A pulsatile 3D flow relevant to thoracic hemodynamics: CFD - 4D MRI comparison

Application Challenge AC7-04 © copyright ERCOFTAC 2021

CFD Simulations

Overview of CFD Simulation

Large Eddy Simulations were carried out using the in-house, massively parallel and multiphysics YALES2BIO solver based on YALES2 [4] developed at CORIA (Rouen, France). YALES2BIO is dedicated to the simulation of blood flows at the macroscopic and microscopic scales. The base is a solver for the incompressible Navier-Stokes equations. The equations are discretised using a finite-volume fourth-order scheme, adapted to unstructured meshes [5,6]. The divergence-free property of the velocity field is ensured thanks to the projection method introduced by Chorin [7]. The velocity field is first advanced in time using a low-storage fourth-order Runge-Kutta scheme [6,8] in a prediction step. This predicted field is then corrected by a pressure gradient, obtained by solving a Poisson equation to calculate pressure. This equation is solved with the Deflated Preconditioned Conjugate Gradient algorithm [9]. YALES2BIO was validated and successfully used in many configurations relevant to cardiovascular biomechanics (see [10] for a list of publications). The boundary conditions applied at the inlet came from the data acquired during the experiment (2D cine PC-MRI).

Simulation Case

Computational Domain

The geometry used in the calculation is the same as the one presented in the test data section. The computational domain has one inlet and one single outlet, so that the flow split at the main tube/collateral junction does not depend on the details of the outlet boundary condition. The mesh (Fig. 5) used to perform the simulation consisted of an unstructured tetrahedral mesh generated with GAMBIT 2.4.6 (ANSYS, Inc., Canonsburg, PA). The mesh includes 3,812,438 cells with an average cell volume of 38 mm³ (representative cell size of 0.7 mm).