Ship Motion Predictions: A Comparison Between a CFD Based Method, a Panel Method and Measurements

In the past 50 years the research into the behavior of ships in seaway has shown a great deal of progress. From analytical solutions of 2-d hydrodynamics as derived by Ursell in 1949 to complex 3-d CFD numerical solutions that can be used nowadays. The first consistent approach for ship motions in time domain was derived by prof. T.F.Ogilvie [] as presented in his seminal paper in Bergen in 1964. Throughout the years however ship hull forms developed and the need for validation of computational procedures for the calculation of the ship motions has never gone away. In this paper we present a selection of results of a benchmark study performed by some 12 companies and universities using their state of the art computational tools. In this benchmark study the results of model tests of a modern container vessel are used. The results presented in this paper show that the panel method as described in this paper, based on the disturbed steady flow, leads to acceptable transfer functions for ship motions. The CFD approach used in this paper also produces acceptable motion transfer functions. However the results from the CFD computation for the internal load transfer functions do show a larger scatter when comparing with the results from model test.