Modelling of the Probabilistic Input for a Nonlinear Wave-Load Simulation Model.

摘要

For the calculation of long-term fatigue and extreme loads in vessels, a statistically correct description of the environmental conditions is required. In order to take into account nonlinear response effects, time domain methods are applied. These methods require a realistic description of an irregular wave surface in the time domain satisfying spectral and statistical properties resulting from oceanographical research. It is shown that in case of nonlinear simulation models, a random wave amplitude model should be applied. Commonly used parameters in the statistical description of the waves are the significant wave height, a characteristic wave period and a wave spectrum. Usually information about the spectral shape is not available. For that reason a modification of the Pierson-Moskowitz spectrum or a JONSWAP spectrum is usually applied. Both spectral shapes have their limitations. The use of a Wallops spectrum eliminates the problem of the choice of a suitable spectrum since the formulation is applicable to general sea states. The spectral shape is a function of the variance and the peak period and has a variable band width. It is shown that the spectrum satisfies theoretical and measured wave height distributions very well.