Machine learning applications to predict two-phase flow patterns

Recent advances in artificial intelligence with traditional machine learning algorithms and deep learning architectures solve complex classification problems. This work presents the performance of different artificial intelligence models to classify two-phase flow patterns, showing the best alternat...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Harold Brayan Arteaga-Arteaga, Alejandro Mora-Rubio, Frank Florez, Nicolas Murcia-Orjuela, Cristhian Eduardo Diaz-Ortega, Simon Orozco-Arias, Melissa delaPava, Mario Alejandro Bravo-Ortíz, Melvin Robinson, Pablo Guillen-Rondon, Reinel Tabares-Soto
Formato:	article
Lenguaje:	EN
Publicado:	PeerJ Inc. 2021
Materias:	Flow patterns classification Machine learning Deep learning Extra trees Feature extraction Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/e1d8468b25c0456492404d1fa7d1a9e8
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e1d8468b25c0456492404d1fa7d1a9e8
record_format	dspace
spelling	oai:doaj.org-article:e1d8468b25c0456492404d1fa7d1a9e82021-12-01T15:05:16ZMachine learning applications to predict two-phase flow patterns10.7717/peerj-cs.7982376-5992https://doaj.org/article/e1d8468b25c0456492404d1fa7d1a9e82021-11-01T00:00:00Zhttps://peerj.com/articles/cs-798.pdfhttps://peerj.com/articles/cs-798/https://doaj.org/toc/2376-5992Recent advances in artificial intelligence with traditional machine learning algorithms and deep learning architectures solve complex classification problems. This work presents the performance of different artificial intelligence models to classify two-phase flow patterns, showing the best alternatives for this specific classification problem using two-phase flow regimes (liquid and gas) in pipes. Flow patterns are affected by physical variables such as superficial velocity, viscosity, density, and superficial tension. They also depend on the construction characteristics of the pipe, such as the angle of inclination and the diameter. We selected 12 databases (9,029 samples) to train and test machine learning models, considering these variables that influence the flow patterns. The primary dataset is Shoham (1982), containing 5,675 samples with six different flow patterns. An extensive set of metrics validated the results obtained. The most relevant characteristics for training the models using Shoham (1982) dataset are gas and liquid superficial velocities, angle of inclination, and diameter. Regarding the algorithms, the Extra Trees model classifies the flow patterns with the highest degree of fidelity, achieving an accuracy of 98.8%.Harold Brayan Arteaga-ArteagaAlejandro Mora-RubioFrank FlorezNicolas Murcia-OrjuelaCristhian Eduardo Diaz-OrtegaSimon Orozco-AriasMelissa delaPavaMario Alejandro Bravo-OrtízMelvin RobinsonPablo Guillen-RondonReinel Tabares-SotoPeerJ Inc.articleFlow patterns classificationMachine learningDeep learningExtra treesFeature extractionElectronic computers. Computer scienceQA75.5-76.95ENPeerJ Computer Science, Vol 7, p e798 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Flow patterns classification Machine learning Deep learning Extra trees Feature extraction Electronic computers. Computer science QA75.5-76.95
spellingShingle	Flow patterns classification Machine learning Deep learning Extra trees Feature extraction Electronic computers. Computer science QA75.5-76.95 Harold Brayan Arteaga-Arteaga Alejandro Mora-Rubio Frank Florez Nicolas Murcia-Orjuela Cristhian Eduardo Diaz-Ortega Simon Orozco-Arias Melissa delaPava Mario Alejandro Bravo-Ortíz Melvin Robinson Pablo Guillen-Rondon Reinel Tabares-Soto Machine learning applications to predict two-phase flow patterns
description	Recent advances in artificial intelligence with traditional machine learning algorithms and deep learning architectures solve complex classification problems. This work presents the performance of different artificial intelligence models to classify two-phase flow patterns, showing the best alternatives for this specific classification problem using two-phase flow regimes (liquid and gas) in pipes. Flow patterns are affected by physical variables such as superficial velocity, viscosity, density, and superficial tension. They also depend on the construction characteristics of the pipe, such as the angle of inclination and the diameter. We selected 12 databases (9,029 samples) to train and test machine learning models, considering these variables that influence the flow patterns. The primary dataset is Shoham (1982), containing 5,675 samples with six different flow patterns. An extensive set of metrics validated the results obtained. The most relevant characteristics for training the models using Shoham (1982) dataset are gas and liquid superficial velocities, angle of inclination, and diameter. Regarding the algorithms, the Extra Trees model classifies the flow patterns with the highest degree of fidelity, achieving an accuracy of 98.8%.
format	article
author	Harold Brayan Arteaga-Arteaga Alejandro Mora-Rubio Frank Florez Nicolas Murcia-Orjuela Cristhian Eduardo Diaz-Ortega Simon Orozco-Arias Melissa delaPava Mario Alejandro Bravo-Ortíz Melvin Robinson Pablo Guillen-Rondon Reinel Tabares-Soto
author_facet	Harold Brayan Arteaga-Arteaga Alejandro Mora-Rubio Frank Florez Nicolas Murcia-Orjuela Cristhian Eduardo Diaz-Ortega Simon Orozco-Arias Melissa delaPava Mario Alejandro Bravo-Ortíz Melvin Robinson Pablo Guillen-Rondon Reinel Tabares-Soto
author_sort	Harold Brayan Arteaga-Arteaga
title	Machine learning applications to predict two-phase flow patterns
title_short	Machine learning applications to predict two-phase flow patterns
title_full	Machine learning applications to predict two-phase flow patterns
title_fullStr	Machine learning applications to predict two-phase flow patterns
title_full_unstemmed	Machine learning applications to predict two-phase flow patterns
title_sort	machine learning applications to predict two-phase flow patterns
publisher	PeerJ Inc.
publishDate	2021
url	https://doaj.org/article/e1d8468b25c0456492404d1fa7d1a9e8
work_keys_str_mv	AT haroldbrayanarteagaarteaga machinelearningapplicationstopredicttwophaseflowpatterns AT alejandromorarubio machinelearningapplicationstopredicttwophaseflowpatterns AT frankflorez machinelearningapplicationstopredicttwophaseflowpatterns AT nicolasmurciaorjuela machinelearningapplicationstopredicttwophaseflowpatterns AT cristhianeduardodiazortega machinelearningapplicationstopredicttwophaseflowpatterns AT simonorozcoarias machinelearningapplicationstopredicttwophaseflowpatterns AT melissadelapava machinelearningapplicationstopredicttwophaseflowpatterns AT marioalejandrobravoortiz machinelearningapplicationstopredicttwophaseflowpatterns AT melvinrobinson machinelearningapplicationstopredicttwophaseflowpatterns AT pabloguillenrondon machinelearningapplicationstopredicttwophaseflowpatterns AT reineltabaressoto machinelearningapplicationstopredicttwophaseflowpatterns
_version_	1718404901651873792

Machine learning applications to predict two-phase flow patterns

Ejemplares similares