About this Webinar:

A Physics AI revolution is brewing right now in the engineering sciences: the availability of massive amounts of data together with recent advances in physics-based AI/ML modeling architectures and the availability of differentiable physics solvers is making this possible. This revolution will have a meaningful and disruptive impact on how aircraft and aircraft components are designed and how distributed engineering teams are organized and work. With advances in large-scale data availability, Physics AI models can be trained in specific domains with inference times around 1–3 seconds but with accuracy in the predictions of the physics and derived quantities within the 1-2% range.

Join Professor Juan Alonso as he explains how Luminary Cloud’s SHIFT Models for aerospace applications can result in efficient 1) exploration of vast design spaces, 2) interactive design, 3) inference-based design optimization, 4) real-time control of physics systems, 5) uncertainty quantification and design under uncertainty, and 6) digital twin applications.

What You'll Learn:

• Learn about recent trends in Physics AI including how key developments are expanding the use cases these models can be leveraged for.
• Discover the importance of generating massive amounts of high-fidelity data and how these data affect model accuracy for both scalar output quantities and field prediction.
• Domain-specific vs foundational models: how far can we currently push Physics AI models and how can we improve model generalizability?
• What are the most applicable use cases for this technology in aerospace engineering?