RANS Maneuvering Simulation of Esso Osaka With Rudder and a Body-Force Propeller

Abstract

A simplified potential theory-based infinite-bladed propeller model is coupled with the Reynolds averaged Navier-Stokes (RANS) code CFDSHIP-IOWA to give a model that interactively determines propeller-hull-rudder interaction without requiring detailed modeling of the propeller geometry. Computations are performed for an open-water propeller, for the Series 60 ship sailing straight ahead and for the appended tanker Esso Osaka in different maneuvering conditions. The results are compared with experimental data, and the tanker data are further used to study the interaction among the propeller, hull, and rudder. A comparison between calculated and measured data for the Series 60 ship shows fair agreement, where the computation captures the trends in the flow, that is, the flow structure and the magnitude of the field quantities together with the integral quantities. For the tanker, the flow study reveals a rather complex flow field in the stern region, where the velocity distribution and propeller loading reflect the flow field changes caused by the different maneuvering conditions. The integral quantities, that is, the propeller, hull, and rudder forces, are in fair agreement with experiments. No formal verification and validation are performed, so the present results are related to previous work with verification and validation of the same model, but without the propeller. For the validated cases, the levels of validation are the same as without the propeller, because the validation uncertainties, that is, the combined experimental and simulation uncertainties, are assumed to be the same for both cases. Based on this, validation is obtained for approximately the same cases as for the without-propeller conditions, but the comparison errors, that is, the difference between experiment and calculation, are different. For instance, the difference between computation and experiment for the ship resistance is generally larger with the propeller than without, whereas the opposite is the case for the rudder drag. Summarizing the results, the method shows encouraging results, and taking the effort related to modeling the propeller into account, the method appears to be useful in connection with studies of rudder-propeller-hull related flow problems, where the real propeller geometry cannot be modeled.