Editor’s note: this article was first published in 2022. Since then, Neural Concept has evolved beyond physics prediction into a platform that brings together the tools, infrastructure and skills AI agents need to understand and act on engineering processes. Engineers can then design better-performing physical products, faster, through AI-native engineering.
AI-based predictive tools give designers near real-time feedback on how their designs perform. The question is: where does the data used to train them come from?
The neural network is fed by both CAD geometries and CAE simulations in order to create predictions for product designers. Since each team provides respectively these raw data without the need to transform them, the tool becomes more than the sum of its parts.
In Neural Concept’s Engineering Intelligence Platform, Geometry Intelligence reads CAD as it is and Physics Intelligence learns from CAE data, alongside the solvers that produce it. CAD designs can be STL/IGES or native formats, while CAE data can come from simulations with arbitrary levels of sophistication; such as multiphase, transient etc. Bringing realistic CAD and CAE representations together this way is a step forward from traditional methods. Indeed, reduced-order modelling traditionally requires to parametrize CAD geometries, while CAD-embedded simulations usually requires downgrading in the CAE fidelity and introduces uncertainties.
This Production App exemplifies the powerful convergence from realistic CAD and CAE data. The example captures a workflow for an automotive HVAC’s fluids and thermal performance. In this design scenario, the inputs to the AI-tool provided by product designers are 1) raw CAD geometries and 2) operating parameters (e.g. inlet air mass flow rate and temperature). The expected outputs are air pressure, velocity and temperature, available in 3 seconds after importing a CAD.
The predictive tool is much more than “easy-to-use software for designers”. Its usage is immediate in a design environment, because it encapsulates a workflow commissioned by the final users and tailored to their personalized needs.
