Application of reinforcement learning in the LHC tune feedback

The Beam-Based Feedback System (BBFS) was primarily responsible for correcting the beam energy, orbit and tune in CERN Large Hadron Collider (LHC). A major code renovation of BBFS planned carried out during LHC Long Shutdown 2 (LS2). This work consists an explorative study to solve a beam-based control problem, feedback (QFB), utilising state-of-the-art Reinforcement Learning (RL). simulation environment created mimic operation QFB. series RL agents were trained, best-performing then subjected set well-designed tests. original controller used QFB reimplemented compare performance classical approach selected test scenarios. Results from simulated show that agent can exceed controller-based paradigm.