A scalable DG solver for the electroneutral Nernst-Planck equations
arXiv:2112.09271 · doi:10.1016/j.jcp.2022.111859
Abstract
The robust, scalable simulation of flowing electrochemical systems is increasingly important due to the synergy between intermittent renewable energy and electrochemical technologies such as energy storage and chemical manufacturing. The high Péclet regime of many such applications prevents the use of off-the-shelf discretization methods. In this work, we present a high-order Discontinuous Galerkin scheme for the electroneutral Nernst-Planck equations. The chosen charge conservation formulation allows for the specific treatment of the different physics: upwinding for advection and migration, and interior penalty for diffusion of ionic species as well the electric potential. Similarly, the formulation enables different treatments in the preconditioner: AMG for the potential blocks and ILU-based methods for the advection-dominated concentration blocks. We evaluate the convergence rate of the discretization scheme through numerical tests. Strong scaling results for two preconditioning approaches are shown for a large 3D flow-plate reactor example.
arXiv comment26 pages, 7 figures
References in corpus (8)
- Firedrake: automating the finite element method by composing abstractions
- A discontinuous Galerkin method for nonlinear parabolic equations and gradient flow problems with interaction potentials
- A free energy satisfying discontinuous Galerkin method for one-dimensional Poisson--Nernst--Planck systems
- Solver composition across the PDE/linear algebra barrier
- Automated generation and symbolic manipulation of tensor product finite elements
- A parallel edge orientation algorithm for quadrilateral meshes
- A structure-exploiting numbering algorithm for finite elements on extruded meshes, and its performance evaluation in Firedrake
- High level implementation of geometric multigrid solvers for finite element problems: applications in atmospheric modelling