Automated experimental design of safe rampdowns via probabilistic machine learning

Viraj Mehta, Jayson Barr, Joseph Abbate, Mark D. Boyer, Ian Char, Willie Neiswanger, Egemen Kolemen, Jeff Schneider

Research output: Contribution to journalArticlepeer-review


Typically the rampdown phase of a shot consists of a decrease in current and injected power and optionally a change in shape, but there is considerable flexibility in the rate, sequencing, and duration of these changes. On the next generation of tokamaks it is essential that this is done safely as the device could be damaged by the stored thermal and electromagnetic energy present in the plasma. This works presents a procedure for automatically choosing experimental rampdown designs to rapidly converge to an effective rampdown trajectory. This procedure uses probabilistic machine learning methods paired with acquisition functions taken from Bayesian optimization. In a set of 2022 experiments at DIII-D, the rampdown designs produced by our method maintained plasma control down to substantially lower current and energy levels than are typically observed. The actions predicted by the model significantly improved as the model was able to explore over the course of the experimental campaign.

Original languageEnglish (US)
Article number046014
JournalNuclear Fusion
Issue number4
StatePublished - Apr 2024

All Science Journal Classification (ASJC) codes

  • Nuclear and High Energy Physics
  • Condensed Matter Physics


  • disruption
  • machine learning
  • rampdown


Dive into the research topics of 'Automated experimental design of safe rampdowns via probabilistic machine learning'. Together they form a unique fingerprint.

Cite this