Optimization of Laser Pulse for Double-cone Ignition via Machine Learning

Abstract

Double-cone ignition (DCI) scheme is a promising fast-ignition approach for the realization of inertial fusion energy. The optimization of laser pulse plays a key role for the fuel compression required by the high gain fusion burning. In this paper, we employ the genetic algorithm in machine learning to optimize the laser pulse in couple with one-dimensional hydrodynamic simulations for the fuel compression. It is found that quantum degeneracy is beneficial for the achievement of high density compression. Analysis from the dataset indicates that the areal density is strongly proportional to the peak density and implosion velocity. Numerical results also suggest that a peak density of 1000 g/cm 3 and an areal density of 3 g/cm 2 could be achieved with a drive energy of 300 kJ in the configuration of double-cone ignition scheme.