Fusion-Learning of Bayesian Network Models for Fault Diagnostics

Bayesian Network (BN) models are being successfully applied to improve fault diagnosis, which in turn can improve equipment uptime and customer service. Most of these BN models are essentially trained using quantitative data obtained from sensors. However, sensors may not be able to cover all faults...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Toyosi Ademujimi, Vittaldas Prabhu
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	fusion-learning Bayesian Network smart maintenance fault diagnostics natural language processing technical language processing Chemical technology TP1-1185
Acceso en línea:	https://doaj.org/article/2468e21bba1f47cbae537d27f9c99e1b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:2468e21bba1f47cbae537d27f9c99e1b
record_format	dspace
spelling	oai:doaj.org-article:2468e21bba1f47cbae537d27f9c99e1b2021-11-25T18:58:05ZFusion-Learning of Bayesian Network Models for Fault Diagnostics10.3390/s212276331424-8220https://doaj.org/article/2468e21bba1f47cbae537d27f9c99e1b2021-11-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/22/7633https://doaj.org/toc/1424-8220Bayesian Network (BN) models are being successfully applied to improve fault diagnosis, which in turn can improve equipment uptime and customer service. Most of these BN models are essentially trained using quantitative data obtained from sensors. However, sensors may not be able to cover all faults and therefore such BN models would be incomplete. Furthermore, many systems have maintenance logs that can serve as qualitative data, potentially containing historic causation information in unstructured natural language replete with technical terms. The motivation of this paper is to leverage all of the data available to improve BN learning. Specifically, we propose a method for fusion-learning of BNs: for quantitative data obtained from sensors, metrology data and qualitative data from maintenance logs, corrective and preventive action reports, and then follow by fusing these two BNs. Furthermore, we propose a human-in-the-loop approach for expert knowledge elicitation of the BN structure aided by logged natural language data instead of relying exclusively on their anecdotal memory. The resulting fused BN model can be expected to provide improved diagnostics as it has a wider fault coverage than the individual BNs. We demonstrate the efficacy of our proposed method using real world data from uninterruptible power supply (UPS) fault diagnostics.Toyosi AdemujimiVittaldas PrabhuMDPI AGarticlefusion-learningBayesian Networksmart maintenancefault diagnosticsnatural language processingtechnical language processingChemical technologyTP1-1185ENSensors, Vol 21, Iss 7633, p 7633 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	fusion-learning Bayesian Network smart maintenance fault diagnostics natural language processing technical language processing Chemical technology TP1-1185
spellingShingle	fusion-learning Bayesian Network smart maintenance fault diagnostics natural language processing technical language processing Chemical technology TP1-1185 Toyosi Ademujimi Vittaldas Prabhu Fusion-Learning of Bayesian Network Models for Fault Diagnostics
description	Bayesian Network (BN) models are being successfully applied to improve fault diagnosis, which in turn can improve equipment uptime and customer service. Most of these BN models are essentially trained using quantitative data obtained from sensors. However, sensors may not be able to cover all faults and therefore such BN models would be incomplete. Furthermore, many systems have maintenance logs that can serve as qualitative data, potentially containing historic causation information in unstructured natural language replete with technical terms. The motivation of this paper is to leverage all of the data available to improve BN learning. Specifically, we propose a method for fusion-learning of BNs: for quantitative data obtained from sensors, metrology data and qualitative data from maintenance logs, corrective and preventive action reports, and then follow by fusing these two BNs. Furthermore, we propose a human-in-the-loop approach for expert knowledge elicitation of the BN structure aided by logged natural language data instead of relying exclusively on their anecdotal memory. The resulting fused BN model can be expected to provide improved diagnostics as it has a wider fault coverage than the individual BNs. We demonstrate the efficacy of our proposed method using real world data from uninterruptible power supply (UPS) fault diagnostics.
format	article
author	Toyosi Ademujimi Vittaldas Prabhu
author_facet	Toyosi Ademujimi Vittaldas Prabhu
author_sort	Toyosi Ademujimi
title	Fusion-Learning of Bayesian Network Models for Fault Diagnostics
title_short	Fusion-Learning of Bayesian Network Models for Fault Diagnostics
title_full	Fusion-Learning of Bayesian Network Models for Fault Diagnostics
title_fullStr	Fusion-Learning of Bayesian Network Models for Fault Diagnostics
title_full_unstemmed	Fusion-Learning of Bayesian Network Models for Fault Diagnostics
title_sort	fusion-learning of bayesian network models for fault diagnostics
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/2468e21bba1f47cbae537d27f9c99e1b
work_keys_str_mv	AT toyosiademujimi fusionlearningofbayesiannetworkmodelsforfaultdiagnostics AT vittaldasprabhu fusionlearningofbayesiannetworkmodelsforfaultdiagnostics
_version_	1718410452058243072

Fusion-Learning of Bayesian Network Models for Fault Diagnostics

Ejemplares similares