Abstract

Getting the best possible accuracy with the lowest possible computational cost is an important factor in the early design stage of ships. Potential flow-based analysis presents such a solution for seakeeping analyses. The accuracy of roll motion in potential flow is however not so good, due to the large influence from vicsous roll damping, which is missing in these calculations. This paper proposes a hybrid method, as a solution to this problem, where the viscous roll damping from Ikeda's semi-empirical method is injected into an existing 3D unsteady fully nonlinear potential flow (FNPF) method. The hybrid method is investigated using roll decay tests with the KVLCC2 test case. This investigation shows that the accuracy of simulated roll motions is significantly improved and also shows good agreement with the corresponding roll decay model tests.

INTRODUCTION

Inviscid potential flow calculations can be used to solve seakeeping problems at very low computational costs. These methods offer far cheaper alternatives than doing for instance model tests or URANS calculations. Potential flow calculations can therefore be used extensively during the early design stage of ships. The pitch and heave motions can be predicted with good accuracy, even with the older linear strip theory methods (Himeno, 1981). The roll motions will however not be very realistic in potential flow, due to high influenced from viscous roll damping. This is very unfortunate as the roll motions is indeed a very important response. The impact of roll motions can be seen from the APL China casualty in 1998, where a post-Panamax C11 class container ship lost almost a third of its containers (France et al., 2001). Another example is the container ship Svendborg Maersk, were 500 containers were lost overboard and 250 containers were damaged as a result of heavy roll motions during a passage from English Channel to Gibraltar (DMA, 2014). A lot of experimental research was conducted during the 1970s and 80s to separate the invicid and viscous roll damping. Semi-empirical formulas were developed to estimate the viscous parts, to be used together with the potential flow methods (Ikeda et al., 1978). The older linear methods can today be replaced by more advanced nonlinear potential flow methods. These newer methods still need some injection of semi-empirical viscous damping to give a fair representation of the roll motions. But is the separation of damping components still valid, considering that these older semi-empirical methods were developed in close connection to linear strip theory? (Falzarano et al., 2015) have shown that the separation of viscous and invicid damping is still valid for a panel method and Ikeda's method to predict the roll motion for the mentioned APL China vessel. (Coslovich et al., 2018) have investigated an even more advanced method, using a fully nonlinear potential flow method (FNPF) (Kjellberg, 2013) combined with Watanabe and Inoue method (W-I) (Watanabe and Inoue, 1957) to predict the viscous damping for the DTC test case (Moctar et al., 2012). This FNPF method is used also in the present paper, but instead of W-I method, Ikeda's method is used instead. Ikeda's method is believed to be a good method for the purpose, based on results from a previous comparisons of a large number of model scale roll decay tests (Alexandersson et al., 2020). The implementation of the proposed hybrid method is introduced in the next section where the underlying Ikeda's method and FNPF method are both presented. The implementation of Ikeda's method is also closely examined and an alternative way to calculate eddy damping is proposed. In the validation study of the hybrid method, roll decay tests from model tests are compared with simulations for the KVLCC2 test case. In the roll decay tests, both decay and frequency can be observed, without the presence of external forces from wind and waves. This gives a good indication of the ship's roll damping and inertia. A more thorough description of the roll decay test is given in the Roll decay test section of this paper.

This content is only available via PDF.
You can access this article if you purchase or spend a download.