| Article | |
| Title | Machine learning based crystal collimator alignment optimization |
| Author(s) | Ricci, Gianmarco (CERN) ; D'Andrea, Marco (CERN) ; Di Castro, Mario (CERN) ; Matheson, Eloise (CERN) ; Mirarchi, Daniele (CERN) ; Mostacci, Andrea (CERN) ; Redaelli, Stefano (CERN) |
| Publication | 2024 |
| Number of pages | 8 |
| In: | Phys. Rev. Accel. Beams 27 (2024) 093001 |
| DOI | 10.1103/PhysRevAccelBeams.27.093001 (publication) |
| Subject category | Accelerators and Storage Rings |
| Abstract | In the CERN Large Hadron Collider (LHC), bent crystals are used to efficiently deflect beam halo particles toward secondary collimators used as absorbers. In this crystal collimation scheme, a crystal with a length of a few millimeters can produce a deflection equivalent to a magnetic field of hundreds of Tesla at LHC top energies, improving the cleaning performance of the machine. However, crystals must be in optimal alignment with respect to the circulating beam to maximize the efficiency of the channeling process. A newly developed machine learning model automatically classifies the channeling condition of crystals using beam loss monitor signals during slow rotation of the crystal. This advancement represents a crucial step toward refining the process of identifying the optimal channeling orientation. The algorithm has been tested for the fist time in operation with Pb ion beams at the record energy of 6.8 Z TeV demonstrating its reliability. |
| Copyright/License | publication: © 2024 authors (License: CC BY 4.0) |
Corresponding record in: Inspire
Record created 2024-10-08, last modified 2024-10-08
