지형복잡도에 따른 지상기반 라이다 바람데이터의 정확도평가

Abstract

The measurement campaign was performed using LiDAR (WIND CUBE v2-Leosphere) and wind sensors on meteorological mast at three measurement sites, Sumang, Gangjeong and Susan, of Jeju Island, South Korea. Each site has a different topographical complexity, and wind data was collected from 11 days to 14 days in the winter. The Ruggedness index, RIX, calculations resulted in 2.91 % (Sumang), 0 % (Gangjeong) and 2.05 % (Susan), respectively. The LiDAR error defined as LiDAR wind speed minus meteorological mast wind speed. After data filtering, the linear regression analysis was performed with 10 minutes concurrent wind data measured at met masts and LiDAR. As a result, The coefficients of determination, R², for wind speed was higher than 0.87 (Sumang), 0.99 (Gangjeong) and 0.94 (Susan), respectively. The average value of LiDAR error were calculated as the absolute values. The result of the average values of LiDAR error were 0.23 m/s at Sumang, 0.06 m/s at Gangjeong and 0.15 m/s at Susan, respectively. The standard deviation of LiDAR error were 0.54 m/s at Sumang, 0.19 m/s at Gangjeong and 0.50 m/s at Susan, respectively. In addition, the turbulence intensity and standard deviation were larger with an increase of the terrain complexity. LiDAR wind speed represented a tendency higher than met mast wind speed. The LiDAR error rate and standard deviation for wind speed showed about 2 - 6 % and 1.7 - 4.6 % difference, respectively. The CFD analysis of the Sumang area with the largest RIX, led to the conclusion that because of nearby hills (Parasitic cones), the normalized velocity at measurement point were about 0.6 at 40 m height and 0.7 at 70 m, respectively. Turbulence intensities were 0.11 at 40 m and 0.12 at 70m, respectively. Also, recirculation region occurred on parasitic cone's lee side with severe slope.