Editing Princeton Plasma Physics Laboratory (section)

===Later designs===
In 1999, the [[National Spherical Torus Experiment]] (NSTX), based on the spherical tokamak concept, came online at the PPPL. 

Odd-parity heating was demonstrated in the 4&nbsp;cm radius PFRC-1 experiment in 2006. PFRC-2 has a plasma radius of 8&nbsp;cm. Studies of electron heating in PFRC-2 reached 500&nbsp;[[electronvolt|eV]] with pulse lengths of 300&nbsp;ms.<ref name=":0">{{Cite web |url=https://www.nextbigfuture.com/2019/06/game-changing-direct-drive-fusion-propulsion-progress.html |title=Game Changing Direct Drive Fusion Propulsion Progress |last=Wang |first=Brian |date=June 22, 2019 |website=NextBigFuture |language=en-US |access-date=2019-06-22}}</ref>

In 2015, PPPL completed an upgrade to NSTX to produce NSTX-U that made it the most powerful experimental fusion facility, or tokamak, of its type in the world.<ref>{{Cite web |url=https://www.pppl.gov/nstx |title=National Spherical Torus Experiment Upgrade (NSTX-U) |website=Princeton Plasma Physics Lab}}</ref>

In 2017, the group received a Phase II NIAC grant along with two NASA STTRs funding the RF subsystem and superconducting coil subsystem.<ref name=":0" />

In 2024, the lab announced MUSE, a new [[stellarator]]. MUSE uses rare-earth permanent magnets with a field strength that can exceed 1.2 [[Tesla (unit)|teslas]]. The device uses quasiaxisymmetry, a subtype of [[quasisymmetry]]. The research team claimed that its use of quasisymmetry was more sophisticated than prior devices.<ref>{{Cite web |last=Paul |first=Andrew |date=2024-04-05 |title=Stellarator fusion reactor gets new life thanks to a creative magnet workaround |url=https://www.popsci.com/environment/stellarator-fusion-reactor/ |access-date=2024-04-11 |website=Popular Science |language=en-US}}</ref> Also in 2024, PPL announced a [[reinforcement learning]] model that could forecast tearing mode instabilities up to 300 milliseconds in advance. That is enough time for the plasma controller to adjust operating parameters to prevent the tear and maintain [[High-confinement mode|H-mode]] performance.<ref>{{Cite web |date=March 4, 2024 |title=AI can predict and prevent fusion plasma instabilities in milliseconds |url=https://www.ans.org/news/article-5835/ai-can-predict-and-prevent-fusion-plasma-instabilities-in-milliseconds/ |access-date=2024-05-20 |website=www.ans.org |language=en}}</ref><ref>{{Cite journal |last=Seo |first=Jaemin |last2=Kim |first2=SangKyeun |last3=Jalalvand |first3=Azarakhsh |last4=Conlin |first4=Rory |last5=Rothstein |first5=Andrew |last6=Abbate |first6=Joseph |last7=Erickson |first7=Keith |last8=Wai |first8=Josiah |last9=Shousha |first9=Ricardo |last10=Kolemen |first10=Egemen |date=2024 |title=Avoiding fusion plasma tearing instability with deep reinforcement learning |url=https://www.nature.com/articles/s41586-024-07024-9 |journal=Nature |language=en |volume=626 |issue=8000 |pages=746–751 |doi=10.1038/s41586-024-07024-9 |issn=1476-4687|pmc=10881383 }}</ref>