Home > Publications database > Active deep learning for nonlinear optics design of a vertical FFA accelerator > print

001     358340
005     20250310164549.0
024 7 _ |a 10.18429/JACOW-IPAC2023-WEPA026
|2 doi
024 7 _ |a 10.15120/GSI-2025-00524
|2 datacite_doi
037 _ _ |a GSI-2025-00524
041 _ _ |a English
100 1 _ |a Santamaria Garcia, Andrea
|0 P:(DE-HGF)0
|b 0
111 2 _ |a 14th International Particle Accelerator Conference
|g IPAC2023
|c Venice
|d 2023-05-07 - 2023-05-12
|w Italy
245 _ _ |a Active deep learning for nonlinear optics design of a vertical FFA accelerator
260 _ _ |a Geneva, Switzerland
|c 2023
|b JACoW Publishing
300 _ _ |a 2709 - 2712
336 7 _ |a CONFERENCE_PAPER
|2 ORCID
336 7 _ |a Conference Paper
|0 33
|2 EndNote
336 7 _ |a INPROCEEDINGS
|2 BibTeX
336 7 _ |a conferenceObject
|2 DRIVER
336 7 _ |a Output Types/Conference Paper
|2 DataCite
336 7 _ |a Contribution to a conference proceedings
|b contrib
|m contrib
|0 PUB:(DE-HGF)8
|s 1741076449_1280108
|2 PUB:(DE-HGF)
500 _ _ |a Published by JACoW Publishing under the terms of the Creative Commons Attribution 4.0 license.
520 _ _ |a Vertical Fixed-Field Alternating Gradient (vFFA) accelerators exhibit particle orbits which move vertically during acceleration. This recently rediscovered circular accelerator type has several advantages over conventional ring accelerators, such as zero momentum compaction factor. At the same time, inherently non-planar orbits and a unique transverse coupling make controlling the beam dynamics a complex task. In general, betatron tune adjustment is crucial to avoid resonances, particularly when space charge effects are present. Due to highly nonlinear magnetic fields in the vFFA, it remains a challenging task to determine an optimal lattice design in terms of maximising the dynamic aperture. This contribution describes a deep learning based algorithm which strongly improves on regular grid scans and random search to find an optimal lattice: a surrogate model is built iteratively from simulations with varying lattice parameters to predict the dynamic aperture. The training of the model follows an active learning paradigm, which thus considerably reduces the number of samples needed from the computationally expensive simulations.
536 _ _ |a 899 - ohne Topic (POF4-899)
|0 G:(DE-HGF)POF4-899
|c POF4-899
|f POF IV
|x 0
588 _ _ |a Dataset connected to DataCite
650 _ 7 |a Accelerator Physics
|2 Other
650 _ 7 |a mc5-beam-dynamics-and-em-fields - MC5: Beam Dynamics and EM Fields
|2 Other
650 _ 7 |a mc5-d13-machine-learning - MC5.D13: Machine Learning
|2 Other
693 _ _ |a theory
|e no experiment theory work (theory)
|1 EXP:(DE-Ds200)theory-20200803
|0 EXP:(DE-Ds200)no_experiment-20200803
|5 EXP:(DE-Ds200)no_experiment-20200803
|x 0
700 1 _ |a Lagrange, Jean-Baptiste
|0 P:(DE-HGF)0
|b 1
700 1 _ |a Hirlaender, Simon
|0 P:(DE-HGF)0
|b 2
700 1 _ |a Oeftiger, Adrian
|0 P:(DE-Ds200)OR9462
|b 3
|e Corresponding author
773 _ _ |a 10.18429/JACOW-IPAC2023-WEPA026
856 4 _ |y OpenAccess
|u https://repository.gsi.de/record/358340/files/WEPA026.pdf
856 4 _ |y OpenAccess
|x pdfa
|u https://repository.gsi.de/record/358340/files/WEPA026.pdf?subformat=pdfa
909 C O |o oai:repository.gsi.de:358340
|p openaire
|p open_access
|p VDB
|p driver
|p dnbdelivery
910 1 _ |a GSI Helmholtzzentrum für Schwerionenforschung GmbH
|0 I:(DE-Ds200)20121206GSI
|k GSI
|b 3
|6 P:(DE-Ds200)OR9462
913 1 _ |a DE-HGF
|b Programmungebundene Forschung
|l ohne Programm
|1 G:(DE-HGF)POF4-890
|0 G:(DE-HGF)POF4-899
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-800
|4 G:(DE-HGF)POF
|v ohne Topic
|x 0
914 1 _ |y 2023
915 _ _ |a OpenAccess
|0 StatID:(DE-HGF)0510
|2 StatID
915 _ _ |a Creative Commons Attribution CC BY 4.0
|0 LIC:(DE-HGF)CCBY4
|2 HGFVOC
920 _ _ |l no
920 1 _ |0 I:(DE-Ds200)APH-20060809OR090
|k APH
|l Accelerator Physics
|x 0
980 _ _ |a contrib
980 _ _ |a VDB
980 _ _ |a UNRESTRICTED
980 _ _ |a I:(DE-Ds200)APH-20060809OR090
980 1 _ |a FullTexts

LibraryCollectionCLSMajorCLSMinorLanguageAuthor
Marc 21