Title: Fusion plasma turbulence simulation with neural network surrogate models
Speaker: Jonathan Citrin (DIFFER)
Time: Dec. 19, 2019, 10:00–11:00
Location: Differ, Alexander-zaal
Plasma energy losses due to turbulent transport is one of the limiting factors for achieving viable fusion energy. Reactor design and plasma scenario optimisation demands both accurate and tractable predictive turbulence calculations.
While high-fidelity direct numerical calculations are available, and agree ever more routinely with experimental observations, these codes are intractable for use in resolving plasma evolution on discharge timescales. We describe a combination of reduced order turbulence modelling and neural network surrogates, which provide a pathway to provide fast and accurate fusion plasma turbulence modelling, bridging 12 orders of magnitude in calculation speed. These developments open up a plethora new possibilities in fusion science for first-principle-based scenario optimization, control-oriented applications, and uncertainty quantification.