Spatially distributed Wind and Turbulence Measurements with a Fleet of Unmanned Aerial Systems

Abstract

This thesis deals with the development of a unique measuring device for wind field measurement in the atmospheric boundary layer and its application to examine spatial turbulence structures in heterogeneous terrain as well as flow measurements around a wind turbine. The innovative measuring system consists of a fleet of 35 quadrotors UAS (unmanned aerial systems), of which a maximum of 20 were used simultaneously. This measuring system enables flexible, simultaneous, spatially distributed measurements of the wind vector in the boundary layer. An algorithm was developed to measure the wind that is based on the position and acceleration sensors of the UAS and does not require additional external wind sensors. The algorithm puts the sensor data in relation to the acting wind forces and is calibrated and validated with the help of reference measurements on a 99-m meteorological mast. The potential of the UAS fleet for wind field and turbulence measurements is shown by comparisons with Doppler wind lidar and ultrasonic anemometer measurement data. Furthermore, a special flight pattern with spatially horizontally distributed measurements was developed to allow for the examination of horizontal turbulence structures. On the one hand, the limit of validity of the Taylor hypothesis of frozen turbulence is tested. On the other hand, it is demonstrated how turbulence structures differ in their horizontal spatial characteristics depending on the atmospheric conditions. Additionally, the correlation of different scales in the frequency domain is examined using coherence. In comparison to models of the decay of coherence, the validity of the models is limited to neutral stratification. Overall, the coherence is smaller for the lateral separation distance than for the longitudinal one. In a final measurement campaign, the knowledge gained and an improved wind algorithm were used to analyze the flow around a wind turbine (WT). At the same time, measurements were carried out in the wake and in the inflow of the WT. Spatially distributed measurements in the near wake of a 2 MW WT clearly show the expected wind speed deficit. Laterally distributed measurements in the wake under stable and near-neutral stratification indicate a double Gaussian distribution of the lateral velocity profile. Under convective conditions, the turbulent mixing is enhanced, which leads to a measurement of a simple Gaussian distribution already in the near wake. Furthermore, horizontal turbulent flow measurements show the expected energy input from outside the wake into the edge areas of the wake. In addition, it could be shown that a turbulent flow from the center of the wake to the edge areas can also be measured in stable and near-neutral stratification. Also, the occurrence of vortices resulting from the pressure differences at the rotor blade tips was investigated.