UAV Icing: Experimental Validation Data for Predicting ice Shapes at Low Reynolds Numbers

Icing is a severe hazard to aircraft and in particular to unmanned aerial vehicles (UAVs). One important activity to understand icing risks is the prediction of ice shapes with simulation tools. Nowadays, several icing computational fluid dynamic (CFD) models exist. Most of these methods have been originally developed for manned aircraft purposes at relatively high Reynolds numbers. In contrast, typical UAV applications experience Reynolds numbers an order of magnitude lower, due to the smaller airframe size and lower airspeeds. This work proposes a set of experimental ice shapes that can serve as validation data for ice prediction methods at low Reynolds numbers. Three ice shapes have been collected at different temperatures during an experimental icing wind tunnel campaign. The obtained ice shapes represent wet (glaze ice, −2 °C), mixed (−4 °C), and dry (rime ice, −10 °C) ice growth regimes. The Reynolds number is between Re=5.6…6.0×105, depending on the temperature. The ice shapes were digitized with structure-from-motion, a photogrammetric method that builds 3D models from 2D image sequences. In addition, ice weight measurements and ice density approximations are available. This validation dataset is used in the 2nd AIAA Ice Prediction Workshop (IPW) as a base case scenario. The IPW is a recurring activity that aims to compare different 3D icing CFD methods about their ability to predict ice shapes. Overall, this work is adding a much-needed validation case for low Reynolds number icing, which will aid in the verification and development of ice prediction models.