MERRA 재해석 자료를 이용한 풍력 발전량 예측 성능평가

Abstract

Using MERRA (Modern Era Retrospective analysis for Research and Application) reanalysis data, wind resource assessment has been performed when there is no wind data measured on met mast. However there are not many studies about reliability of MERRA reanalysis data in South Korea. The investigation was carried out for evaluating how much MERRA reanalysis data is reliable for wind resource assessment. The evaluation result was compared with that obtained using wind data measured on met mast. The real annual energy production (AEP) was obtained from Hankyung wind farm with 21 MW capacity of wind turbines, which was used as a reference wind farm. The measurement period was all from January 13, 2010 to January 12, 2011. The collecting points for MERRA reanalysis data and met mast data were 24 km and 14.1 km away from Hankyung wind farm, respectively. The wind farm design software, WindPRO, was used for AEP prediction. The results show that using MERRA reanalysis data caused -16.25 % error for AEP, and -17.40 % error for Capacity Factor. On the other hand, the prediction using met mast data resulted in -7.19 % error for AEP, and -8.42 % for Capacity Factor. Based on this result, it is not recommended to use MERRA reanalysis data whose collecting point is as far away as this investigation for wind resource assessment. However it could bring more accurate AEP prediction to use MERRA reanalysis data from collecting point closer to a given site.