This dataset comprises the average and standard deviation of wind speed, collected from 120 turbines in an inland wind farm, for the years of 2009 and 2010. Missing data in the original dataset are imputed by using the iterative singular value decomposition. Two data files are associated with each year---one contains the hourly average wind speed, and the other contains the hourly standard deviation of wind speed. The naming convention makes it clear which year a file is associated with and whether it is for the average speed (Ave) or for the standard deviation (Stdev). The data arrangement in these four files is as follows---the columns are the 120 turbines and the rows are times, starting from 12 a.m. on January 1 of a respective year as the first data row, followed by the subsequent hours in that year. The fifth file in this dataset contains the coordinates of the 120 turbines. To protect the wind farm's identity, the coordinates have been transformed by an undisclosed mapping, so that their absolute values are no longer meaningful but the turbine-to-turbine relative distances are maintained. This dataset is used in Chapters 3 and 4 of the Data Science for Wind Energy book.

{"references": ["Ding, Y. (2019) Data Science for Wind Energy, Chapman & Hall/CRC Press, Boca Raton, FL"]}