A multi-source transfer learning method for new mode monitoring in industrial processes

被引：0

作者：

Wang, Kai ^{[1
]}

Zhou, Wenxuan ^{[1
]}

Liu, Chenliang ^{[1
]}

Yuan, Xiaofeng ^{[1
]}

Wang, Yalin ^{[1
]}

机构：

[1] Cent South Univ, Sch Automat, Changsha 410083, Peoples R China

来源：

2022 8TH INTERNATIONAL CONFERENCE ON CONTROL, DECISION AND INFORMATION TECHNOLOGIES (CODIT'22) | 2022年

基金：

中国国家自然科学基金;

关键词：

Process monitoring; Multi-mode data; Few samples; Transfer learning; Common subspace; MULTIMODE PROCESS; EXTRACTION; COMMON;

D O I：

10.1109/CODIT55151.2022.9804089

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Since the change of operation condition is common in industrial processes, it could cause historical process data multimodal characteristics. When the working conditions are switched, the new mode will suffer small sample problem in the initial stage of the working mode switching, which brings difficulties to the monitoring of the new mode. Different from traditional modeling method which only considers the new mode data, this paper proposes a novel multi-source transfer learning method that considers both the historical multimode and new mode data. First, the common features of historical multimode data are extracted. Then, the extracted features are transformed into the model of new mode data. In order to alleviate the problem of insufficient samples of the current working mode, the common subspace of the new mode is obtained by combining the common features of the historical multimode with the new mode data. Finally, a numerical case and a real industrial hydrocracking process are used to validate the effectiveness of the proposed method.

引用

页码：124 / 129

页数：6

共 23 条

[1] Multimodal process monitoring based on variational Bayesian PCA and Kullback-Leibler divergence between mixture models [J].

Cao, Yue ;

Jan, Nabil Magbool ;

Huang, Biao ;

Fang, Mengqi ;

Wang, Yalin ;

Gui, Weihua .

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2021, 210

[2] Gaussian process model based multi-source labeled data transfer learning for reducing cost of modeling target chemical processes with unlabeled data [J].

Chan, Lester Lik Teck ;

Chen, Junghui .

CONTROL ENGINEERING PRACTICE, 2021, 117 (117)

[3] Adaptive Transfer Learning of Cross-Spatiotemporal Canonical Correlation Analysis for Plant-Wide Process Monitoring [J].

Cheng, Hongchao ;

Liu, Yiqi ;

Huang, Daoping ;

Pan, Yongping ;

Wang, Qilin .

INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2020, 59 (49) :21602-21614

[4]

Gower J.C., 2004, Procrustes Problems

[5]

GuoYang L. C. L., 2011, T CHINA ELECTROTECHN, V11

[6] Transfer Dictionary Learning Method for Cross-Domain Multimode Process Monitoring and Fault Isolation [J].

Huang, Keke ;

Wen, Haofei ;

Zhou, Can ;

Yang, Chunhua ;

Gui, Weihua .

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2020, 69 (11) :8713-8724

[7] Tensor Decompositions and Applications [J].

Kolda, Tamara G. ;

Bader, Brett W. .

SIAM REVIEW, 2009, 51 (03) :455-500

[8]

Liu C, 2021, IEEE Trans. Ind. Inform

[9] Deep learning with nonlocal and local structure preserving stacked autoencoder for soft sensor in industrial processes [J].

Liu, Chenliang ;

Wang, Yalin ;

Wang, Kai ;

Yuan, Xiaofeng .

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2021, 104

[10] Dual-layer feature extraction based soft sensor methods and applications to industrial polyethylene processes [J].

Liu, Jingxiang ;

Hou, Jie ;

Chen, Junghui .

COMPUTERS & CHEMICAL ENGINEERING, 2021, 154

← 1 2 3 →