Annons

AERODYNAMICS: On Siemens’ and PhysicsX’s Ambitions to Spearhead AI-Based Deep Physics Simulation

CFD and FEA: Pre-trained MODELS built on STAR-CCM+ and NASTRAN data. Siemens' purchase of the AI, simulation and analysis company Altair gives an interesting signal on how the company values software capabilities in these areas: The future points to a continued strong increase in the needs for sharp capabilities in all three domains, with emphasis on the AI side. And Siemens' PLM division, Siemens Digital Industries Software, has the ambition to spearhead the simulation train into the future. It is even clearer today after the company's announcement last week of an ongoing collaboration with PhysicsX, a startup that uses generative AI to enable groundbreaking engineering in advanced industries such as Aerospace & Defense. Together, the companies now aim to create the next generation of AI-based deep physics simulation to accelerate performance prediction and optimization. PhysicsX builds its latest pre-trained deep physics model for aerodynamics on high-fidelity simulation data generated with the Siemens Xcelerator portfolio's Simcenter platform.
With this and with the purchase of Altair–a trailblazing pioneer in unleashing the power of AI across the entire product lifecycle–a new higher standard is being set in AI-driven CAE technology. The goal is to continue on the path initiated by Altair; to seamlessly embed AI into design and simulation tools with integrated user-friendly workflows, to together with on-demand high-performance computing (HPC) resources and Siemens PLM division's already existing rich Xcelerator portfolio on the Simcenter platform, empower teams to extend AI with perhaps the market's strongest combined PLM-integrated total solution for the CAE domain.
Add to this what PhysicsX solutions and their promise of the dramatic acceleration that AI can bring to simulation through the Large Geometry Model (LGM). For example, PhysicsX's LGM-Aero is trained on a corpus with more than 25 million geometries and associated physics simulation. The training data contains tens of billions of mesh elements and tens of thousands of Computational Fluid Dynamics (CFD) and Finite Element Analysis (FEA) simulations made with Siemens Simcenter STAR-CCM+ and Nastran software. One result is that users can fine-tune deep learning models for their own applications with as little as a few tens of simulations. "AI represents a truly transformative opportunity for the simulation community, and we are exploring its potential at an accelerating pace," said Jean-Claude Ercolanelli, senior VP of Simulation and Test Solutions, Siemens Digital Industries Software. He continues: "To scale AI use, open collaboration is key and we're excited for PhysicsX to reveal to the world what they're working on. Together, we're now exploring an AI-enhanced simulation industry that has the potential to change the way products are designed and engineered.”

Robin Tuluie, Chairman and Founder of PhysicsX comments that the release of Ai.rplane, an open access reference application, ”demonstrates the power of LGM-Aero to design innovative avionics solutions.”

Highest quality of syntetized data and robust integration
He further claims that the collaboration with Siemens is taking a step to the next level of fundamental technology to improve the way engineering is practiced:
“The foundation for successful AI deployment in engineering is based on the highest quality synthetic data, robust integrations between CAE and AI, and building on the trust customers have in our respective technologies. We are excited to build into the opportunity space with Siemens through the launch of our latest Large Geometry Model (LGM) for aerodynamics, LGM-Aero, trained on high-quality data (hi-fi data) generated with the Siemens Xcelerator portfolio tool.”

PhysicsX’s Ai.rplane is the world’s first large physics model (LPM) for flight, a fully trained model that generalizes to a wide set of aeroelastic applications. IMAGE: PhysicsX

The ai.rplane application handles complex tasks in seconds
In general about PhysicsX, it can be noted that the company’s ambition is, ”a quest to enable groundbreaking engineering across advanced industries with AI.” For this, the company is building a new type of engineering platform that will be able to take AI-based simulation and optimization through the entire product development life cycle, from design to manufacturing to operation. According to the company, the solutions have also been used by some of the most advanced companies in the semiconductor, aviation, automotive, energy and material processing segments.
As noted above, PhysicsX has thus released Ai.rplane. This is a free public reference application that offers a set of exploratory tools focused on aerodynamics and aircraft structures with unique AI shape generation. Powered by LGM-Aero, Ai.rplane demonstrates the power of a fully trained model that generalizes to a wide set of aeroelastic applications. LGM-Aero is used as a zero shot model to derive flight performance, flight stability and structural stress for all generated airfoils. The application demonstrates the creation of geometry and the assessment of physics performance in less than a second, compared to the several hours required for traditional numerical simulations.

PhysicsX has released Ai.rplane, a free-to-use technology demonstrator for the public that offers a set of exploratory tools focused on aerodynamics and aircraft structures with unique AI shape generation – built on training data created with Siemens Simcenter Nastran and STAR-CCM+ software. IMAGE: PhysicsX

Starting with a basic set of instructions, users can explore the full generative design space offered by Ai.rplane, then modify or optimize for a desired performance characteristic in seconds. Over time, PhysicsX aims to add new features, including cargo packaging, selection of commercial powertrains, and controls.

In addition, Ai.rplane was developed using a comprehensive set of simulation techniques from Siemens to automate the generation of high-quality training data. Both LGM-Aero and Ai.rplane are available on the PhysicsX AI engineering platform.

To access Ai.rplane, go to airplane.physicsx.ai.

Print Friendly, PDF & Email

Success Stories

Industriellt

Success Stories

Intressant på PLM TV News

PLM TV News

PLM TV News

PLM TV News

PLM TV News

PLM TV News

Aktuell ANALYS

Aktuell Analys

Aktuell Analys

3D-printing

Block title