Multi-phase LSTM-AE fault monitoring of batch processes based on diffusion distance and entropy FCM

被引：0
|
作者：
Gao X.-J. [1 ,2 ,3 ,4 ]
Li X.-F. [1 ,2 ,3 ,4 ]
Qi Y.-S. [5 ]
机构：
[1] Faculty of Information Technology, Beijing University of Technology, Beijing
[2] Engineering Research Center of Digital Community, Ministry of Education, Beijing
[3] Beijing Laboratory for Urban Mass Transit, Beijing
[4] Beijing Laboratory of Computational Intelligence and Intelligent System, Beijing
[5] School of Electric Power, Inner Mongolia University of Technology, Hohhot
来源：
Gao Xiao Hua Xue Gong Cheng Xue Bao/Journal of Chemical Engineering of Chinese Universities | 2023年 / 37卷 / 01期
关键词：
batch process; diffusion distance; fault monitoring; nonlinearity; phase division;
D O I：
10.3969/j.issn.1003-9015.2023.01.015
中图分类号：
学科分类号：
摘要：
To solve the problem of low fault monitoring accuracy in batch process due to ignoring the nonlinearity of data in phase division, a fault monitoring method for multi-stage LSTM-autoencoder of batch process based on diffusion distance and entropy fuzzy C-means was proposed. Firstly, the information entropy was used to describe the two-dimensional time slice matrix after batch processing for identifying the number of clusters automatically. Then the fuzzy C-means clustering (FCM) was improved by diffusion distance to solve the problem that Euclidean distance could not represent the data nonlinearity, and effectively divided the stable phases of batch process and the transition phases by silhouette coefficient. Finally, a multi - phase LSTM-AE monitoring model was established. The proposed method was verified by using penicillin fermentation data and E. coli production of interleukin-2. The results showed that the proposed method not only improved the performance of phase division, but also could monitor faults more accurately. © 2023 Zhejiang University. All rights reserved.
引用
收藏
页码：120 / 130
页数：10
相关论文
共 32 条
  • [1] YAO Y, GAO F., A survey on multiphase statistical modeling methods for batch processes, Annual Reviews in Control, 33, 2, pp. 172-183, (2009)
  • [2] SIMOGLOU A, GEORGIEVA P, MARTIN E B, Et al., On-line monitoring of a sugar crystallization process, Computers & Chemical Engineering, 29, 6, pp. 1411-1422, (2005)
  • [3] LAWRENCE X Y., Pharmaceutical quality by design: Product and process development, understanding, and control, Pharmaceutical Research, 25, 10, pp. 781-791, (2008)
  • [4] LU N, GAO F, YANG Y, Et al., PCA-based modeling and on-line monitoring strategy for uneven-length batch processes, Industrial & Engineering Chemistry Research, 43, 13, pp. 3343-3352, (2004)
  • [5] ZHANG S M, ZHAO C H, GAO F R., Incipient fault detection for multiphase batch processes with limited batches, IEEE Transactions on Control Systems Technology, 27, 1, pp. 103-117, (2017)
  • [6] JIANG Q C, YAN S F, YAN X F, Et al., Data-driven two-dimensional deep correlated representation learning for nonlinear batch process monitoring, IEEE Transactions on Industrial Informatics, 16, 4, pp. 2839-2848, (2020)
  • [7] RENDALL R, CHIANG L H, REIS M S., Data-driven methods for batch data analysis - A critical overview and mapping on the complexity scale, Computers & Chemical Engineering, 124, pp. 1-13, (2019)
  • [8] ZHU J L, GAO F R., Similar batch process monitoring with orthogonal subspace alignment, IEEE Transactions on Industrial Electronics, 65, 10, pp. 8173-8183, (2018)
  • [9] LUO L K, YAO Y, GAO F R, Et al., Mixed-effects Gaussian process modeling approach with application in injection molding processes, Journal of Process Control, 62, pp. 37-43, (2018)
  • [10] GAO X J, MENG L J, WANG H, Et al., Fault monitoring of batch processes based on substep time-space JITL-MKPLS, Journal of Chemical Engineering of Chinese Universities, 35, 1, pp. 127-139, (2021)
1234