Three threads, one problem: disruptive plasma instabilities
Fusion devices fail in ways that are predictable in principle and costly in practice. I work on modeling what those failures do to the machine, on seeing them coming early enough to steer away, and on asking what all of this is worth to the future of fusion energy.
Consequences: electromagnetic loads induced in W7-X
What happens to the machine when the plasma pops.
Whenever the magnetic field inside a fusion device changes quickly, currents are induced in the conducting structures. Those currents interact with the background field and pull on the machine. Predicting the resulting forces is a prerequisite for designing future tokamaks and stellarators and safely operating the ones we have now.
02 / PhD thesisCauses: interpretable machine learning
Clarifying the density limit threshold across five tokamaks.
The density limit is one of the fundamental bounds on tokamak operating space, and for forty years it has been estimated with a scaling that does not involve the plasma edge conditions. Assembling a database across five machines showed that edge collisionality is the primary organizing parameter for the limit, and that a two-parameter dimensionless boundary predicts the threshold far better. I measured and applied feedback on that boundary in real time to avoid disruptions at DIII-D.
03 / OngoingContext: economic impact of disruptions
What plasma instabilities cost.
Disruption research is usually justified on physics grounds and evaluated on physics terms. But whether a disruption is a nuisance or a showstopper is an economic question, and answering it changes which physics problems are worth working on.