Computation of ship hull resistance using reynolds-averaged navier-stokes equations (RaNSE) in calm water

Abstract

Determination of the ship hull resistance has always been a challenging task for the ship designers. Although experimental methods such as towing tank tests are very accurate and reliable, conducting towing tank tests is very time consuming and expensive. For these reasons, now-a-days Computational Fluid Dynamics (CFD) techniques are frequently used for prediction of ship hull resistance. The goal of this thesis is to carry out a detailed numerical investigation by utilizing two RaNSE based dedicated naval CFD solvers WISDAM (Wave vIScous ow Dif- ference Accurate Method) and steadyNavalFoam for prediction of ship resistance in calm water. Computations for two benchmark container ship models KCS (KRISO Container Ship) and SR108 are made. In the present work, the viscous ow around the ship hull are represented by the Reynolds-averaged Navier-Stokes Equations (RaNSE) and the continuity equation for uid velocity and pressure. The uid is assumed viscous and incompressible, and the ow eld were considered turbulent. These numerical codes utilize the nite volume method to discretize the governing equations. Structured hexahe- dral mesh is used in case of WISDAM. In steadyNavalFoam, both structured and unstructured hexahedral mesh were used. For free surface treatment, WISDAM utilizes the Marker Density Function Method (MDF) and steadyNavalFoam utilizes the Volume of Fluid (VOF) approach. For turbulence modeling, the k ð€€€ ! SST model is used in the steadyNavalFoam code. In WISDAM, a combination of the Dynamic sub grid scale (DSGS) model and the Baldwin-Lomax turbulence model is used. The computational results are presented in a non-dimensional form. The contours of ow properties are also visualized. The results from KCS were compared with ex- perimental results from Tokyo CFD Workshop 2015. However, lack of experimental data made it di cult to validate the resistance computed for SR108. The results obtained for the calm water resistance for KCS by steadyNavalFoam is satisfactory when compared to experimental data. However, WISDAM results show a large deviation from the experimental data for KCS. It has also been found that com- puted values of calm water resistance by WISDAM for SR108 were signi cantly less than those computed by steadyNavalFoam. The geometry of the two containership hulls also played a dominant role in the wave elevation and pressure and velocity distribution.