This dataset includes three sets, one corresponding to an actual vortex generator installation and two corresponding to an artificial pitch angle adjustment. Two pairs of wind turbines from the same inland wind farm, as used in Chapter 5 of the Data Science for Wind Energy book, are chosen to provide the data, each pair consisting of two wind turbines, together with a nearby met mast. The turbine that undergoes an upgrade in a pair is referred to as the experimental turbine, the reference turbine, or the test turbine, whereas the one that does not have the upgrade is referred to as the control turbine. In both pairs, the test turbine and the control turbine are practically identical and were put into service at the same time. This wind farm is on a reasonably flat terrain. The power output, y, is measured on individual turbines, whereas the environmental variables in x (i.e., the weather covariates) are measured by sensors at the nearby mast. For this dataset, there are five variables in x and they are the same as those in the Inland Wind Farm Dataset1. For the vortex generator installation pair, there are 14 months' worth of data in the period before the upgrade and around eight weeks of data after the upgrade. For the pitch angle adjustment pair, there are about eight months of data before the upgrade and eight and a half weeks after the upgrade. Note that the pitch angle adjustment is not physically carried out, but rather simulated on the respective test turbine. The following data modification is done to the test turbine data. The actual test turbine data, including both power production data and environmental measurements, are taken from the actual turbine pair operation. Then, the power production from the designated test turbine on the range of wind speed over 9 m/s is increased by 5%, namely multiplied by a factor of 1.05, while all other variables are kept the same. No data modification of any kind is done to the data affiliated with the control turbine in the pitch angle adjustment pair. The third column of a respective dataset is the upgrade status variable, of which a zero means the test turbine is not modified yet, while a one means that the test turbine is modified. The upgrade status has no impact on the control turbine, as the control turbine remains unmodified throughout. The vortex generator installation takes effect on June 20, 2011, and the pitch angle adjustment takes effect on April 25, 2011.

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