Transfer learning for process fault diagnosis: Knowledge transfer from simulation to physical processes

doi:10.1016/j.compchemeng.2020.106904

	Transfer learning for process fault diagnosis: Knowledge transfer from simulation to physical processes
	Li, Weijun 1; Gu, Sai 1; Zhang, Xiangping 2; Chen, Tao 1
刊名	COMPUTERS & CHEMICAL ENGINEERING
	2020-08-04
卷号	139 页码:10
关键词	Fault diagnosis Transfer learning Model-process mismatch Deep learning Computer simulation Domain adaptation
ISSN号	0098-1354
DOI	10.1016/j.compchemeng.2020.106904
英文摘要	Deep learning has shown great promise in process fault diagnosis. However, due to the lack of sufficient labelled fault data, its application has been limited. This limitation may be overcome by using the data generated from computer simulations. In this study, we consider using simulated data to train deep neural network models. As there inevitably is model-process mismatch, we further apply transfer learning approach to reduce the discrepancies between the simulation and physical domains. This approach will allow the diagnostic knowledge contained in the computer simulation being applied to the physical process. To this end, a deep transfer learning network is designed by integrating the convolutional neural network and advanced domain adaptation techniques. Two case studies are used to illustrate the effectiveness of the proposed method for fault diagnosis: a continuously stirred tank reactor and the pulp mill plant benchmark problem. (c) 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license. (http://creativecommons.org/licenses/by/4.0/)
资助项目	EPSRC[EP/R001588/1] ; BBSRC[BB/S020896/1] ; Unilever-IPE-Surrey collaborative doctoral training programme
WOS关键词	MODEL-PLANT MISMATCH ; STATE
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	PERGAMON-ELSEVIER SCIENCE LTD
WOS记录号	WOS:000555543100013
资助机构	EPSRC ; BBSRC ; Unilever-IPE-Surrey collaborative doctoral training programme
内容类型	期刊论文
源URL	[http://ir.ipe.ac.cn/handle/122111/41597]
专题	中国科学院过程工程研究所
通讯作者	Chen, Tao
作者单位	1.Univ Surrey, Dept Chem & Proc Engn, Guildford GU2 7XH, Surrey, England 2.Chinese Acad Sci, Inst Proc Engn, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Li, Weijun,Gu, Sai,Zhang, Xiangping,et al. Transfer learning for process fault diagnosis: Knowledge transfer from simulation to physical processes[J]. COMPUTERS & CHEMICAL ENGINEERING,2020,139:10.
APA	Li, Weijun,Gu, Sai,Zhang, Xiangping,&Chen, Tao.(2020).Transfer learning for process fault diagnosis: Knowledge transfer from simulation to physical processes.COMPUTERS & CHEMICAL ENGINEERING,139,10.
MLA	Li, Weijun,et al."Transfer learning for process fault diagnosis: Knowledge transfer from simulation to physical processes".COMPUTERS & CHEMICAL ENGINEERING 139(2020):10.