Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction

Zhuang Yu; Zhang Zhongyi; Tao Jin; Li Yi; Li Fan; Wang Yu; Zhang Lei; Du Jian

doi:10.1016/j.cjche.2025.09.016

Your Location：

Home >

Browse articles >

Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction

Updated：2026-03-13

- Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction
- Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction
- 中国化学工程学报（英文） 2026年89卷第1期页码：132-144
- Affiliations：
  
  Institute of Chemical Process Systems Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024, China
  COFCO Biotechnology Co., Ltd., Beijing 100005, China
- Author bio：
  
  Corresponding author. E-mail address: dujian@dlut.edu.cn(J. Du).
- Funds：
  
  the Special Foundation for State Major Basic Research Program of China(2021YFD2101005);National Nat-ural Science Foundation of China.(22478057;22178045)
- DOI：10.1016/j.cjche.2025.09.016
  中图分类号：
- 收稿：2025-06-13，
  
  修回：2025-08-27，
  
  录用：2025-09-01，
  
  网络首发：2025-10-27，
  
  纸质出版：2026-01
- Accepted：
Scan QR Code
Zhuang Yu, Zhang Zhongyi, Tao Jin, 等. Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction[J]. 中国化学工程学报（英文）, 2026,89(1):132-144.

Zhuang Yu, Zhang Zhongyi, Tao Jin, et al. Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction[J]. Chinese Journal of Chemical Engineering, 2026, 89(1): 132-144.
Zhuang Yu, Zhang Zhongyi, Tao Jin, 等. Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction[J]. 中国化学工程学报（英文）, 2026,89(1):132-144. DOI： 10.1016/j.cjche.2025.09.016.

Zhuang Yu, Zhang Zhongyi, Tao Jin, et al. Attention-enhanced multi-time scale LSTM for soft sensor modeling of corn starch liquefaction[J]. Chinese Journal of Chemical Engineering, 2026, 89(1): 132-144. DOI： 10.1016/j.cjche.2025.09.016.

摘要

Abstract

Data-driven deep learning modeling has been increasingly applied to quality prediction in complex chemical processes. However

the data show complex temporal features due to different residence times and strong coupling relationships among chemical entities. This study proposes a multi-scale temporal feature extraction module to extract local dynamic temporal features across different time scales and combines it with long short-term memory (LSTM) networks to capture global temporal patterns

thereby taking full advantage of available data. In addition

variable-wise channel attention is integrated into the model to enhance attention on the essential parts of the feature maps and improve predictive performance. Furthermore

by analyzing the attention weights

the model quickly identifies the key variables that significantly affect the predictions. Finally

the model is applied to a real corn starch liquefaction process and achieves an accurate product quality prediction with an

value of 0.9392

which represents a 4% to 9% improvement over traditional models and demonstrates the superiority of the proposed approach.

关键词

Keywords

references

D.J. Rose, G.E. Inglett, S.X. Liu, Utilisation of corn (Zea mays) bran and corn fiber in the production of food components, J. Sci. Food Agric. 90 (6) (2010) 915—924.

Q.F. Min, Y.G. Lu, Z.Y. Liu, C. Su, B. Wang, Machine learning based digital twin framework for production optimization in petrochemical industry, Int. J. Inf. Manag. 49(2019)502—519.

P.P. Mondal, A. Galodha, V.K. Verma, V. Singh, P.L. Show, M.K. Awasthi, B. Lall, S. Anees, K. Pollmann, R. Jain, Review on machine learning-based bioprocess optimization, monitoring, and control systems, Bioresour. Technol. 370(2023) 128523.

H. Cartwright, S. Curteanu, Neural networks applied in chemistry. II. Neuroevolutionary techniques in process modeling and optimization, Ind. Eng. Chem. Res. 52 (36) (2013) 12673—12688.

M. Kano, Y. Nakagawa, Data-based process monitoring, process control, and quality improvement: recent developments and applications in steel industry, Comput. Chem. Eng. 32 (1—2) (2008) 12—24.

Z.Y. Pang, P. Huang, C. Lian, C. Peng, X.C. Fang, H.L. Liu, Data-driven prediction of product yields and control framework of hydrocracking unit, Chem. Eng. Sci. 283(2024) 119386.

Y.C. Jiang, S. Yin, J.W. Dong, O. Kaynak, A review on soft sensors for monitoring, control, and optimization of industrial processes, IEEE Sens. J. 21 (11) (2020) 12868—12881.

Z.Q. Ge, Z.H. Song, S.X. Ding, B. Huang, Data mining and analytics in the process industry: the role of machine learning, IEEE Access 5 (2017) 20590—20616.

M.R. Dobbelaere, P.P. Plehiers, R. Van de Vijver, C.V. Stevens, K.M. Van Geem, Machine learning in chemical engineering: strengths, weaknesses, opportunities, and threats, Engineering 7 (9) (2021) 1201—1211.

R. Sharmin, U. Sundararaj, S. Shah, L. Vande Griend, Y.J. Sun, Inferential sensors for estimation of polymer quality parameters: industrial application of a PLS-based soft sensor for a LDPE plant, Chem. Eng. Sci. 61 (19) (2006) 6372—6384.

E. Marengo, M. Bobba, E. Robotti, M.C. Liparota, Modeling of the polluting emissions from a cement production plant by partial least-squares, principal component regression, and artificial neural networks, Environ. Sci. Technol. 40 (1)(2006)272—280.

Z. Zhou, C. Qiu, Y.F. Zhang, A comparative analysis of linear regression, neural networks and random forest regression for predicting air ozone employing soft sensor models, Sci. Rep. 13(1) (2023)22420.

N. Parveen, S. Zaidi, M. Danish, Development of SVR-based model and comparative analysis with MLR and ANN models for predicting the sorption capacity of Cr(VI), Process. Saf. Environ. Prot. 107(2017)428—437.

M.M. Zhang, X.G. Liu, A soft sensor based on adaptive fuzzy neural network and support vector regression for industrial melt index prediction, Chemometr. Intell. Lab. Syst. 126 (2013) 83—90.

Y.R. Xu, X.H. Zeng, S. Bernard, Z. He, Data-driven prediction of neutralizer pH and valve position towards precise control of chemical dosage in a wastewater treatment plant, J. Clean. Prod. 348 (2022) 131360.

J.C.B. Gonzaga, L.A.C. Meleiro, C. Kiang, R. Maciel Filho, ANN-based softsensor for real-time process monitoring and control of an industrial polymerization process, Comput. Chem. Eng. 33 (1) (2009) 43—49.

Y.L. He, Z.Q. Geng, Q.X. Zhu, Data driven soft sensor development for complex chemical processes using extreme learning machine, Chem. Eng. Res. Des. 102 (2015)1—11.

C. Shang, F. Yang, D.X. Huang, W.X. Lyu, Data-driven soft sensor development based on deep learning technique, J. Process Control 24 (3)(2014)223—233.

R.M. Xie, N.M. Jan, K.R. Hao, L. Chen, B. Huang, Supervised variational autoencoders for soft sensor modeling with missing data, IEEE Trans. Ind. Inf. 16(4)(2020)2820—2828.

X.F. Yuan, C. Ou, Y.L. Wang, C.H. Yang, W.H. Gui, A novel semi-supervised pretraining strategy for deep networks and its application for quality variable prediction in industrial processes, Chem. Eng. Sci. 217 (2020) 115509.

Y.M. Ha n, Y. Wang, Z.W. Chen, Y. Lu, X. Hu, L.C. Chen, Z.Q. Geng, Multiscale variational autoencoder regressor for production prediction and energy saving of industrial processes, Chem. Eng. Sci. 284(2024)119529.

K.C. Wang, C. Shang, L. Liu, Y.H. Jiang, D.X. Huang, F. Yang, Dynamic soft sensor development based on convolutional neural networks, Ind. Eng. Chem. Res. 58 (26)(2019) 11521—11531.

Y.M. Han, C.Y. Fan, M. Xu, Z.Q. Geng, Y.H. Zhong, Production capacity analysis and energy saving of complex chemical processes using LSTM based on attention mechanism, Appl. Therm. Eng. 160(2019)114072.

W.S. Ke, D.X. Huang, F. Yang, Y.H. Jiang, Soft sensor development and applications based on LSTM in deep neural networks, in: 2017 IEEE Symposium Series on Computational Intelligence (SSCI). Honolulu, HI, USA, IEEE, 2017.

X.F. Yuan, L. Li, Y.A.W. Shardt, Y.L. Wang, C.H. Yang, Deep learning with spatiotemporal attention-based LSTM for industrial soft sensor model development, IEEE Trans. Ind. Electron. 68(5)(2021)4404—4414.

X.F. Yuan, S.B. Qi, Y.L. Wang, K. Wang, C.H. Yang, L.J. Ye, Quality variable prediction for nonlinear dynamic industrial processes based on temporal convolutional networks, IEEE Sens. J. 21 (18)(2021) 20493—20503.

Y.M. Bai, J.S. Zhao, A novel transformer-based multi-variable multi-step prediction method for chemical process fault prognosis, Process. Saf. Environ. Prot. 169 (2023) 937—947.

J. Hong, W.D. Tian, Prediction in c atalytic cracking process based on swarm intelligence algorithm optimization of LSTM, Processes 11(5)(2023)1454.

W.S. Zha, Y.P. Liu, Y.J. Wan, R.L. Luo, D.L. Li, S. Yang, Y.M. Xu, Forecasting monthly gas field production based on the CNN-LSTM model, Energy 260 (2022)124889.

X.F. Yuan, L.F. Huang, L.J. Ye, Y.L. Wang, K. Wang, C.H. Yang, W.H. Gui, F.F. Shen, Quality prediction modeling for industrial processes using multiscale attention-based convolutional neural network, IEEE Trans. Cybern. 54 (5) (2024)2696—2707.

J.H. Wang, G.F. Lin, M.J. Chang, I.H. Huang, Y.R. Chen, Real-time water-level forecasting using dilated causal convolutional neural networks, Water Resour. Manag. 33(11)(2019)3759—3780.

X.T. Bi, J.S. Zhao, A novel orthogonal self-attentive variational autoencoder method for interpretable chemical process fault detection and identification, Process. Saf. Environ. Prot. 156(2021)581—597.

Y.J. Wang, C. Qian, S.J. Qin, Attention-mechanism based DiPLS-LSTM and its application in industrial process time series big data prediction, Comput. Chem. Eng. 176(2023)108296.

Z.Y. Yang, K. Wang, L.J. Ye, X.F. Yuan, Y.L. Wang, C.H. Yang, W.H. Gui, A difference metric attention with position distance-based weighting for transformer in data sequence modeling of industrial processes, IEEE Trans. Ind. Inf. 21 (2)(2025)1803—1812.

X.F. Yuan, N. Xu, L.J. Ye, K. Wang, F.F. Shen, Y.L. Wang, C.H. Yang, W.H. Gui, Attention- based interval aided networks for data modeling of heterogeneous sampling sequences with missing values in process industry, IEEE Trans. Ind. Inf. 20(4)(2024)5253—5262.

X.F. Yuan, L. Li, Y.L. Wang, C.H. Yang, W.H. Gui, Deep learning for quality prediction of nonlinear dynamic processes with variable attention-based long short-term memory network, Can. J. Chem. Eng. 98 (6) (2020) 1377—1389.

Y. Tong, M. Shu, M.X. Li, Y.W. Liu, R. Tao, C.C. Zhou, Y. Zhao, G.X. Zhao, Y. Li, Y. C. Dong, L. Zhang, L.L. Liu, J. Du, A neural network-based production process modeling and variable importance analysis approach in corn to sugar factory, Front. Chem. Sci. Eng. 17(3)(2023)358—371.

M.P. Maples, D.E. Reichart, N.C. Konz, T.A. Berger, A.S. Trotter, J.R. Martin, D. A. Dutton, M.L. Paggen, R.E. Joyner, C.P. Salemi, Robust chauvenet outlier rejection, Astrophys. J. Suppl. 238(1)(2018)2.

S. Xu, B. Lu, M. Baldea, T.F. Edgar, W. Wojsznis, T. Blevins, M. Nixon, Data cleaning in the process industries, Rev. Chem. Eng. 31 (5) (2015) 453—490.

S. Morales, H. A'lvarez, C. Sa'nchez, Dynamic models for the production of glucose syrups from cassava starch, Food Bioprod. Process. 86 (1) (2008) 25—30.

C.M. Li, D. Fang, Z.F. Li, Z.B. Gu, Q.W. Yang, L. Cheng, Y. Hong, An improved twostep saccharification of high-concentration corn starch slurries by granular starch hydrolyzing enzyme, Ind. Crops Prod. 94 (2016) 259—265.

J.G. Monroy, E.J. P alomo, E. Lo'pez-Rubio, J. Gonzalez-Jimenez, Continuous chemical classification in uncontrolled environments with sliding windows, Chemometr. Intell. Lab. Syst. 158(2016)117—129.

J. Wu, X.Y. Chen, H. Zhang, L.D. Xiong, H. Lei, S.H. Deng, Hyperparameter optimization for machine learning models based on Bayesian optimization, J. Electron. Sci. Technol. 17 (2019)26—40.

T. Akiba, S. Sano, T. Yanase, T. Ohta, M. Koyama, Optuna: a next-generation hyperparameter optimization framework, in: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Anchorage AK USA, ACM, 2019.

W.W. Guo, J.L. Zhu, X.Y. Yu, M.W. Jia, Y. Liu, Temporal graph convolutional network soft sensor for molecular weight distribution prediction, Chemometr. Intell. Lab. Syst. 252 (2024) 105196.

X.Y. Lin, Z.H. Li, Y.M. Han, Z.W. Chen, Z.Q. Geng, Novel spatiotemporal graph attention model for production prediction and energy structure optimization of propylene production processes, Comput. Chem. Eng. 181(2024)108507.

Y. Wang, F.F. Shen, L.J. Ye, A knowledge-refined hybrid graph model for quality prediction of industrial processes, Eng. Appl. Artif. Intell. 139 (2025) 109711.

浏览量

Downloads

CSCD

文章被引用时，请邮件提醒。

Submit

工具集

关联资源

From data to insight: Building a knowledge graph for risk analysis of hazardous chemical accidents

Quality related fault detection based on dynamic-inner convolutional autoencoder and partial least squares and its application to ironmaking process

SmdaNet: A hierarchical hard sample mining and domain adaptation neural network for fault diagnosis in industrial process

Large language model-based multi-objective modeling framework for vacuum gas oil hydrotreating

Process fault root cause diagnosis through state evolution mapping based on temporal unit shapelets