Siemens Digital Industries Software has announced an ongoing collaboration with the startup PhysicsX, which is utilizing generative artificial intelligence (AI) to enhance engineering in advanced industries. This partnership focuses on developing the next generation of AI-based deep physics simulation, aimed at improving performance prediction and optimization.
PhysicsX is currently constructing its latest pretrained deep physics model for aerodynamics, leveraging high-fidelity simulation data generated with the Siemens Xcelerator portfolio. Jean-Claude Ercolanelli, senior vice president of Simulation and Test Solutions at Siemens, emphasized the transformative potential of AI in the simulation community and the importance of open collaboration in scaling AI adoption.
Robin Tuluie, founder and chairman of PhysicsX, highlighted that successful AI deployment in engineering relies on high-quality synthetic data and robust integration between computer-aided engineering (CAE) and AI technologies. The latest model, Large Geometry Model (LGM)-Aero, is trained on over 25 million geometries and associated physics simulations, utilizing advanced software such as Simcenter STAR-CCM+ and Simcenter Nastran.
Additionally, PhysicsX has introduced Ai.rplane, a free public-access reference application designed to assist in aerodynamics and aircraft structures with AI-driven shape generation. This application, powered by LGM-Aero, can infer aerodynamic performance, flight stability, and structural stress for various flying shapes in under a second.
Users of Ai.rplane can explore generative design options and optimize for desired performance characteristics quickly. Future updates to the application will include new features such as cargo packaging and commercial powertrain selection. Both LGM-Aero and Ai.rplane are available on the PhysicsX AI engineering platform.