Parametric Finite Element Methods

• Parametric Finite Element Approach is a technique that explicitly parameterizes geometry and solution fields from a reference domain to address evolving interfaces in PDEs.
• It employs piecewise polynomial finite element spaces and variational formulations to achieve convergence, stability, and structure preservation even with sharp discontinuities.
• The method is applied to problems like multi-phase flows and curvature-driven phenomena, ensuring energy dissipation and volume conservation through Lagrange-multiplier constraints.

A parametric finite element approach utilizes domain parameterization in the construction and discretization of finite element spaces for solving PDEs, especially where geometry, material data, or governing equations depend on parameters or exhibit geometric evolution. This methodology enables robust treatment of moving interfaces, parametric dependence, and even physically or geometrically sharp discontinuities, while retaining well-established theory for convergence, stability, and structure preservation. In the last decade, parametric finite element strategies have advanced application domains including interface evolution, multi-phase flows, geometric/curvature-driven phenomena, and PDE-constrained optimization.

1. Foundational Principles of Parametric Finite Element Methods

The parametric finite element method (PFEM) relies on explicit parameterization of geometry and solution fields via mappings from reference domains. For interface-evolving problems, one typically constructs a parametrization $X(\cdot, t): \mathfrak{X} \to \Omega$ mapping a fixed reference manifold $\mathfrak{X}$ into the physical domain $\Omega\subset\mathbb R^d$; the moving interface $\Gamma(t)$ is then given by $X(\mathfrak{X},t)$. Piecewise polynomial finite element spaces—often with mass-lumped inner products—are defined over the reference mesh and mapped to evolving geometries (Garcke et al., 17 Aug 2025). This framework accommodates both fitted (interface-conforming mesh) and unfitted strategies (fixed background mesh), and generalizes from stationary domains to highly general time-dependent surfaces (Barrett et al., 2019).

Weak formulations are central: instead of discretizing PDEs directly, one derives variational forms that include explicit parameter dependence, curvature, and boundary behavior, often using surface gradients, Laplace–Beltrami operators, and surface divergence (Bonito et al., 2019). For time-dependent evolutions, temporal schemes (backward Euler, BDF, Crank–Nicolson, predictor–corrector) are layered atop the parametric spatial discretization (Li et al., 2024, Gan et al., 18 Oct 2025).

2. Structure-Preservation: Conservation Laws and Energy Stability

Recent advances in parametric finite element methods reveal the centrality of structure-preserving discretizations—schemes that replicate key physical conservation laws and monotonicity identities at the discrete algebraic level (Garcke et al., 17 Aug 2025, Bao et al., 2021, Bao et al., 2022).

Many problems—two-phase Stokes/Navier–Stokes, geometric interface flows, surface diffusion—feature both a conserved (area/volume) quantity and a decaying energy functional. Structure preservation is achieved by augmenting the interface dynamics with Lagrange-multiplier constraints (enforcing discrete volume conservation and discrete energy decay), and by specialized choices for tangential velocities or normals, e.g., Simpson or midpoint averaging. For example, in two-phase Stokes, multipliers $\lambda_E,\lambda_V$ enforce at each time step exact algebraic conservation:

• Energy: $E^{m+1} -E^m +