Applicability Analysis of Sea Surface Wind Field Data for Yangjiang Offshore Wind Farm in Guangdong Province

Abstract

[Introduction] In order to test the applicability of sea surface wind field data set in Yangjiang offshore wind farm area, this paper tests and evaluates the 10 m wind field of daily gridded advanced scatterometer (DASCAT), European centre for medium-range weather forecasts reanalysis v5 (ERA5) and final reanalysis data (FNL). [Method] The research was based on the 10 m wind fields at four sites in the Yangjiang offshore wind farm area in Guangdong province. [Result] The results demonstrate five pionts: (1) The wind speed correlations are above 0.8, with ERA5 the highest. The RMSE of wind speed are within 2.6 m/s, with DASCAT being the best in SWZ and ERA5 being the best in DWZ. Both FNL and ERA5 show significant underestimation of wind speed in SWZ, while DASCAT has smaller mean wind speed deviation than those of FNL and ERA5, and ERA5 mean wind speed is closer to those of observation. (2) The wind direction correlation reaches more than 0.75, with ERA5 being the highest in SWZ and FNL being the highest in DWZ. The RMSE of wind direction are within 35°, and ERA5 errors are the minimum. However, DASCAT and FNL are both closer to the observed predominant wind direction. (3) The statistics of the RMSE of wind speed in each wind speed period show that FNL is the minimum in the low wind speed period while DASCAT is the minimum in the medium and high wind speed periods in SWZ. ERA5 is the minimum in all wind speed period in DWZ. In both SWZ and DWZ, ERA5 has the highest wind speed correlation in both the low and medium wind speed periods, and FNL has the highest wind speed correlation in the high wind speed period. (4) The monthly wind speed errors of DASCAT and ERA5 are small and distributed relatively close to each other in SWZ, with peaks in April-May and October; ERA5 errors are smallest in DWZ, with a peak in July and a trough in December-January of the following year. (5) The distribution characteristics of the multi-year average wind speed for 10 m show that the wind speed increases from north to south and from west to east, and the wind speed gradient is large in SWZ. [Conclusion] Overall, the 10 m sea surface wind field data set of ERA5 performs better in the study area, and the deficiency of its systematic low mean wind speed can be corrected by DASCAT.

Keywords

• sea surface wind field
• offshore wind farm
• dascat
• era5
• fnl
• applicability research