A Numerical Trim Variation Study for Ships Operating in Off-design Conditions. the Validation of Rans Code Results and the Explanation of the Physical Effects of the Flow around a Hull Subject to Trim Variation

Fuel efficiency is an important factor for the shipping industry both regarding new build vessels and existing vessels. A method to reduce the fuel consumption of existing vessels operating in off design conditions is to trim a vessel in the most optimum trim condition. The current approach to determine the most efficient trim condition is to perform propulsion tests under different trim conditions in a towing tank. To determine the usability of the CFD program PARNASSOS a trim variation study is performed on a vessel that is already tested in one of the tanks of MARIN. The trend of the propulsion from the tank tests is used to validate the trend of the resistance from the CFD calculations. Before validating the trend of the resistance with the trend of the propulsion it was checked whether or not the change of the resistance is dominant over the change in propulsive efficiency. It turned out that the change of the resistance is dominant over the change of the propulsive efficiency and therefore the trend of the propulsive power can be used to validate the trend of the resistance. The uncertainty of the CFD calculations is determined using a grid refinement study according to the method developed by Luis Eça and Martin Hoekstra [Eça and Hoekstra, 2014]. According to this method the uncertainty of the performed calculations is estimated at 3.5%. The uncertainty of the tank tests are estimated at 0.9% according to Martijn van Rijsbergen [van Rijsbergen, 2014]. Unfortunately the uncertainty value obtained using the current method for estimating the uncertainty is not small enough to validate the trend of the resistance with the trend of the propulsion from the towing tank tests. The fact that the uncertainty determined is not small enough to validate the trend obtained from the CFD calculations doesn’t necessarily invalidate the assumption that the results of the CFD calculations are correct. It is expected that the uncertainty of the difference is smaller than the relative uncertainty, an expectation that is supported by the fact that the fits of the power series estimation for the resistance coefficients of the even keel condition and the 1.5m aft trim condition show the same trend and the two fits do not cross each other. There are at least two ways to further reduce the estimate of the uncertainty. The first one is to use finer grids. The second one is to further optimize the method to determine the uncertainty of these CFD calculations and reduce the influence of data scatter and non similarity of the grids on the estimate of the uncertainty. The results of the CFD calculations are analyzed as if the data is validated. Trimming the vessel aft resulted in an increase of the total resistance. Trimming the vessel forward reduced the total resistance. The total resistance, the frictional resistance and the hydrodynamic pressure resistance increase when the vessel is trimmed aft while the hydrostatic pressure resistance reduces. These results show that the change in frictional resistance and the hydrodynamic pressure resistance are dominant over the change in hydrostatic pressure resistance. For moderate changes of trim the change in wetted surface is dominant over the change of shear stress regarding the change in frictional resistance. For extreme changes in trim the cause of the increase of the frictional resistance is a sheet vortex developing at the bow of the vessel and running aft along the bilge of the vessel. This sheet vortex influences the local thickness of the boundary layer and therefore influences the local shear stress. At the transom of the vessel the local hydrodynamic pressure resistance is reduced caused by the submergence of the transom. Trimming the vessel aft reduced the pressure recovery at the stern of the vessel which has a negative influence on the local hydrodynamic pressure resistance. At the forward shoulder of the vessel the hydrodynamic pressure resistance increases as well when the vessel is trimmed aft. Trimming the vessel aft resulted in an increase of the hydrodynamic pressure resistance which shows that the effect of the pressure recovery at the stern of the vessel and the increase of hydrodynamic pressure at the forward shoulder of the vessel are dominant over the effect of the presence of a dead water zone behind the transom of the vessel.