A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring

Data-driven process-monitoring methods have been the mainstream for complex industrial systems due to their universality and the reduced need for reaction mechanisms and first-principles knowledge. However, most data-driven process-monitoring methods assume that historical training data and online t...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Chunhua Yang, Huiping Liang, Keke Huang, Yonggang Li, Weihua Gui
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2021
Materias:	Process monitoring Multimode process Dictionary learning Transfer learning Engineering (General). Civil engineering (General) TA1-2040
Acceso en línea:	https://doaj.org/article/801db0aa512b49229d7754e159648070
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:801db0aa512b49229d7754e159648070
record_format	dspace
spelling	oai:doaj.org-article:801db0aa512b49229d7754e1596480702021-11-14T04:32:18ZA Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring2095-809910.1016/j.eng.2020.08.028https://doaj.org/article/801db0aa512b49229d7754e1596480702021-09-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2095809921003118https://doaj.org/toc/2095-8099Data-driven process-monitoring methods have been the mainstream for complex industrial systems due to their universality and the reduced need for reaction mechanisms and first-principles knowledge. However, most data-driven process-monitoring methods assume that historical training data and online testing data follow the same distribution. In fact, due to the harsh environment of industrial systems, the collected data from real industrial processes are always affected by many factors, such as the changeable operating environment, variation in the raw materials, and production indexes. These factors often cause the distributions of online monitoring data and historical training data to differ, which induces a model mismatch in the process-monitoring task. Thus, it is difficult to achieve accurate process monitoring when a model learned from training data is applied to actual online monitoring. In order to resolve the problem of the distribution divergence between historical training data and online testing data that is induced by changeable operation environments, a robust transfer dictionary learning (RTDL) algorithm is proposed in this paper for industrial process monitoring. The RTDL is a synergy of representative learning and domain adaptive transfer learning. The proposed method regards historical training data and online testing data as the source domain and the target domain, respectively, in the transfer learning problem. Maximum mean discrepancy regularization and linear discriminant analysis-like regularization are then incorporated into the dictionary learning framework, which can reduce the distribution divergence between the source domain and target domain. In this way, a robust dictionary can be learned even if the characteristics of the source domain and target domain are evidently different under the interference of a realistic and changeable operation environment. Such a dictionary can effectively improve the performance of process monitoring and mode classification. Extensive experiments including a numerical simulation and two industrial systems are conducted to verify the efficiency and superiority of the proposed method.Chunhua YangHuiping LiangKeke HuangYonggang LiWeihua GuiElsevierarticleProcess monitoringMultimode processDictionary learningTransfer learningEngineering (General). Civil engineering (General)TA1-2040ENEngineering, Vol 7, Iss 9, Pp 1262-1273 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Process monitoring Multimode process Dictionary learning Transfer learning Engineering (General). Civil engineering (General) TA1-2040
spellingShingle	Process monitoring Multimode process Dictionary learning Transfer learning Engineering (General). Civil engineering (General) TA1-2040 Chunhua Yang Huiping Liang Keke Huang Yonggang Li Weihua Gui A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring
description	Data-driven process-monitoring methods have been the mainstream for complex industrial systems due to their universality and the reduced need for reaction mechanisms and first-principles knowledge. However, most data-driven process-monitoring methods assume that historical training data and online testing data follow the same distribution. In fact, due to the harsh environment of industrial systems, the collected data from real industrial processes are always affected by many factors, such as the changeable operating environment, variation in the raw materials, and production indexes. These factors often cause the distributions of online monitoring data and historical training data to differ, which induces a model mismatch in the process-monitoring task. Thus, it is difficult to achieve accurate process monitoring when a model learned from training data is applied to actual online monitoring. In order to resolve the problem of the distribution divergence between historical training data and online testing data that is induced by changeable operation environments, a robust transfer dictionary learning (RTDL) algorithm is proposed in this paper for industrial process monitoring. The RTDL is a synergy of representative learning and domain adaptive transfer learning. The proposed method regards historical training data and online testing data as the source domain and the target domain, respectively, in the transfer learning problem. Maximum mean discrepancy regularization and linear discriminant analysis-like regularization are then incorporated into the dictionary learning framework, which can reduce the distribution divergence between the source domain and target domain. In this way, a robust dictionary can be learned even if the characteristics of the source domain and target domain are evidently different under the interference of a realistic and changeable operation environment. Such a dictionary can effectively improve the performance of process monitoring and mode classification. Extensive experiments including a numerical simulation and two industrial systems are conducted to verify the efficiency and superiority of the proposed method.
format	article
author	Chunhua Yang Huiping Liang Keke Huang Yonggang Li Weihua Gui
author_facet	Chunhua Yang Huiping Liang Keke Huang Yonggang Li Weihua Gui
author_sort	Chunhua Yang
title	A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring
title_short	A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring
title_full	A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring
title_fullStr	A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring
title_full_unstemmed	A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring
title_sort	robust transfer dictionary learning algorithm for industrial process monitoring
publisher	Elsevier
publishDate	2021
url	https://doaj.org/article/801db0aa512b49229d7754e159648070
work_keys_str_mv	AT chunhuayang arobusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT huipingliang arobusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT kekehuang arobusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT yonggangli arobusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT weihuagui arobusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT chunhuayang robusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT huipingliang robusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT kekehuang robusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT yonggangli robusttransferdictionarylearningalgorithmforindustrialprocessmonitoring AT weihuagui robusttransferdictionarylearningalgorithmforindustrialprocessmonitoring
_version_	1718429965367640064

A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring

Ejemplares similares