Datum
2024-07-02Schlagwort
500 Naturwissenschaften 600 Technik FT-IR-SpektroskopieDiskriminanzanalyseMastixMaschinelles LernenText MiningAlterungMetadata
Zur Langanzeige
Aufsatz
FTIR spectroscopy and multivariate discriminant analysis for classifying bituminous mastics: Exploring aging states and mastic composition
Zusammenfassung
This study aimed to differentiate among various bituminous mastics through the analysis of FTIR spectroscopy data. The mastics, incorporating three filler types, a 50/70 penetration index binder, and foamed bitumen, underwent chemical examination via FTIR spectroscopy after exposure to various aging cycles. Employing different multivariate discriminant analysis approaches, the investigation identified an optimal method, i.e., PLSR-LDA, which achieved a prediction precision of 100 % in distinguishing between various types of mixtures. The governing wavenumbers in the classification were determined through PLSR-LDA analysis by calculating VIP scores. Next, integrals associated with the highlighted wavenumber regions served as input to the DT classifier, resulting in classification with the prediction precisions of 96 % and 88 % for mastic types and aging states, respectively. The subsequent analysis underscored the significance of the CaCo₃ absorption band in discriminating between different mastic types. Similarly, C-H stretching of CH₃ and carbonyl bands were identified as pivotal regions governing the classification based on aging states.
Zitierform
In: Construction and Building Materials Volume 438 (2024-07-02) eissn:1879-0526Förderhinweis
Gefördert im Rahmen des Projekts DEALZitieren
@article{doi:10.17170/kobra-2024070910501,
author={Motevalizadeh, Seyed Mohsen and Mollenhauer, Konrad and Wetekam, Jens},
title={FTIR spectroscopy and multivariate discriminant analysis for classifying bituminous mastics: Exploring aging states and mastic composition},
journal={Construction and Building Materials},
year={2024}
}
0500 Oax 0501 Text $btxt$2rdacontent 0502 Computermedien $bc$2rdacarrier 1100 2024$n2024 1500 1/eng 2050 ##0##http://hdl.handle.net/123456789/15933 3000 Motevalizadeh, Seyed Mohsen 3010 Mollenhauer, Konrad 3010 Wetekam, Jens 4000 FTIR spectroscopy and multivariate discriminant analysis for classifying bituminous mastics: Exploring aging states and mastic composition / Motevalizadeh, Seyed Mohsen 4030 4060 Online-Ressource 4085 ##0##=u http://nbn-resolving.de/http://hdl.handle.net/123456789/15933=x R 4204 \$dAufsatz 4170 5550 {{FT-IR-Spektroskopie}} 5550 {{Diskriminanzanalyse}} 5550 {{Mastix}} 5550 {{Maschinelles Lernen}} 5550 {{Text Mining}} 5550 {{Alterung}} 7136 ##0##http://hdl.handle.net/123456789/15933
2024-07-22T15:03:03Z 2024-07-22T15:03:03Z 2024-07-02 doi:10.17170/kobra-2024070910501 http://hdl.handle.net/123456789/15933 Gefördert im Rahmen des Projekts DEAL eng Namensnennung 4.0 International http://creativecommons.org/licenses/by/4.0/ bituminous mastics FTIR spectroscopy multivariate discriminant analysis machine learning data mining 500 600 FTIR spectroscopy and multivariate discriminant analysis for classifying bituminous mastics: Exploring aging states and mastic composition Aufsatz This study aimed to differentiate among various bituminous mastics through the analysis of FTIR spectroscopy data. The mastics, incorporating three filler types, a 50/70 penetration index binder, and foamed bitumen, underwent chemical examination via FTIR spectroscopy after exposure to various aging cycles. Employing different multivariate discriminant analysis approaches, the investigation identified an optimal method, i.e., PLSR-LDA, which achieved a prediction precision of 100 % in distinguishing between various types of mixtures. The governing wavenumbers in the classification were determined through PLSR-LDA analysis by calculating VIP scores. Next, integrals associated with the highlighted wavenumber regions served as input to the DT classifier, resulting in classification with the prediction precisions of 96 % and 88 % for mastic types and aging states, respectively. The subsequent analysis underscored the significance of the CaCo₃ absorption band in discriminating between different mastic types. Similarly, C-H stretching of CH₃ and carbonyl bands were identified as pivotal regions governing the classification based on aging states. open access Motevalizadeh, Seyed Mohsen Mollenhauer, Konrad Wetekam, Jens doi:10.1016/j.conbuildmat.2024.137188 FT-IR-Spektroskopie Diskriminanzanalyse Mastix Maschinelles Lernen Text Mining Alterung publishedVersion eissn:1879-0526 Construction and Building Materials Volume 438 false 137188
Die folgenden Lizenzbestimmungen sind mit dieser Ressource verbunden: