The Simulation And The Statistical Property For Deep Water Nonlinear Long Crest Wave

Nowadays, ocean development of China is stepping into deep water. The random waves in real sea are likely to evolve into waves with strong nonlinearity. In engineering, it attracts a lot of attentions on the amplitude of these nonlinear wave crests (troughs) and wave particle velocity at the peaks.The multidirectionality of the deep water random waves is not concerned in this thesis.We simulate the deep water long crest waves by using deep water Green-Naghdi (GN)model.1) In the simulation of strong nonlinear irregular waves, we find the level of GN model affects the simulating results significantly if the representative wave number is specified.Different level of GN model depends on velocity assumptions which describe the vertical structure of flow field. Higher level GN model means better description for velocity field.By using the 3rd level GN model,a good result could be achieved for the simulation of strong nonlinear irregular waves whose significant wave height is as high as 10 meters. The representative wave number corresponding to half energy of spectrum deems as the best.2) In this thesis, we also simulate the evolution of gravity wave groups with moderate wave steepness to investigate the wave-wave interactions. Experimental and numerical methods are both used in this investigation to verify the ability to simulate this problem by using the GN model. It is shown that the deep water GN model could well simulate the separation of wave packet and the growing of nonlinearity during the evolution of gravity wave groups. And strong nonlinear waves whose wave steepness grows twice than the initial wave could be observed during the simulations.3) Finally, the 3rd level GN (GN-3) model and ITTC double parameter spectrum are used to simulate the long crest wave. Based on the rules,we choose a series of working conditions combined with different significant wave height and eigenperiod from the 3rd to 9th sea states corresponding to different ITTC spectra. Statistical property of the amplitude and particle velocity of wave crest or wave trough is obtained under corresponding sea states. Then this property is compared with Rayleigh distribution which is based on the linear theory. It is shown that the GN distribution is more likely to deviate from Rayleigh distribution when the sea state grows higher. The Rayleigh distribution seems to be insufficient prediction for crest distribution. For trough distribution, it is likely to be over prediction. The reduction in eigenperiod and increasing in significant wave height will contribute to the deviation. In addition,based on the numerical results of GN model,a simplified fitting function is proposed. It is very convenient to use in engineering. From the comparison of numerical results between GN-3 and GN-4 models, we find the GN-3 model could give convergent results over all.