Interpreting neural networks trained to predict plasma temperature from optical emission spectra

Simple item page
dc.contributor.authorKépeš, Erikcs
dc.contributor.authorSaeidfirouzeh, Homacs
dc.contributor.authorLaitl, Vojtěchcs
dc.contributor.authorVrábel, Jakubcs
dc.contributor.authorKubelík, Petrcs
dc.contributor.authorPořízka, Pavelcs
dc.contributor.authorFerus, Martincs
dc.contributor.authorKaiser, Jozefcs
dc.coverage.issue4cs
dc.coverage.volume39cs
dc.date.accessioned2025-02-03T14:48:32Z
dc.date.available2025-02-03T14:48:32Z
dc.date.issued2024-04-03cs
dc.description.abstractWe explore the application of artificial neural networks (ANNs) for predicting plasma temperatures in Laser-Induced Breakdown Spectroscopy (LIBS) analysis. Estimating plasma temperature from emission spectra is often challenging due to spectral interference and matrix effects. Traditional methods like the Boltzmann plot technique have limitations, both in applicability due to various matrix effects and in accuracy owing to the uncertainty of the underlying spectroscopic constants. Consequently, ANNs have already been successfully demonstrated as a viable alternative for plasma temperature prediction. We leverage synthetic data to isolate temperature effects from other factors and study the relationship between the LIBS spectra and temperature learnt by the ANN. We employ various post-hoc model interpretation techniques, including gradient-based methods, to verify that ANNs learn meaningful spectroscopic features for temperature prediction. Our findings demonstrate the potential of ANNs to learn complex relationships in LIBS spectra, offering a promising avenue for improved plasma temperature estimation and enhancing the overall accuracy of LIBS analysis. ANN can learn spectroscopic trends widely used by domain experts for plasma temperature estimation using emission spectra.en
dc.formattextcs
dc.format.extent1160-1174cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationJournal of Analytical Atomic Spectrometry. 2024, vol. 39, issue 4, p. 1160-1174.en
dc.identifier.doi10.1039/d3ja00363acs
dc.identifier.issn1364-5544cs
dc.identifier.orcid0000-0002-7086-2613cs
dc.identifier.orcid0000-0001-5629-3314cs
dc.identifier.orcid0000-0002-8604-7365cs
dc.identifier.orcid0000-0002-7397-125Xcs
dc.identifier.other188826cs
dc.identifier.researcheridF-2136-2018cs
dc.identifier.researcheridG-9463-2014cs
dc.identifier.researcheridD-6800-2012cs
dc.identifier.scopus57190620988cs
dc.identifier.scopus55312098800cs
dc.identifier.scopus7402184758cs
dc.identifier.urihttps://hdl.handle.net/11012/249970
dc.language.isoencs
dc.publisherROYAL SOC CHEMISTRYcs
dc.relation.ispartofJournal of Analytical Atomic Spectrometrycs
dc.relation.urihttps://pubs.rsc.org/en/content/articlelanding/2024/ja/d3ja00363acs
dc.rightsCreative Commons Attribution-NonCommercial 3.0 Unportedcs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/1364-5544/cs
dc.rights.urihttp://creativecommons.org/licenses/by-nc/3.0/cs
dc.subjectINDUCED BREAKDOWN SPECTROSCOPYen
dc.subjectLASER-INDUCED PLASMAen
dc.subjectCHEMCAM INSTRUMENT SUITEen
dc.subjectLINEen
dc.subjectSCIENCEen
dc.subjectSPECTROMETRYen
dc.subjectPARAMETERSen
dc.subjectPRECISIONen
dc.subjectDENSITYen
dc.subjectTHOMSONen
dc.titleInterpreting neural networks trained to predict plasma temperature from optical emission spectraen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.grantNumberinfo:eu-repo/grantAgreement/GA0/GF/GF23-05186Kcs
sync.item.dbidVAV-188826en
sync.item.dbtypeVAVen
sync.item.insts2025.02.03 15:48:32en
sync.item.modts2025.01.17 16:42:29en
thesis.grantorVysoké učení technické v Brně. Fakulta strojního inženýrství. Ústav fyzikálního inženýrstvícs
thesis.grantorVysoké učení technické v Brně. Středoevropský technologický institut VUT. Pokročilé instrumentace a metody pro charakterizace materiálůcs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
d3ja00363a.pdf
Size:
2.11 MB
Format:
Adobe Portable Document Format
Description:
file d3ja00363a.pdf
Download
Collections
Ústav fyzikálního inženýrství
Pokročilé instrumentace a metody pro charakterizace materiálů