VALIDATION OF WAVE PROPAGATION IN NUMERICAL WAVE TANKS Tim Bunnik MARIN René Huijsmans MARIN ABSTRACT During the last few years there has been a strong growth in the availability and capabilities of numerical wave tanks. In order to assess the accuracy of such methods, a validation study was carried out. The study focuses on two types of numerical wave tanks: 1. A numerical wave tank based a non-linear potential flow algorithm. 2. A numerical wave tank based on a Volume of Fluid algorithm. The first algorithm uses a structured grid with triangular elements and a surface tracking technique. The second algorithm uses a structured, Cartesian grid and a surface capturing technique. Validation material is available by means of waves measured at multiple locations in two different model test basins. The first method is capable of generating waves up to the break limit. Wave absorption is therefore modeled by means of a numerical beach and not by mean of the parabolic beach that is used in the model basin. The second method is capable of modeling wave breaking. Therefore, the parabolic beach in the model test basin can be modeled and has also been included. Energy dissipation therefore takes place according to physics which are more related to the situation in the model test basin. Three types of waves are generated in the model test basin and in the numerical wave tanks. All these waves are generated on basin scale. The following waves are considered: 1. A scaled 100-year North-Sea wave (Hs=0.24 meters, Tp=2.0 seconds) in deep water (5 meters) 2. A scaled operational wave (Hs=0.086 meters, Tp=1.69 seconds) at intermediate water depth (0.86 meters) generated by a flap-type wave generator. 3. A scaled operational wave (Hs=0.046 meters, Tp=1.2 seconds) in shallow water (0.35 meters) generated by a piston-type wave generator. The waves are generated by means of a flap or piston-type wave generator. The motions of the wave generator in the simulations (either rotational or translational) are identical to the motions in the model test basin. Furthermore, in the simulations with intermediate water depth, the non-flat contour of the basin bottom (ramp) is accurately modeled. A comparison is made between the measured and computed wave elevation at several locations in the basin. The comparison focuses on: 1. Reflection characteristics of the model test basin and the numerical wave tanks. 2. The accuracy in the prediction of steep waves. 3. Second order effects like set-down in intermediate and shallow water depth Furthermore, a convergence study is presented to check the grid independence of the wave tank predictions. INTRODUCTION It becomes more and more important to validate and use prediction tools for wave generation because: 1. They help to better understand the physics 2. Thy help to improve the performance of test basins 3. They can be used as coupling methods to other tools (for example to predict wave impact). This paper compares the results of two different methods for wave generation with the results of wave measurements in two different model test basins. The first method uses the Navier-Stokes equations as the basis of the numerical model. The second one applies potential flow theory. Both methods take into account non-linear free surface effects. This paper describes the mathematical and numerical features of both models, the layout of the model test basins, and finally compares computed and measured wave trains in several ways. The numerical simulation of nonlinear waves follows from the mathematical model that is used to describe the wave dynamics. Based on the continuum hypothesis, the 1 Copyright © 2005 by ASME Proceedings of OMAE2005 24th International Conference on Offshore Mechanics and Arctic Engineering (OMAE 2005) June 12-17, 2005, Halkidiki, Greece OMAE2005-67221