Shown are fusion yields in NIF experiments. Capsule types were plastic (CH), beryllium (Be), or high-density carbon (HDC). Pulses were so-called low-foot (LF), high-foot (HF), or big-foot (BF). LGF indicates low gas fill in the capsules. The record results came from combining HDC capsules with BF pulses.
(Download Image)Can machine learning improve computer models enough to ignite internal confinement fusion?
Laser-fusion researchers have turned to machine-learning techniques to seek the combinations of laser pulse characteristics and target design needed to optimize target implosions for inertial confinement fusion. Machine learning has tripled the fusion yield of the simpler direct-drive approach being studied with the OMEGA laser at the University of Rochester. In principle, the Rochester technique could increase fusion yields with LLNL's National Ignition Facility, and an LLNL team is using machine learning to improve their experimental results. Read more at Laser Focus World.
