Posters

Abstract

Accurate representation of breaking wave hydrodynamics is essential for assessing extreme metocean impacts on offshore wind infrastructure under changing climate conditions. Breaking waves exhibit complex physical phenomena including violent free-surface deformation, air entrainment, turbulent mixing, and flow separation—processes that challenge traditional computational modeling approaches. This study presents a systematic comparison of two fundamentally different computational fluid dynamics (CFD) methodologies for breaking wave physics simulation validated against large-scale experimental data from Germany's GWK wave facility. METHODOLOGY: Two advanced CFD approaches were evaluated: the mesh-based REEF3D solver employing level set interface tracking with fifth-order WENO spatial discretization and k-ω SST turbulence modeling, and the particle-based DualSPHysics framework using Lagrangian weakly-compressible SPH formulation with GPU acceleration. Both methods simulated breaking waves impacting a 0.7 m diameter vertical cylinder across five distinct scenarios (steep approaching waves, direct structural breaking, upstream breaking impacts, developed plunging breakers, and post-breaking bores) with wave heights of 1.1–1.7 m. REEF3D employed adaptive mesh refinement localizing computational resolution around critical regions, while DualSPHysics naturally handled violent free-surface fragmentation through its meshless Lagrangian formulation. INNOVATIVE CONTRIBUTIONS: This represents the first systematic comparison of Eulerian level set and Lagrangian SPH methods specifically for breaking wave hydrodynamics on offshore structures. The comparative framework evaluates hydrodynamic representation accuracy, computational efficiency characteristics, and practical applicability across different breaking wave regimes relevant to metocean impact studies. RESULTS: Validation against experimental data demonstrated both methods achieve peak force predictions within 5% deviation from measured values. REEF3D showed 40% computational efficiency improvement through adaptive mesh refinement for the validated scenarios, with typical runtimes of 18–25 hours per case on 24 cores. Breaking wave physics characterization revealed both approaches successfully captured wave steepening, crest overturning, and jet formation dynamics, with differences in air entrainment representation: REEF3D's level set method provided smooth interface tracking suitable for pressure field analysis, while DualSPHysics naturally represented fragmented free-surface regions and splash-up phenomena without numerical diffusion. Free-surface elevation predictions showed excellent agreement with experimental measurements (correlation coefficients >0.95) for both methods across all scenarios. Computational resource requirements differed significantly: REEF3D required 15.12 million cells for uniform mesh validation cases versus DualSPHysics' 550,000 particles, with comparable accuracy. Pressure distribution analysis indicated REEF3D provided smoother spatial gradients beneficial for structural load assessment, while DualSPHysics excelled in capturing post-impact flow separation and reattachment patterns through Lagrangian particle tracking. CONCLUSIONS: Both CFD approaches demonstrate capability for accurate breaking wave hydrodynamics modeling applicable to metocean impact assessment. Method selection depends critically on application requirements: REEF3D offers computational efficiency advantages and smooth pressure fields for integrated force calculations, while DualSPHysics provides superior representation of violent free-surface fragmentation and complex post-breaking flow patterns. The validated comparative framework provides evidence-based guidance for selecting appropriate computational tools for breaking wave studies in climate change impact assessments and offshore wind infrastructure design under evolving metocean conditions.