IPAC2021 - List of Authors (Johnston, S.C.)

Paper	Title	Page
WEPAB244	Optimization and Machine Learning Applied to the RF Manipulations of Proton Beams in the CERN PS	3201
	A. Lasheen, H. Damerau, S.C. Johnston CERN, Meyrin, Switzerland
	The 25 ns bunch spacing in the LHC is defined by a sequence of RF manipulations in the Proton Synchrotron (PS). Multiple RF systems covering a large range of revolution harmonics (7 to 21, 42, 84, 168) allow performing RF manipulations such as beam splitting, and non-adiabatic bunch shortening. For the nominal beam sent to LHC, each bunch is split in 12 in the PS. The relative amplitude and phase settings of the RF systems need to be precisely adjusted to minimize the bunch-by-bunch variations in intensity, longitudinal emittance, and bunch shape. However, due to transient beam-loading, the ideal settings, as well as the best achievable beam quality, vary with beam intensity. Slow drifts of the hardware may also affect beam quality. In this paper, automatized optimization routines based on particle simulations with intensity effects are presented, together with the first considerations of machine learning. The optimization routines are used to assess the best achievable longitudinal beam quality expected with the PS RF systems upgrades, in the framework of the LHC Injector Upgrade project.
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB244
About •	paper received ※ 19 May 2021 paper accepted ※ 01 July 2021 issue date ※ 24 August 2021
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Page

Optimization and Machine Learning Applied to the RF Manipulations of Proton Beams in the CERN PS

3201

A. Lasheen, H. Damerau, S.C. Johnston
CERN, Meyrin, Switzerland

The 25 ns bunch spacing in the LHC is defined by a sequence of RF manipulations in the Proton Synchrotron (PS). Multiple RF systems covering a large range of revolution harmonics (7 to 21, 42, 84, 168) allow performing RF manipulations such as beam splitting, and non-adiabatic bunch shortening. For the nominal beam sent to LHC, each bunch is split in 12 in the PS. The relative amplitude and phase settings of the RF systems need to be precisely adjusted to minimize the bunch-by-bunch variations in intensity, longitudinal emittance, and bunch shape. However, due to transient beam-loading, the ideal settings, as well as the best achievable beam quality, vary with beam intensity. Slow drifts of the hardware may also affect beam quality. In this paper, automatized optimization routines based on particle simulations with intensity effects are presented, together with the first considerations of machine learning. The optimization routines are used to assess the best achievable longitudinal beam quality expected with the PS RF systems upgrades, in the framework of the LHC Injector Upgrade project.

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB244

About •

paper received ※ 19 May 2021 paper accepted ※ 01 July 2021 issue date ※ 24 August 2021

Export •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)