Rapid machine learning-based diagnostic analysis for high-energy-density experiments on high repetition rate laser systems

Abstract

High intensity, high-repetition rate (HRR) lasers, that is lasers that can operate on the order of 1 Hz or faster, are quickly coming on-line around the world. High intensity lasers have long been an impactful tool in high energy density (HED) science since they are capable of creating matter at extreme temperatures and pressures relevant to this field. The advent of HRR technology enhances to this capability since HRR enables these types of these experiments to be performed faster, thus leading to an acceleration in the rate of learning in fundamental HED science. However, in order to use the full potential of HRR systems, high repetition rate diagnostics in addition to real-time analysis tools must be developed to process experimental measurements and outputs at a rate that matches the laser. Towards this goal, we present an automated machine learning based analysis for a synthetic X-ray spectrometer, which is a common diagnostic in HED experiments.