Abstract
We performed the regional frequency analysis to estimate the extreme wind speed in southern coastal region of Korea using annual maximum wind speed (AMWS) and linear-moments method proposed by Hosking. Data at all station is found to be accordance. It is confirmed that all data could represent one homogeneous region through the heterogeneity test. Generalized Normal (GNO) and Generalized Extreme Value (GEV) distributions were selected as probability distributions representing the southern coastal region from the goodness-of-fit tests. The stability of the two distributions was evaluated using the relative error (RB) and the root mean square error (RRMSE). It is found that the GNO distribution is more stable than the GEV distribution. Finally, the extreme wind speed of the stations is estimated use the quantile representing the southern coast region, the average, median value, and location parameters of each station. The regional frequency analysis is effective to estimate the extreme wind speed at station where data are insufficient or not measured.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
