Simulation of unsteady blood flows in a patient-specific compliant pulmonary artery with a highly parallel monolithically coupled fluid-structure interaction algorithm
Fande Kong,Vitaly Kheyfets,Ender Finol,Xiao-Chuan Cai
DOI: https://doi.org/10.48550/arXiv.1810.04289
2018-10-10
Abstract:Computational fluid dynamics (CFD) is increasingly used to study blood flows in patient-specific arteries for understanding certain cardiovascular diseases. The techniques work quite well for relatively simple problems, but need improvements when the problems become harder in the case when (1) the geometry becomes complex (from a few branches to a full pulmonary artery), (2) the model becomes more complex (from fluid-only calculation to coupled fluid-structure interaction calculation), (3) both the fluid and wall models become highly nonlinear, and (4) the computer on which we run the simulation is a supercomputer with tens of thousands of processor cores. To push the limit of CFD in all four fronts, in this paper, we develop and study a highly parallel algorithm for solving a monolithically coupled fluid-structure system for the modeling of the interaction of the blood flow and the arterial wall. As a case study, we consider a patient-specific, full size pulmonary artery obtained from CT (Computed Tomography) images, with an artificially added layer of wall with a fixed thickness. The fluid is modeled with a system of incompressible Navier-Stokes equations and the wall is modeled by a geometrically nonlinear elasticity equation. As far as we know this is the first time the unsteady blood flow in a full pulmonary artery is simulated without assuming a rigid wall. The proposed numerical algorithm and software scale well beyond 10,000 processor cores on a supercomputer for solving the fluid-structure interaction problem discretized with a stabilized finite element method in space and an implicit scheme in time involving hundreds of millions of unknowns.
Computational Physics,Computational Engineering, Finance, and Science,Distributed, Parallel, and Cluster Computing,Numerical Analysis