Comparative Analysis of High-Fidelity and Reduced-Order Models for Nonlinear Wave–Bathymetry and Wave–Structure Interactions

This paper presents a computational study of wave–bathymetry and wave–structure interaction problems using advanced numerical techniques based on high-fidelity, two-phase Navier–Stokes (TpNS) flow and reduced-order, fully nonlinear potential flow models. For high-fidelity simulations, the TpNS equations are discretized using the finite-element method, with free-surface evolution captured through a hybrid level-set (LS) and volume-of-fluid (VOF) formulation. A monolithic, phase-conservative LS equation is introduced to mitigate mass loss and interface smearing, combined with a semi-implicit projection scheme. Hydrodynamic forces are resolved using a high-order, phase-resolving cut finite-element method (CutFEM), which enables the representation of complex solid geometries within a fixed background mesh. An equivalent polynomial of Heaviside and Dirac distributions ensures accurate evaluation of surface and volume integrals. Hence, no explicit generation of cut cell meshes, adaptive quadrature, or local refinement is required. For reduced-order modeling, a fast regularized boundary integral method (RBIM) is employed to solve the fully nonlinear potential flow. Singular and near-singular integrals are treated using a subtract-and-addition technique based on auxiliary functions derived from Stokes’ theorem, allowing direct application of high-order quadrature without conventional boundary element discretization. An arbitrary Lagrangian–Eulerian (ALE) formulation is adopted to enforce free-surface boundary conditions while avoiding excessive mesh distortion. The proposed approaches are applied to investigate highly nonlinear wave transformation over complex bathymetry and wave-induced dynamics of floating structures, including eddy-making damping effects. Numerical results are validated against experimental measurements. These two modeling approaches represent complementary levels of physical fidelity and computational efficiency, and their systematic comparison clarifies the trade-offs between computational accuracy, efficiency, and cost for practical marine problems.