Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

MACHINE LEARNING METHODS FOR SINGLE SHOT RF TUNING

Lundquist, J. S. LU ; Werin, S. LU and Milas, N. (2021) 10th International Beam Instrumentation Conference, IBIC 2021 In CERN-Proceedings p.313-316
Abstract

The European Spallation Source, currently under construction in Lund, Sweden, will be the world’s most powerful neutron source. It is driven by a proton linac with a current of 62.5 mA, 2.86 ms long pulses at 14 Hz. The final section of its normal-conducting front-end consists of a 39 m long drift tube linac (DTL) divided into five tanks, designed to accelerate the proton beam from 3.6 MeV to 90 MeV. The high beam current and power impose challenges to the design and tuning of the machine and the RF amplitude and phase have to be set within 1% and 1◦ of the design values. The usual method used to define the RF set-point is signature matching, which can be a challenging process, and new techniques to meet the growing... (More)

The European Spallation Source, currently under construction in Lund, Sweden, will be the world’s most powerful neutron source. It is driven by a proton linac with a current of 62.5 mA, 2.86 ms long pulses at 14 Hz. The final section of its normal-conducting front-end consists of a 39 m long drift tube linac (DTL) divided into five tanks, designed to accelerate the proton beam from 3.6 MeV to 90 MeV. The high beam current and power impose challenges to the design and tuning of the machine and the RF amplitude and phase have to be set within 1% and 1◦ of the design values. The usual method used to define the RF set-point is signature matching, which can be a challenging process, and new techniques to meet the growing complexity of accelerator facilities are highly desirable. In this paper we study the use of ML to determine the RF optimum amplitude and phase, using a single pass of the beam through the ESS DTL1 tank. This novel method is compared with the more established methods using scans over RF phase, providing similar results in terms of accuracy for simulated data with errors. We also discuss the results and future extension of the method to the whole ESS DTL.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
10th International Beam Instrumentation Conference, IBIC 2021 - Proceedings
series title
CERN-Proceedings
editor
Kim, Changbum ; Kim, Dong-Eon ; Lee, Jaeyu and Schaa, Volker RW
pages
4 pages
publisher
CERN
conference name
10th International Beam Instrumentation Conference, IBIC 2021
conference location
Virtual, Online, Korea, Republic of
conference dates
2021-09-13 - 2021-09-17
external identifiers
  • scopus:85130591076
ISSN
2078-8835
ISBN
9783954502301
DOI
10.18429/JACoW-IBIC2021-TUPP41
language
English
LU publication?
yes
id
2f50f45d-ad0d-400a-a6af-dc544a229eef
date added to LUP
2022-07-08 15:18:37
date last changed
2022-07-11 14:49:24
@inproceedings{2f50f45d-ad0d-400a-a6af-dc544a229eef,
  abstract     = {{<p>The European Spallation Source, currently under construction in Lund, Sweden, will be the world’s most powerful neutron source. It is driven by a proton linac with a current of 62.5 mA, 2.86 ms long pulses at 14 Hz. The final section of its normal-conducting front-end consists of a 39 m long drift tube linac (DTL) divided into five tanks, designed to accelerate the proton beam from 3.6 MeV to 90 MeV. The high beam current and power impose challenges to the design and tuning of the machine and the RF amplitude and phase have to be set within 1% and 1<sup>◦</sup> of the design values. The usual method used to define the RF set-point is signature matching, which can be a challenging process, and new techniques to meet the growing complexity of accelerator facilities are highly desirable. In this paper we study the use of ML to determine the RF optimum amplitude and phase, using a single pass of the beam through the ESS DTL1 tank. This novel method is compared with the more established methods using scans over RF phase, providing similar results in terms of accuracy for simulated data with errors. We also discuss the results and future extension of the method to the whole ESS DTL.</p>}},
  author       = {{Lundquist, J. S. and Werin, S. and Milas, N.}},
  booktitle    = {{10th International Beam Instrumentation Conference, IBIC 2021 - Proceedings}},
  editor       = {{Kim, Changbum and Kim, Dong-Eon and Lee, Jaeyu and Schaa, Volker RW}},
  isbn         = {{9783954502301}},
  issn         = {{2078-8835}},
  language     = {{eng}},
  pages        = {{313--316}},
  publisher    = {{CERN}},
  series       = {{CERN-Proceedings}},
  title        = {{MACHINE LEARNING METHODS FOR SINGLE SHOT RF TUNING}},
  url          = {{http://dx.doi.org/10.18429/JACoW-IBIC2021-TUPP41}},
  doi          = {{10.18429/JACoW-IBIC2021-TUPP41}},
  year         = {{2021}},
}