Supervised Nonlinear Dynamic System for Soft Sensor Application Aided by Variational Auto-Encoder

被引：56

作者：

Shen, Bingbing ^{[1
,2
]}

Ge, Zhiqiang ^{[1
,2
]}

机构：

[1] Zhejiang Univ, Coll Control Sci & Engn, State Key Lab Ind Control Technol, Hangzhou 310027, Peoples R China

[2] Alibaba Zhejiang Univ, Joint Inst Frontier Technol, Hangzhou 310027, Peoples R China

来源：

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT | 2020年 / 69卷 / 09期

基金：

中国国家自然科学基金;

关键词：

Data models; Nonlinear dynamical systems; Probabilistic logic; Feature extraction; Numerical models; Neural networks; Principal component analysis; Dynamic data modeling; nonlinear features; probabilistic latent variable model; supervised nonlinear dynamic system (NDS); variational autoencoder (VAE); LATENT VARIABLE MODELS; REGRESSION; AUTOENCODER; ANALYTICS; DIAGNOSIS; FRAMEWORK; TUTORIAL;

D O I：

10.1109/TIM.2020.2968162

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Dynamic data modeling has been attracting much attention from researchers and has been introduced into the probabilistic latent variable model in the process industry. It is a huge challenge to extend these dynamic probabilistic latent variable models to nonlinear forms. In this article, a supervised nonlinear dynamic system (NDS) based on variational auto-encoder (VAE) is introduced for processes with dynamic behaviors and nonlinear characteristics. Based on the framework of VAE, which has a probabilistic data representation and a high fitting ability, the supervised NDS can extract effective nonlinear features for latent variable regression. The feasibility of the proposed supervised NDS is tested on two numerical examples and an industrial case. Detailed comparisons verify the effectiveness and superiority of the proposed model.

引用

页码：6132 / 6142

页数：11

共 43 条

[21] Process monitoring using variational autoencoder for high-dimensional nonlinear processes [J].

Lee, Seulki ;

Kwak, Mingu ;

Tsui, Kwok-Leung ;

Kim, Seoung Bum .

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2019, 83 :13-27

[22] A New Method of Dynamic Latent-Variable Modeling for Process Monitoring [J].

Li, Gang ;

Qin, S. Joe ;

Zhou, Donghua .

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2014, 61 (11) :6438-6445

[23] A systematic approach for soft sensor development [J].

Lin, Bao ;

Recke, Bodil ;

Knudsen, Jorgen K. H. ;

Jorgensen, Sten Bay .

COMPUTERS & CHEMICAL ENGINEERING, 2007, 31 (5-6) :419-425

[24]

Nasrabadi N. M., 2007, Springer google schola, V16, P49901, DOI [10.1117/1.2819119, DOI 10.18637/JSS.V017.B05, 10.5555/1162264]

[25]

Nyamundanda G, 2014, J ROY STAT SOC C, V63, P763, DOI 10.1111/rssc.12060

[26] Nonlinear dynamic data reconciliation and parameter estimation through particle swarm optimization: Application for an industrial polypropylene reactor [J].

Prata, Diego Martinez ;

Schwaab, Marcio ;

Lima, Enrique Luis ;

Pinto, Jose Carlos .

CHEMICAL ENGINEERING SCIENCE, 2009, 64 (18) :3953-3967

[27] Data-driven soft sensor development based on deep learning technique [J].

Shang, Chao ;

Yang, Fan ;

Huang, Dexian ;

Lyu, Wenxiang .

JOURNAL OF PROCESS CONTROL, 2014, 24 (03) :223-233

[28] Novel Bayesian Framework for Dynamic Soft Sensor Based on Support Vector Machine With Finite Impulse Response [J].

Shang, Chao ;

Gao, Xinqing ;

Yang, Fan ;

Huang, Dexian .

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, 2014, 22 (04) :1550-1557

[29] Semi-supervised selective ensemble learning based on distance to model for nonlinear soft sensor development [J].

Shao, Weiming ;

Tian, Xuemin .

NEUROCOMPUTING, 2017, 222 :91-104

[30] Nonlinear probabilistic latent variable regression models for soft sensor application: From shallow to deep structure [J].

Shen, Bingbing ;

Yao, Le ;

Ge, Zhiqiang .

CONTROL ENGINEERING PRACTICE, 2020, 94

← 1 2 3 4 5 →