Application of Machine Learning in Longitudinal Phase Space Prediction at the European XFEL

Abstract

Beam longitudinal phase space (LPS) distribution is the crucial property to driving the high-brightness free-electron laser. However, the beam LPS diagnostics is often de- structive and the relevant physical simulation is too time- consuming to be involved in the control room. Therefore, we explored applying the machine learning models to facilitate the virtual diagnostic of the LPS distribution at European XFEL. Two different model designs are proposed and the performance demonstrates its feasibility based on the sim- ulations. This work lays the further investigation of the real-time virtual diagnostics in the operational machine.

Zihan Zhu et al., Application of Machine Learning in Longitudinal Phase Space Prediction at the European XFEL, Proceedings of the 40th International Free Electron Laser Conference (FEL2022)