artery [ 44 ], model assumptions can affect mathematical simulation of blood flow different. Laminar flow has been assumed to give a reasonable approximation of blood flow through larger vessels [ 43 ] and has been applied in most analyses of stenotic flow and wall shear stress [ 32 , 38 ]. Re values calculated at the compression site in the present study fell between 2100 and 4000, meaning that our CFD model, rather than assuming laminar flow as in previous studies, needed to assume transitional flow.…”

“…It is therefore important to use an appropriate turbulence model to calculate the flow field and obtain the hemodynamic parameters. Two turbulence models frequently used to study disturbance at the post stenotic region in arteries are the k-ω and k-ε turbulence models [ 38 ]. In this study, we first investigated the ability of these two models to predict flow behavior in venous stenoses.…”

Calibration of patient-specific boundary conditions for coupled CFD models of the aorta derived from 4D Flow-MRI

“…The governing Navier Stokes equations solved numerically in Ansys Fluent ® utilizing a finite volume method, a standard k-ω turbulence model, and the Pressure-Implicit with Splitting of Operators (PISO) algorithm at 10 iterations per time step (dt = 0.001s) ( López et al, 2015 ). The standard k-ω turbulence model was utilized since it has been shown to be more accurate when compared to experimental data and maintain stability in regions of stagnation and high fluid acceleration which are common in AD cases ( Song et al, 2003 ; Kabir et al, 2020 ). CFD Simulations were performed on a single node of the ARCHIE-WeSt cluster at the University of Strathclyde.…”