17–22 May 2026
C.I.D
Europe/Zurich timezone

Data-Driven Optimization of Open Loop Control Functions at the CERN Super Proton Synchrotron

WEP6016
20 May 2026, 16:00
2h
C.I.D

C.I.D

Deauville, France
Poster Presentation MC6.D13: Instrumentation: Artificial Intelligence Poster session

Speaker

Adrian Menor de Onate (European Organization for Nuclear Research)

Description

To minimize beam intensity loss during a cycle in the CERN Super Proton Synchrotron (SPS), several machine parameters must be adjusted as functions of cycle time, spanning injection, injection plateau, acceleration, and extraction plateau. Today, these functions are typically tuned manually – a cumbersome procedure that can require hours of operator effort. This paper presents the progress towards automatically tuning time-dependent parameter functions. Using Bayesian optimization (BO), we aim to minimize intensity loss throughout the cycle with intensity measurements as the primary feedback signal. We report results from applying this method to an intentionally detuned machine development beam in the SPS, as a step towards deployment on the operational fixed-target beams. The approach is generic and applicable to time-dependent parameter optimization problems in other machines.

In which format do you inted to submit your paper? LaTeX
Preprint marking on your proceeding paper I wish my paper to be marked as preprint.

Author

Adrian Menor de Onate (European Organization for Nuclear Research)

Co-authors

Michael Schenk (European Organization for Nuclear Research) Borja Rodriguez Mateos (European Organization for Nuclear Research) Tirsi Prebibaj (European Organization for Nuclear Research) Verena Kain (European Organization for Nuclear Research) Ingrid Mases (European Organization for Nuclear Research)

Presentation materials

There are no materials yet.