Ship Airwake Sensitivities to Modeling Parameters

Abstract

Accurate models for the prediction of ship airwake flowfields are critical to the development of realistic flight simulation tools for aircraft carrier launch and recovery operations. The accurate computation of the ship airwake can be very challenging due to the complexity of the ship geometry, the size of and difficulty in generating a suitable computational mesh, and the large range of length and time scales present in the unsteady flowfield. The present paper investigates the sensitivity of the airwake solution to several modeling parameters, including geometric complexity and the resolution of boundary layers, with the aim of determining the level of fidelity required to obtain an accurate solution.Results are compared to wind tunnel experimental measurements. The results of these studies show that, in general, a majority of the airwake flow features are characterized by bluff body shedding from the larger geometric entities that comprise the ship geometry. Depending on the requirements and intended use of the solution, a certain tradeoff can be reached between solution turn-around/grid generation time and solution accuracy.

Pressure and velocity measurements of air flow past a proposed generic Aircraft carrier geometry

“…19,23,27 (iii) Since the GAC is intended as a simplified, geometric benchmark model for aircraft carrier studies, intricate details like sensors and minor equipment mounted on the island or the flight deck have been omitted and only the major structures are retained. Studies by Shipman et al 32 support this argument, that such small structures would not create effects of a magnitude that is of interest to a generalized study like the present work. Thus, the GAC has resulted in a simplified geometry that is easy to fabricate and duplicate experimentally or simulate numerically for CFD studies, meshing requirements around the canonical forms used in the GAC are far more forgiving than complex three-dimensional geometries.…”

Application and verification of internal boundary conditions for antenna mast wake predictions

“…However, the size of the wake behind the yardarm poles is again over-predicted even for this case. In addition to applying the IBC approach, a parallel effort was executed that employed standard surface gridding techniques to grid the actual antenna structure (Shipman et al, 2005). The CFD code CRUNCH was used for this portion of the study.…”

Implicit large eddy simulation of ship airwakes

“…Results by Shipman et al (46) investigated the effects of geometrical representation of the ships on the quality of the airwake prediction, concluding that reasonable simplifications can be made to the ships superstructure without greatly affecting the majority of the airwake. Additionally, they note that including boundary layer resolving grid components only marginally improves the simulation quality.…”

“…Note that in generating the grids no attempt was made to model the boundary layer attached to the ship. This simplification of the meshing requirements makes the prediction of flow separation from gradually curved surfaces less accurate, however previous studies by Shipman et al (46) investigated in depth the effects of resolving the boundary layer and concluded it has minimal influence on the quality of the wake prediction in terms of mean and fluctuating velocities. It is not clear in the paper of Shipman et al whether the resolved boundary layer study had the same number of points, and same computational time as the unresolved case.…”

“…There are inevitably trade-offs required to allow computation of many different wind directions for such a complex geometry. Computational uncertainties due to reductions in formal order of accuracy in highly skewed portions of the grid (for a detailed description see Jordan (25) ), and lack of perfect boundary layer resolution (see, for example Shipman et al (46) ) will not be explored in detail.…”