Combined multifractal analysis of wind power production and atmospheric fields using simultaneous measurement of high-resolution data

International audienceAtmospheric fields are known to exhibit extreme variability over wide range of temporal and spatial scales, which makes them complex to characterize. When it comes to wind power production, the power available at atmosphere and power extracted by turbines at multiple scales are affected by corresponding variations in coexisting fields. Understanding their variability and correlations helps in quantifying uncertainties in modeling as well as real data analysis. Here, we aim to characterize the variability and correlations across scales of wind power production, and atmospheric fields including 3D wind, rainfall and air density using simultaneous measurements in a wind farm relying on the framework of Universal Multifractal (UM) analysis. It is a widely used, physically based, scale invariant framework for characterizing and simulating geophysical fields over wide range of scales.Towards this, high-resolution atmospheric data collected from a meteorological mast located in the wind farm of Pays d'Othe operated by Boralex (110 km south-east of Paris, France) is used. The data is being collected under the project RW-Turb (https://hmco.enpc.fr/portfolio-archive/rw-turb/; supported by the French National Research Agency (ANR-19-CE05-0022). The campaign utilizes multiple 3D sonic anemometers (manufactured by Thies), mini meteorological stations (manufactured by Thies), and disdrometers (Parsivel2, manufactured by OTT) installed at turbine hub height along with turbines in the wind farm. The temporal resolution is 100 Hz for the 3D sonic anemometers, 1 Hz for the meteorological stations and 30 s for the disdrometers. Variability in power production is examined according to different meteorological conditions using the framework of UM and consequences of their correlations are discussed. In the process we also make short commentary on the actual sampling resolution at which fields should be considered for extracting useful statistical information about their variability