A general high-order finite element method for solving partial differential equations, called the COnstraints Oriented Library (COOL) method, is presented. This hp approach takes into account the underlying nature of the corresponding physical problem, and thus avoids the generation of nonphysical solutions. In the COOL method, all terms in a variational form are represented by the same functional dependence and by the same regularity, thus eliminating regularity constraints imposed by standard numerical methods. External constraints, such as the incompressibility condition appearing in the Maxwell or Navier-Stokes equations, can then be satisfied identically and are eliminated algebraically. This reduces the number of variables, and leads to well-conditioned matrix problems. The consequence is that only physically relevant solutions remain. The COOL method also satisfies automatically internal constraints, such as occur for the grad(div) and curl(curl) operators, and this for any geometry. This approach can be applied to a wide range of physical problems, including fluid flows, electromagnetics, material sciences, ideal linear magnetohydrodynamic stability analysis, and Alfvén wave heating of fusion plasmas. Results obtained by applying the COOL method to the grad(div) and curl(curl) operators, the Stokes problem, and the steady and unsteady Navier-Stokes equations are presented.
This paper presents the results of a benchmark study that compares a number of numerical models applied to a specific problem in the context of hydrogen flow and transport in a nuclear waste repository. The processes modelled are two-phase (water and hydrogen) immiscible compressible two-component transient flow in a heterogeneous porous medium under isothermal conditions. The three-dimensional (3D) model represents a module of a repository for high-level waste in a clay host rock. An upscaling technique and a vertex-centred finite-volume method are employed to yield very accurate solutions. Since the full range of results required in the benchmark is too large to be displayed in this paper, we focus on the evolution of the pressures, the saturations, the fluxes and the comparison of the numerical results with the other participants. A homemade C++ upscaling code and the parallel multiphase flow simulator DuMuX have been adopted for this study.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.