Parallel Simulation of Blood Flows in 3D Patient-specific Cerebral/Coronary Arteries
We develop a parallel solution algorithm for the simulation of blood flow in 3D cerebral and coronary arteries. In this study, patient-specific geometries, hemodynamic parameters, and resistive boundary conditions are considered and examined. The problem is described by the incompressible Navier-Stokes equations and discretized using a stabilized finite element method in space with a fully implicit backward scheme in time. A parallel Newton-Krylov-Schwarz method is employed to solve the nonlinear system at each time step. To improve the convergence of the Newton method, we apply an adaptive physics-based nonlinear elimination preconditioner to remove the local high nonlinearities. Numerical results are presented to demonstrate the robustness and efficiency of the proposed algorithm with respect to some physical and numerical parameters. We also report the parallel scalability of the proposed algorithm on a supercomputer with thousands of processor cores.