Based on an FEM / CFD model and the geometry, a generative network (GAN) or a reinforcement learning agent creates solution variants, and each of them is verified by our solvers.


We describe the computational environment and the geometric model and generate many variants of shapes and actions. The solver’s assessment is fed back to the network as a reward, so the proposals improve with each iteration. Earlier results from publications, projects and research are used for supervised learning and speed up the first, random attempts.
Let’s talk to define your needs and prepare an offer.
We use necessary cookies to make the website work properly. With your consent, we may also use analytics cookies that help us improve the website. Details are in our privacy policy.