000358340 001__ 358340
000358340 005__ 20250310164549.0
000358340 0247_ $$2doi$$a10.18429/JACOW-IPAC2023-WEPA026
000358340 0247_ $$2datacite_doi$$a10.15120/GSI-2025-00524
000358340 037__ $$aGSI-2025-00524
000358340 041__ $$aEnglish
000358340 1001_ $$0P:(DE-HGF)0$$aSantamaria Garcia, Andrea$$b0
000358340 1112_ $$a14th International Particle Accelerator Conference$$cVenice$$d2023-05-07 - 2023-05-12$$gIPAC2023$$wItaly
000358340 245__ $$aActive deep learning for nonlinear optics design of a vertical FFA accelerator
000358340 260__ $$aGeneva, Switzerland$$bJACoW Publishing$$c2023
000358340 300__ $$a2709 - 2712
000358340 3367_ $$2ORCID$$aCONFERENCE_PAPER
000358340 3367_ $$033$$2EndNote$$aConference Paper
000358340 3367_ $$2BibTeX$$aINPROCEEDINGS
000358340 3367_ $$2DRIVER$$aconferenceObject
000358340 3367_ $$2DataCite$$aOutput Types/Conference Paper
000358340 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1741076449_1280108
000358340 500__ $$aPublished by JACoW Publishing under the terms of the Creative Commons Attribution 4.0 license.
000358340 520__ $$aVertical 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.
000358340 536__ $$0G:(DE-HGF)POF4-899$$a899 - ohne Topic (POF4-899)$$cPOF4-899$$fPOF IV$$x0
000358340 588__ $$aDataset connected to DataCite
000358340 650_7 $$2Other$$aAccelerator Physics
000358340 650_7 $$2Other$$amc5-beam-dynamics-and-em-fields - MC5: Beam Dynamics and EM Fields
000358340 650_7 $$2Other$$amc5-d13-machine-learning - MC5.D13: Machine Learning
000358340 693__ $$0EXP:(DE-Ds200)no_experiment-20200803$$1EXP:(DE-Ds200)theory-20200803$$5EXP:(DE-Ds200)no_experiment-20200803$$atheory$$eno experiment theory work (theory)$$x0
000358340 7001_ $$0P:(DE-HGF)0$$aLagrange, Jean-Baptiste$$b1
000358340 7001_ $$0P:(DE-HGF)0$$aHirlaender, Simon$$b2
000358340 7001_ $$0P:(DE-Ds200)OR9462$$aOeftiger, Adrian$$b3$$eCorresponding author
000358340 773__ $$a10.18429/JACOW-IPAC2023-WEPA026
000358340 8564_ $$uhttps://repository.gsi.de/record/358340/files/WEPA026.pdf$$yOpenAccess
000358340 8564_ $$uhttps://repository.gsi.de/record/358340/files/WEPA026.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000358340 909CO $$ooai:repository.gsi.de:358340$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire
000358340 9101_ $$0I:(DE-Ds200)20121206GSI$$6P:(DE-Ds200)OR9462$$aGSI Helmholtzzentrum für Schwerionenforschung GmbH$$b3$$kGSI
000358340 9131_ $$0G:(DE-HGF)POF4-899$$1G:(DE-HGF)POF4-890$$2G:(DE-HGF)POF4-800$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bProgrammungebundene Forschung$$lohne Programm$$vohne Topic$$x0
000358340 9141_ $$y2023
000358340 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000358340 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000358340 920__ $$lno
000358340 9201_ $$0I:(DE-Ds200)APH-20060809OR090$$kAPH$$lAccelerator Physics$$x0
000358340 980__ $$acontrib
000358340 980__ $$aVDB
000358340 980__ $$aUNRESTRICTED
000358340 980__ $$aI:(DE-Ds200)APH-20060809OR090
000358340 9801_ $$aFullTexts