Wind Sector Classification Method based on Spatial Similarity of Wind Vector using Spatio-temporal Wind Data

Abstract

Measurement of wind is required for wind resource assessment. In region with complex topography, uniformity of wind speed and wind direction significantly decreases and a local wind system is formed. Therefore, clear understanding of local wind system is necessary for accurate wind resource assessment. In this study, wind sector classification method, applying cluster analysis based on spatial similarity of wind vector, was proposed to describe the local wind system more clearly. Wind sectors were classified using wind resource map and the validity of classification method was examined. Wind sector classification was applied to Jeju island and Busan, where Jeju island has relatively simple terrain and Busan has complex one. Classification result corresponds to the study areas topography and local wind system. Furthermore, wind characteristics of identical wind sector uniformly appeared and wind characteristics of distinct wind sectors appeared differently. The general validity of wind sector classification method was verified. The proposed wind sector classification method is helpful to describe the local wind system. Moreover, this method can provide a basic information for wind resource assessment.