DIMENSION-REDUCTION METHOD FOR MUTIDIRECTIONAL IRREGULAR WAVE SIMULATION

Based on three kinds of conventional Monte Carlo simulation methods for multidirectional irregular waves, the corresponding original spectral representations of a spatio-temporal random field are proposed. Then three dimension-reduction methods for simulating multidirectional irregular waves are derived by introducing the dimension-reduction idea of a random function. As a result, the random field of multidirectional irregular waves is accurately represented by employing merely 2-4 elementary random variables. Numerical examples show that the effective wave heights of the representative samples simulated by the three dimension-reduction methods can satisfy the normative requirements. Meanwhile, the second-order statistics (simulated values) of the representative sample set fit well with the target values. In general, the accuracy of the proposed method is higher than that of the conventional Monte Carlo scheme, which verifies the validity of the former. In addition, by analyzing the simulation accuracy and efficiency of the three dimension-reduction methods, the most suitable scheme of multidirectional irregular waves for engineering application is proposed. This provides an effective way, combining the probability density evolution theory, for the refined dynamic analysis of offshore engineering structures subjected to random waves.