001006426 001__ 1006426
001006426 005__ 20240711114048.0
001006426 0247_ $$2doi$$a10.1016/j.nme.2023.101396
001006426 0247_ $$2Handle$$a2128/34270
001006426 0247_ $$2WOS$$aWOS:000990102500001
001006426 037__ $$aFZJ-2023-01661
001006426 082__ $$a624
001006426 1001_ $$0P:(DE-Juel1)176921$$aDasbach, Stefan$$b0$$eCorresponding author
001006426 245__ $$aTowards fast surrogate models for interpolation of tokamak edge plasmas
001006426 260__ $$aAmsterdam [u.a.]$$bElsevier$$c2023
001006426 3367_ $$2DRIVER$$aarticle
001006426 3367_ $$2DataCite$$aOutput Types/Journal article
001006426 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1681202262_13485
001006426 3367_ $$2BibTeX$$aARTICLE
001006426 3367_ $$2ORCID$$aJOURNAL_ARTICLE
001006426 3367_ $$00$$2EndNote$$aJournal Article
001006426 520__ $$aOne of the major design limitations for tokamak fusion reactors is the heat load that can be sustained by the materials at the divertor target. Developing a full understanding of how machine or operation parameters affect the conditions at the divertor requires an enormous number of simulations. A promising approach to circumvent this is to use machine learning models trained on simulation data as surrogate models. Once trained such surrogate models can make fast predictions for any scenario in the design parameter space. In future such simulation based surrogate models could be used in system codes for rapid design studies of future fusion power plants. This work presents the first steps towards the development of such surrogate models for plasma exhaust and the datasets required for their training. Machine learning models like neural networks usually require several thousand data points for training, but the exact amount of data required varies from case to case. Due to the long runtimes of simulations we aim at finding the minimal amount of training data required. A preliminary dataset based on SOLPS-ITER simulations with varying tokamak design parameters, including the major radius, magnetic field strength and neutral density is constructed. To be able to generate more training data within reasonable computation time the simulations in the dataset use fluid neutral simulations and no fluid drift effects. The dataset is used to train a simple neural network and Gradient Boosted Regression Trees and test how the performance depends on the number of training simulations.
001006426 536__ $$0G:(DE-HGF)POF4-134$$a134 - Plasma-Wand-Wechselwirkung (POF4-134)$$cPOF4-134$$fPOF IV$$x0
001006426 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
001006426 7001_ $$0P:(DE-Juel1)5247$$aWiesen, Sven$$b1
001006426 773__ $$0PERI:(DE-600)2808888-8$$a10.1016/j.nme.2023.101396$$gVol. 34, p. 101396 -$$p101396 -$$tNuclear materials and energy$$v34$$x2352-1791$$y2023
001006426 8564_ $$uhttps://juser.fz-juelich.de/record/1006426/files/1-s2.0-S2352179123000352-main.pdf$$yOpenAccess
001006426 8767_ $$d2023-04-03$$eAPC$$jZahlung erfolgt
001006426 909CO $$ooai:juser.fz-juelich.de:1006426$$pdnbdelivery$$popenCost$$pVDB$$pdriver$$pOpenAPC$$popen_access$$popenaire
001006426 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)176921$$aForschungszentrum Jülich$$b0$$kFZJ
001006426 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)5247$$aForschungszentrum Jülich$$b1$$kFZJ
001006426 9131_ $$0G:(DE-HGF)POF4-134$$1G:(DE-HGF)POF4-130$$2G:(DE-HGF)POF4-100$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bForschungsbereich Energie$$lFusion$$vPlasma-Wand-Wechselwirkung$$x0
001006426 9141_ $$y2023
001006426 915pc $$0PC:(DE-HGF)0000$$2APC$$aAPC keys set
001006426 915pc $$0PC:(DE-HGF)0001$$2APC$$aLocal Funding
001006426 915pc $$0PC:(DE-HGF)0002$$2APC$$aDFG OA Publikationskosten
001006426 915pc $$0PC:(DE-HGF)0003$$2APC$$aDOAJ Journal
001006426 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
001006426 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2023-03-30
001006426 915__ $$0StatID:(DE-HGF)0700$$2StatID$$aFees$$d2023-03-30
001006426 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001006426 915__ $$0StatID:(DE-HGF)0561$$2StatID$$aArticle Processing Charges$$d2023-03-30
001006426 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2023-03-30
001006426 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bNUCL MATER ENERGY : 2022$$d2023-08-22
001006426 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-08-22
001006426 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2023-05-02T08:51:15Z
001006426 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2023-05-02T08:51:15Z
001006426 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Peer review$$d2023-05-02T08:51:15Z
001006426 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-08-22
001006426 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-08-22
001006426 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology$$d2023-08-22
001006426 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5$$d2023-08-22
001006426 920__ $$lyes
001006426 9201_ $$0I:(DE-Juel1)IEK-4-20101013$$kIEK-4$$lPlasmaphysik$$x0
001006426 9801_ $$aAPC
001006426 9801_ $$aFullTexts
001006426 980__ $$ajournal
001006426 980__ $$aVDB
001006426 980__ $$aUNRESTRICTED
001006426 980__ $$aI:(DE-Juel1)IEK-4-20101013
001006426 980__ $$aAPC
001006426 981__ $$aI:(DE-Juel1)IFN-1-20101013