The results of a study on the statistic properties of the stochastic processes of both irregular wave and riser response were presented in the paper OMAE2014-23196. Following on from that work, the present paper is devoted to the statistic properties of the mean-crossing maxima of the stochastic process of riser response and the statistic properties of the largest maxima, resulting from a number of 3-hour simulations with different wave seeds. The objective of the study is to find a more efficient method for evaluating the most probable maximum (MPM) value of riser response, i.e. using as few as possible simulations to find a reliable expectation to the MPM.
Global dynamic analysis with irregular wave is widely used in riser system design, though it is very time-consuming compared with regular wave. This is because irregular wave can more realistically model the sea state in the field of the riser system compared with regular wave. However time consuming irregular wave analysis may be there remains a perplexing issue, which is the randomness in the analysis results induced by wave seed; i.e. the extreme value of riser response resulting from a 3-hour simulation varies with the wave seed which is arbitrarily selected for the simulation. If a number of 3-hour simulations are performed, then a statistic method is used to evaluate the MPM value, two questions must be answered. First, how many simulations should be run, and second, what statistic method should be used?
Based on the study to the statistic properties of the maxima of riser response and the largest maxima, answers to the above questions are proposed. For universal applicability the study was conducted reflecting four risers which had different configurations and water depths.