Supporting data - Physics-guided machine learning for wind-farm power prediction: Toward interpretability and generalizability

Supporting data for the publication: Title: Physics-Guided Machine Learning for Wind-Farm Power Prediction: Toward Interpretability and Generalizability Authors: Navid Zehtabiyan-Rezaie, Alexandros Iosifidis, Mahdi Abkar PRX Energy 2, 013009 – Published 30 March 2023 DOI: https://doi.org/10.1103/PRXEnergy.2.013009 Intro: Here, we provide datasets containing turbine-level efficiencies for seven wind farm cases, based on the layouts of operational wind farms, such as the Horns Rev 1 (HR1) offshore wind farm in Denmark and the Lillgrund offshore wind farm in Sweden. Details on the specifications of turbines, farms, and the characteristics of the inflow in each case are provided in Table I in our PRX paper. Structures of the datasets: The datasets include efficiency values for all turbines in each case for different incoming wind directions, with a step of 2 degrees. The turbine efficiency is the ratio of its power to the extracted power if it were standing upstream. Turbine-level data for Cases 1-4 contain 28,800 data points. Case 5 adds 14,400 data points, and Cases 6 and 7 add 17,280 data points to the dataset (60,480 data points in total). The datasets include 11 different features of data for each turbine: - Turbine's ID- Turbine's x coordinate- Turbine's y coordinate- Turbine's rotor diameter- Turbine's hub height- Turbine's thrust coefficient- Turbine's power coefficient- Inflow velocity at the hub height- Inflow turbulence level at the hub height- Wind direction- Turbine's efficiency How to cite: To cite our datasets, please use the following Bibtex records:@article{PRXEnergy.2.013009, title = {Physics-Guided Machine Learning for Wind-Farm Power Prediction: Toward Interpretability and Generalizability}, author = {Zehtabiyan-Rezaie, Navid and Iosifidis, Alexandros and Abkar, Mahdi}, journal = {PRX Energy}, volume = {2}, issue = {1}, pages = {013009}, numpages = {17}, year = {2023}, month = {Mar}, publisher = {American Physical Society}, doi = {10.1103/PRXEnergy.2.013009}, url = {https://link.aps.org/doi/10.1103/PRXEnergy.2.013009}}