Meaning
Numerical procedure that subdivides a continuous geometric domain into a finite number of smaller, distinct subdomains called elements to solve differential equations. Utilizing finite element discretization allows developers to solve complex thermal and electrochemical equations across complex cell geometries. The method converts infinite-dimensional problems into solvable systems of algebraic equations.
Geometric Partition
Solid models are split into mesh networks to resolve the governing equations at the nodal points. When applying finite element discretization, the mesh density must be higher near boundaries where gradients are steepest, such as at the current collector tabs. This partitioning ensures that rapid thermal gradients are captured accurately.
Resolution Impact
Coarse grids reduce computing time but can introduce numerical errors in the thermal profiles. If the finite element discretization is too fine, the simulation may take days to complete without offering any meaningful increase in practical accuracy. Finding a balance involves conducting mesh independence studies to locate the threshold where the solution becomes stable.
Computational Demand
Modeling the mechanical expansion of battery packs requires large computing resources due to the non-linear contact forces between cells. Parallel processing can accelerate these calculations. Calculations must run on multi-core workstations to deliver timely results during the development phase.
