Nonlinear dynamic process monitoring based on latent mapping embedding deep neural networks

被引:0
|
作者
Yu, Zhenhua [1 ]
Wang, Wenjing [1 ]
Wang, Xueting [2 ]
Jiang, Qingchao [1 ]
Wang, Guan [2 ]
机构
[1] East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai 200237, Peoples R China
[2] East China Univ Sci & Technol, State Key Lab Bioreactor Engn, Shanghai 200237, Peoples R China
来源
CANADIAN JOURNAL OF CHEMICAL ENGINEERING | 2025年 / 103卷 / 04期
基金
中国国家自然科学基金;
关键词
date-driven process monitoring; latent mapping; nonlinear system; stacked autoencoder; FAULT-DETECTION; DIAGNOSIS; MODEL; KPCA; AUTOENCODERS; RELEVANT;
D O I
10.1002/cjce.25461
中图分类号
TQ [化学工业];
学科分类号
0817 ;
摘要
In industrial processes, complex nonlinearity and dynamics generally exist, making it challenging to achieve good results using conventional process monitoring methods. In this paper, a latent mapping embedding neural network method (LMNN) is proposed for efficient monitoring of nonlinear dynamic processes. First, a deep neural network (DNN) is employed to acquire features of state variables from nonlinear process data and expand them along with the input to a new feature subspace. Second, a latent mapping (LM) method is used to map the high-dimensional feature subspace to a low-dimensional subspace that includes the most beneficial time series information. Then the entire neural network and regression parameters are obtained through an end-to-end learning manner, through which the nonlinearity and process dynamics are well characterized. Subsequently, prediction error-based residual is generated and the monitoring model is established. The performance of the proposed method is verified through a simulation of penicillin production process and an actual fermentation process of penicillin. Comparisons with state-of-the-art methods are carried out, and results validate the effectiveness and superiority of the proposed method.
引用
收藏
页码:1802 / 1812
页数:11
相关论文
共 50 条
  • [21] Soft Sensors to Monitoring a Multivariate Nonlinear Process Using Neural Networks
    Nathalia Arthur Brunet Monteiro
    Jaidilson Jó da Silva
    José Sérgio da Rocha Neto
    Journal of Control, Automation and Electrical Systems, 2019, 30 : 54 - 62
  • [22] Latent Backdoor Attacks on Deep Neural Networks
    Yao, Yuanshun
    Li, Huiying
    Zheng, Haitao
    Zhao, Ben Y.
    PROCEEDINGS OF THE 2019 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY (CCS'19), 2019, : 2041 - 2055
  • [23] Dynamic process monitoring based on orthogonal dynamic inner neighborhood preserving embedding model
    Chen, Xiaoxia
    Tong, Chudong
    Lan, Ting
    Luo, Lijia
    CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2019, 193
  • [24] Nonlinear data mapping by neural networks
    Duin, RPW
    1996 CERN SCHOOL OF COMPUTING, 1996, 96 (08): : 11 - 15
  • [25] MAXIMUM LIKELIHOOD NONLINEAR TRANSFORMATIONS BASED ON DEEP NEURAL NETWORKS
    Cui, Xiaodong
    Goel, Vaibhava
    2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP), 2015, : 4320 - 4324
  • [26] Maximum Likelihood Nonlinear Transformations Based on Deep Neural Networks
    Cui, Xiaodong
    Goel, Vaibhava
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2016, 24 (11) : 2023 - 2031
  • [27] Monitoring and Classification of Nonlinear Loads based on Artificial Neural Networks
    Stosovic, Miona Andrejevic
    Stevanovic, Dejan
    Dimitrijevic, Marko
    2017 13TH INTERNATIONAL CONFERENCE ON ADVANCED TECHNOLOGIES, SYSTEMS AND SERVICES IN TELECOMMUNICATIONS (TELSIKS), 2017, : 443 - 446
  • [28] Multicriteria Gear Monitoring System Based on Deep Neural Networks
    Lai, Chia-Hung
    Wu, Ting-En
    SENSORS AND MATERIALS, 2023, 35 (12) : 4481 - 4489
  • [29] Gear Grinding Monitoring based on Deep Convolutional Neural Networks
    Liu, Chenyu
    Mauricio, Alexandre
    Chen, Zhuyun
    Declercq, Katrien
    Meerten, Yannick
    Vonderscher, Yann
    Gryllias, Konstantinos
    IFAC PAPERSONLINE, 2020, 53 (02): : 10324 - 10329
  • [30] Embedding Principle of Loss Landscape of Deep Neural Networks
    Zhang, Yaoyu
    Zhang, Zhongwang
    Luo, Tao
    Xu, Zhi-Qin John
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34