BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal//recurring_events_ical//2.0//EN
BEGIN:VEVENT
UID:7a994930-12b3-4dbd-ad66-5ef53437af7d@support.access-ci.org
DTSTAMP:20260324T115615Z
DTSTART:20260407T150000Z
DTEND:20260407T160000Z
SUMMARY:Building a Validated Digital Twin for Fusion
DESCRIPTION:In this talk, we will describe an AI-enabled digital twin of th
 e DIII-D National Fusion Facility that transforms fusion energy research b
 y reducing plasma simulation times from weeks to seconds. This effort is a
  collaboration between NVIDIA, General Atomics, and research partners incl
 uding San Diego Supercomputer Center, Argonne Leadership Computing Facilit
 y, and NERSC. The project addresses fusion's critical challenge of control
 ling plasma at hundreds of millions of degrees while predicting its behavi
 or rapidly enough to maintain reactor stability. Built within NVIDIA Omniv
 erse and powered by GPU-accelerated AI surrogate models—including EFIT f
 or plasma equilibrium, CAKE for plasma boundary prediction, and ION ORB fo
 r ion heat density—the digital twin integrates real-time sensor data fro
 m the physical reactor with physics-based simulations and engineering mode
 ls. Trained using the Polaris and Perlmutter supercomputers on decades of 
 experimental data, this interactive virtual environment enables 700 scient
 ists from over 100 organizations to conduct risk-free "what-if" scenario t
 esting and rapidly iterate reactor designs without endangering physical in
 frastructure. This "fusion accelerator" represents a paradigm shift from p
 urely physics-driven experimentation to AI-augmented computational discove
 ry, significantly accelerating the path toward commercial fusion energy.Ev
 ent details and registration HERE. See a complete list of SDSC's upcoming
  training and events here.
URL:https://support.access-ci.org/events/9024
END:VEVENT
END:VCALENDAR