Optimizing the Performance of Direct-Drive Implosion Experiments Using Transformer Neural Processes and Meta-Look-Ahead

Finding a laser pulse shape that optimizes the Lawson parameter [1,2,3] for a given target is a challenging problem in inertial confinement fusion due to the predictive gap between simulations and experiments. The Lawson parameter is typically related to the yield and ρR of the implosion and requires an increase in both. Optimizing the yield of cryogenic implosions on OMEGA using a data-driven predictive machine-learning (ML) approach [4,5] has met with considerable success, but increasing the ρR has proven more challenging. It is likely that this is in part due to hydrodynamic instabilities, but is likely also due to the increased sensitivity of the ρR to fine details of the shock timing and entropy spatial profile of the implosion, which in turn are highly sensitive to the front end of the laser pulse. If simulations used for implosion design [6] fail to capture the instability growth, shock transit, or adiabat-setting behavior of the implosion correctly, the response surface between simulations and experiments will sharply differ, making implosion optimization challenging with limited experimental data. We present the use of meta-learning methods for ICF implosion optimization. ML optimization algorithms as such can properly leverage the high sample rate envisioned for future facilities to effectively optimize in the presence of unknown nonidealities and can provide estimates for the workflows and shot rate needed to validate novel inertial fusion energy schemes. This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award No. DE-NA0004144, Department of Energy under Award Nos. DE-SC0024381, DE-SC0022132,DE-SC0021072 and DE-SC0024456, the University of Rochester, and the New York State Energy Research and Development Authority.