AirShaper

Very happy to see the results in the final instalment of Daniel Riley's (YouTube: rctestflight) propeller competition! 🎉✨ In the last round, we had submitted an optimized toroidal propeller, which significantly improved on the baseline design, as well as our manually designed initial toroidal propeller, but could not compete with the more efficient long and high aspect ratio bladed propellers of other submissions. So once more, we joined forces with Wouter Remmerie 🏁 from AirShaper and used CFD and machine learning to generate an optimized conventional propeller design. It performed great, to say the least, but see for yourself in the video below! As a bonus, both of our submissions performed extremely well in terms of generated noise! PS: Get more detailed insights about the design process for these propeller at the CAESES user conference (https://lnkd.in/dVQPDKyg) in September 2024.

+21% propeller efficiency through AI! Read more 👇 or watch ▶ https://lnkd.in/eNGtz3TU AirShaper and CAESES made an AI-generated propeller! Here's how it works: 1. 3D model generatoin CAESES develops software to automatically change the shape of parametric 3D models. 2. CFD Analysis AirShaper then analyse the hydrodynamics of every shape generated by CAESES and sends back the flow field & performance data. 3. Training the surrogate model CAESES builds a surrogate model based on the findings of all these simulations 4. Predicting the best possible propeller Then, using the lightweight surrogate model, CAESES predicts the best possible propeller design. This approach works for small & large propellers, and even non-propeller applications! Drop a comment if you're curious to learn more :) A massive thanks to Simon Hauschulz, Heinrich von Zadow and Claus Abt of CAESES to make this possible, and Daniel Riley of RCTESTFLIGHT!

How do run a #CFD #simulation on a #drone? Check this 14-minute tutorial ▶ https://lnkd.in/ePHZaGjW In this video, we used a public 3D model of a DJI Phantom, modified it a little, and uploaded it to AirShaper for a CFD analysis. Curious to find out what you can learn from this? Then check out the video and get in touch if you want to run your own simulation!

In our previous posts, we often mention #simulations made in our in-house simulation tool. This time, Tomáš Hájek, who heads the analysis and testing team at Mejzlik Propellers, in his latest publication took #windtunnel test data for one of our existing propellers, the 21x13 CCW 2B E L, and compared it to our in-house simulation software and AirShaper, the online #aerodynamics platform. Our in-house tool combines Lifting Line Theory, Blade Element Momentum (BEM) Theory, and Vortex Theory, while AirShaper applies full blown CFD (Computational Fluid Dynamics) simulations. The full article is available at AirShaper https://lnkd.in/eZy-T5xg These results show a near-exact agreement with the wind tunnel data, especially for the CFD simulations, significantly enhancing our propeller and #UAV design capabilities. The ability to apply CFD methods to full UAVs allows us to optimize propeller-airframe interactions and offer comprehensive aerodynamic design advice.

#windtunnel data versus #aerodynamic #simulation - Check our Mejzlik Propellers validation! 💡 https://lnkd.in/geC-_KCu Predicting the performance of a #propeller is not easy - tip vortices, boundary layer separation, ... all play their part. So Tomáš Hájek, head of the analysis & testing team at Mejzlik Propellers, benchmarked AirShaper against both their in-house simulation tools and wind tunnel data. The results? A near exact match for a large part of the operating regime! Check out this article to learn more and a massive thanks to Tomáš Hájek and Michal Lošťák for making this happen!

How would YOU improve the winglets of this drone? Drop a comment! https://lnkd.in/e336a-yW #aerodynamics #drone #cfd #simulation Thanks a lot for this great collaboration Kavin Gustafson!