The Calculation of Extreme Wind Speed Based on Bayesian Method

Wind speed is one of the main factors to be considered in the design of offshore structures. However, due to the long period fluctuation of typhoons and climate change, the return levels of the wind speed derived using the samples of different years are considerably different, which can lead to unstable results. To solve this problem, in this study, the extreme wind speed is calculated based on the Bayesian method. In the Bayesian parameter estimation method, the parameters of the distribution are considered as random variables, and thus, the priori information can be fully exploited for parameter estimations. The data utilized in this study are deriveds from the best data set of the northwest Pacific tropical cyclone. A typical point in the South China Sea is selected as the calculation point. The samples are determined are determined using a parametric wind model developed by Cardone. Considering the influence of the difference samples of in different time instants, the samples are divided into early and recent samples, which are used as priori and posteriori information, respectively. The normal distribution is used as the priori distribution, and the estimator of the parameters is the expectation of the posteriori distribution, which is calculated using the Markov Chain Monte Carlo (MCMC) algorithm. By comparing the results pertaining to between Bayesian and maximum likelihood method, it can be observed that the former is more stable and reliable.