风电场测风数据插补的不确定度研究

Study on uncertainty of wind data interpolation of wind farm

在开展风资源实测的过程中,对于缺测和无效的测风数据需要采取插补方法提高数据的完整率和测风时长。文章利用我国不同区域、不同地形的1137座测风塔收集的满足规范要求的测风时长和数据完整率的超大规模测风数据,对当前风电行业中插补测风数据常用的线性最小二乘法、总最小二乘法和方差比率法的插补准确性进行对比分析,研究了测风时长和决定性系数对测风数据插补带来的不确定度的影响。研究结果表明,测风数据插补带来的不确定度会随着测风时长和相关性的增加而减小,同时,MERRA2数据插补带来的不确定度远大于异塔插补带来的不确定度,工程中应优先选择异塔插补来提高目标数据的质量。

During the actual measurement of wind resources,the wind interpolation method is quite important for missing and invalid wind measurement data to increase the data integrity rate and wind measurement duration.The ultra-large-scale wind measurement data were collected by a total of 1137 wind measurement towers in different regions and different terrains in China,which meet the requirements of the specifications and the wind measurement time and data integrity rate.The interpolation accuracy of the three wind interpolation methods,that is the linear least squares method,total least squares method and the variance ratio method square method,were compared and analyzed,and the influence of the wind measurement time and the determining coefficient on the uncertainty of the wind measurement data interpolation was studied.The research results show that the uncertainty of wind measurement data interpolation will decrease with the increase of wind measurement time and correlation.At the same time,the uncertainty caused by MERRA2 data interpolation is much greater than the uncertainty caused by interpolation of different towers.In the engineering,the different tower interpolation should be selected first to improve the quality of the target data.