Satellites are subject to very strong thermal exchanges and which are often very complex to model. As opposed to previous methods, Neural Concept Shape is able to take into account thermal properties to optimise any given shape, to minimize its weight while ensuring the mechanical stability.
Formula 1 requires a very high degree of accuracy, where even the slightest improvement can be a game-changer. Neural Concept Shape is able to match these requirements, and brings real-time design interaction to the engineer.
Backward-Facing Step flows are a very common benchmark problem to assess the reliability of a software for industrial applications. Neural Concept explored such an application: the generalisation to volumetric fields predictions.
By interfacing NCS with AirShaper, our optimization library was able to explore extensively the space of 3D drone shapes, while respecting some given constraints. Based on those learnings, the software started seeing trends and improved its understanding of the application. Soon, it started making predictions on what could be an even better aerodynamic shape! The result? A more efficient drone that will fly further on the same battery charge!
Hydrofoils make everything from water skis to sailboats to giant ferries faster (much faster). Using Neural Concept Shape, these hydrofoils were optimized to reach better performances, allowing to fly faster, longer and safer.
Cycling up to 136.7 km/h, with the only power of your legs? This is the incredible achievement of the IUT Annecy team, using our favorite aero-bike, optimized using Neural Concept Shape.