Extreme value frequency analysis of annual wind for grate Khorasan

The annual fastest wind analysis of the important requirements in the design of tall building, Dams, etc. The annual fastest winds of grate Khorasan province synoptic stations selected for statistically analysis. The data were screened and the basic tests (randomness, homogeneity, independence, etc.) were investigated. The data of four station have outliers and not random. Five stations were not stationery and heterogeneous. The seven distribution functions Gumble type 1, gamma 2 parameters, lognormal 2 and 3 parameters, Generalized Pareto, generalized Extreme values and Pearson type 3 with five estimation methods (conventional moments, maximum likelihood, probable weighted moments, maximum entropy and modified moments) were fitted. The Ks goodness of fit test applied to determine the suitability of the chosen distribution and superior distribution. Three priorities for each station were established. Because of the physical behavior of this phenomenon is well justified. Generalized Pareto 21 times and General Extreme value 11 times were the best fitted distributions. Probability weighted moments were the best estimation methods in all distributions.