Approaches to Validation of CFD Models for Far Ship Wake

The centerline wake of surface ships can extend to tens of kilometers on synthetic aperture radar (SAR) images. However, the hydrodynamics of far wakes of ships are not well understood. Our assumption is that far from the ship, the pattern of flows comprising the wake is represented by longitudinal coherent vortices, which gross parameters only slowly change in the direction along the wake. In order to model this process, we have run a set of non-hydrostatic simulations. The simulations also included dynamics of freshwater plumes in the upper ocean because such types of formations often produce sharp fronts, which can be confused with ship wakes. We have used 2D and 3D setups with slippery and free upper boundary and with several turbulence closure schemes including k-ω, standard LES, and Hybrid LES. The models have been implemented in CFD Fluent and simulated such processes as wind-wake and wind-plume interactions and formation of sharp frontal lines on the sea surface. In order to justify the choice of grid and model parameters, we have performed a series of validation tests. These tests included grid and time convergence, sensitivity to geometric parameters, and comparison to available experimental data including photo and SAR images of ship wakes. The Hybrid LES turbulence model has demonstrated a more realistic performance than the other two tested turbulence closure models. The application of CFD to these problems has resulted in a qualitative level of information. Providing information on the level of absolute quantities requires validation with the data from specialized field and laboratory experiments.