Speaker
Description
This study proposes a machine learning approach to analyze the correlation between beam position monitor (BPM) measurements and output laser power in the Hefei Infrared Free-Electron Laser (FEL) facility. Using bunch-by-bunch data of transverse position, charge, and longitudinal phase collected from upstream undulator BPM probes, we develop a predictive model to evaluate whether BPM measurements can effectively forecast the infrared FEL's laser power output.
If the model demonstrates significant predictive capability, we will decompose the network to identify the most influential bunches or parameters, providing targeted optimization strategies for beam tuning experiments. This data-driven approach reduces reliance on empirical tuning methods and improves accelerator operational stability.
| I have read and accept the Conference Policies | Yes |
|---|
