Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes

This paper implements machine learning (ML) classification algorithms on microstructural chemical maps to predict the constituent phases. Intensities of chemical species (Ca, Al, Si, etc.), and in some cases the nanomechanical properties measured at the corresponding points, form the input to the ML...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Emily Ford, Kailasnath Maneparambil, Narayanan Neithalath
Formato:	article
Lenguaje:	EN
Publicado:	Pouyan Press 2021
Materias:	machine learning nanoindentation chemical mapping microstructure cement pastes ultra-high performance concrete Technology T
Acceso en línea:	https://doaj.org/article/c79dd694882744c4a5afd8680a2948dc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c79dd694882744c4a5afd8680a2948dc
record_format	dspace
spelling	oai:doaj.org-article:c79dd694882744c4a5afd8680a2948dc2021-12-03T15:12:29ZMachine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes2588-287210.22115/scce.2021.302400.1357https://doaj.org/article/c79dd694882744c4a5afd8680a2948dc2021-10-01T00:00:00Zhttp://www.jsoftcivil.com/article_136759_ac647438733f6b7d898cc892fd60aa09.pdfhttps://doaj.org/toc/2588-2872This paper implements machine learning (ML) classification algorithms on microstructural chemical maps to predict the constituent phases. Intensities of chemical species (Ca, Al, Si, etc.), and in some cases the nanomechanical properties measured at the corresponding points, form the input to the ML model, which predicts the phase label (LD or HD C-S-H, clinker etc.) belonging to that location. Artificial neural networks (ANN) and forest ensemble methods are used for classification. Confusion matrices and receiver-operator characteristic (ROC) curves are used to analyze the classification efficiency. It is shown that, for complex microstructures such as those of ultra-high performance (UHP) pastes, the classifier performs well when nanomechanical information augments the chemical intensity data. For simpler systems such as well-hydrated plain cement pastes, the classifier accurately predicts the phase label from the intensities of Ca, Al, and Si alone. The work enables fast-and-efficient phase identification and property forecasting from microstructural chemical maps.Emily FordKailasnath ManeparambilNarayanan NeithalathPouyan Pressarticlemachine learningnanoindentationchemical mappingmicrostructurecement pastesultra-high performance concreteTechnologyTENJournal of Soft Computing in Civil Engineering, Vol 5, Iss 4, Pp 1-20 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	machine learning nanoindentation chemical mapping microstructure cement pastes ultra-high performance concrete Technology T
spellingShingle	machine learning nanoindentation chemical mapping microstructure cement pastes ultra-high performance concrete Technology T Emily Ford Kailasnath Maneparambil Narayanan Neithalath Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes
description	This paper implements machine learning (ML) classification algorithms on microstructural chemical maps to predict the constituent phases. Intensities of chemical species (Ca, Al, Si, etc.), and in some cases the nanomechanical properties measured at the corresponding points, form the input to the ML model, which predicts the phase label (LD or HD C-S-H, clinker etc.) belonging to that location. Artificial neural networks (ANN) and forest ensemble methods are used for classification. Confusion matrices and receiver-operator characteristic (ROC) curves are used to analyze the classification efficiency. It is shown that, for complex microstructures such as those of ultra-high performance (UHP) pastes, the classifier performs well when nanomechanical information augments the chemical intensity data. For simpler systems such as well-hydrated plain cement pastes, the classifier accurately predicts the phase label from the intensities of Ca, Al, and Si alone. The work enables fast-and-efficient phase identification and property forecasting from microstructural chemical maps.
format	article
author	Emily Ford Kailasnath Maneparambil Narayanan Neithalath
author_facet	Emily Ford Kailasnath Maneparambil Narayanan Neithalath
author_sort	Emily Ford
title	Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes
title_short	Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes
title_full	Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes
title_fullStr	Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes
title_full_unstemmed	Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes
title_sort	machine learning on microstructural chemical maps to classify component phases in cement pastes
publisher	Pouyan Press
publishDate	2021
url	https://doaj.org/article/c79dd694882744c4a5afd8680a2948dc
work_keys_str_mv	AT emilyford machinelearningonmicrostructuralchemicalmapstoclassifycomponentphasesincementpastes AT kailasnathmaneparambil machinelearningonmicrostructuralchemicalmapstoclassifycomponentphasesincementpastes AT narayananneithalath machinelearningonmicrostructuralchemicalmapstoclassifycomponentphasesincementpastes
_version_	1718373173824585728

Machine Learning on Microstructural Chemical Maps to Classify Component Phases in Cement Pastes

Ejemplares similares