Development of Deep-Learning-Based Automated Analysis for Diagnostics for High-Repetition Rate Laser Driven Acceleration Experiments

 

Advancements in high-repetition rate (HRR) laser systems, that is lasers that can fire at rates greater than ~1 Hz, as well as machine learning tools provide novel opportunities to accelerate the rate of discovery on short-pulse laser systems relevant for laser driven particle acceleration and high-energy density science more broadly. However, to realize the full potential of these HRR systems, HRR diagnostics and diagnostic analysis must be developed to measure and analyze the resulting particle sources at a rate that matches the laser. Towards this goal, we present a deep-learning based framework for real-time analysis of a differential filter-based x-ray spectrometer that is common on short-pulse laser experiments. The analysis framework was trained with a large repository of synthetic data to retrieve key experimental metrics, such as slope temperature. With traditional analysis methods, these quantities would have to be extracted from data using a time-intensive and manual analysis. This framework was developed for this specific diagnostic but is applicable to a wide variety of diagnostics common to laser experiments and thus will be especially important for the development of high-repetition rate diagnostics for HRR laser systems that are coming online.