Developing Digital Twins for Weather, Climate, and Energy
, Senior AI Developer Technology Engineer, NVIDIA
, Senior AI Developer Technologist, AI-HPC, NVIDIA
高度評價
Digital twins of the Earth are critical to understand climate change and its impacts at local scales, and to explore the consequences of human actions and climate change mitigation strategies in real time. High-fidelity digital twins will allow climate scientists, researchers, and non-expert users to unfold the wealth of information in climate simulations and AI model predictions, facilitating planning and decision-making in climate monitoring, modeling, mitigation, and adaptation. We'll elaborate on using Omniverse toward building digital twins and Modulus for developing the physics-based AI models that drive them. We'll focus on two case studies that showcase preliminary steps toward building digital twins: • Emulating extreme weather phenomena using the Fourier neural operator deep learning model; and • Super-resolving idealized wind turbine flows using a physics-informed deep learning super-resolution model.