Speaker
Description
Beam commissioning of slowly-extracted beams from the CERN Super Proton Synchrotron (SPS) to the North Experimental Area targets requires trajectory control through multiple transfer lines using corrector magnets. This is a process that traditionally demands significant expert intervention. Previous work demonstrated reinforcement learning (RL) for automated trajectory correction based on secondary emission monitor (SEM) split-foil intensity measurements, successfully centering the beam on target under nominal conditions. However, this approach fails when the beam lies outside the SEM aperture or is lost along the line.
We present an extended automation framework that allows: automated beam threading when the trajectory is outside the SEM acceptance, and online identification of corrector polarity configurations. The threading algorithm employs a random search combined with Bayesian optimisation to center the beam in the SEMs before using RL, while the automated polarity determination resolves sign ambiguities in the correctors, removing a common source of commissioning delays when using RL.
| In which format do you inted to submit your paper? | LaTeX |
|---|