femto-PIXAR: a neural network method for reconstruction of femtosecond X-ray free electron laser pulse energy

TL;DR

This paper introduces femto-PIXAR, a neural network approach that accurately reconstructs femtosecond X-ray free electron laser pulse energies from experimental data, enabling shot-by-shot characterization of weak pulses without simulations.

Contribution

The paper presents a physics-based U-net neural network model for reconstructing femtosecond X-FEL pulse profiles from experimental data, a novel method for weak pulse analysis.

Findings

Successfully reconstructs pulse profiles for weak X-FEL pulses

Operates without the need for simulations

Works on shot-by-shot basis for experimental data

Abstract

X-ray Free Electron Lasers (X\nobreakdash-FELs) operate in a wide range of lasing configurations for a broad variety of scientific applications at ultrafast time-scales such as structural biology, materials science, and atomic and molecular physics. Shot-by-shot characterization of the X-FEL pulses is crucial for analysis of many experiments as well as tuning the X-FEL performance. However, for the weak pulses found in advanced configurations, e.g. those needed for coherent, two-pulse studies of quantum materials, there is no current method for reliably resolving pulse profiles. Here we show that a physics-based U-net model can reconstruct the individual pulse power profiles for sub-picosecond pulse separation without the need for simulations. Using experimental data from weak X-FEL pulse pairs, we demonstrate we can learn the pulse characteristics on a shot-by-shot basis when conventional methods fail.