In-flight measurement of kite deformations with inertial sensors

The control and design of kites for airborne wind energy requires an extensive understanding of kite dynamics. While these have been thoroughly analyzed with fluid-structure interaction models, very little experimental data exists to validate them. Specifically, most experiments on kites so far have not obtained local measurements that can be compared to these models. Measuring deformations of a kite during flight can provide a basis to validate computational models and additional insights on its dynamics. In this study, a kite shape measurement method using inertial measurement units is developed. In parallel, a stereo-photogrammetry technique is implemented in the software Blender and used to validate the new method.The inertial sensors are placed along the leading edge of a leading-edge inflatable kite to obtain the orientations of its tubular elements. The misalignment between the sensors and the leading edge elements is determined using a short photogrammetric measurement of the kite shape. With the corrected orientations, the shape of the kite is reconstituted using an algorithm that uses a reference shape of the leading edge and a basic structural model. Both measurement methods are used on a small-scale setup with a kite-surfing kite. Photogrammetry and the inertial shape measurement are compared and show a moderate agreement, but the two measurements could not properly aligned, making the comparison unreliable. Nonetheless, the inertial measurement is consistent with the turn rate, according to theory, and the new method is considered partially validated.The inertial shape measurement is used to establish a link between the tip deflection and three flight parameters: the apparent wind speed, the turn rate, and the roll angle of the kite. The data is fitted to a function with physically relevant parameters that describe the sensitivity of the deflection to the different variables. The fitted function accurately follows the main trend of the data, but many outlier points cause a significant error, due to the poor quality of the measured data. This analysis still shows a possibility of empirically developing a simple description of kite deformations from shape measurements.