A systematic data-driven modelling framework for nonlinear distillation processes incorporating data intervals clustering and new integrated learning algorithm

doi:10.1016/j.cjche.2025.02.013

Chinese Journal of Chemical Engineering ›› 2025, Vol. 81 ›› Issue (5): 182-199.DOI: 10.1016/j.cjche.2025.02.013

Previous Articles Next Articles

A systematic data-driven modelling framework for nonlinear distillation processes incorporating data intervals clustering and new integrated learning algorithm

Zhe Wang¹, Renchu He², Jian Long¹

1. Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
2. Department of Automation, China University of Petroleum, Beijing 102249, China

Received:2024-11-11 Revised:2025-02-17 Accepted:2025-02-20 Online:2025-03-08 Published:2025-05-28
Contact: Renchu He,E-mail:rche@cup.edu.cn;Jian Long,E-mail:longjian@ecust.edu.cn
Supported by:
This work was supported by the National Key Research and Development Program of China (2023YFB3307801), the National Natural Science Foundation of China (62394343, 62373155,62073142), Major Science and Technology Project of Xinjiang (No. 2022A01006-4), the Programme of Introducing Talents of Discipline to Universities (the 111 Project) under Grant B17017, the Fundamental Research Funds for the Central Universities, Science Foundation of China University of Petroleum, Beijing (No. 2462024YJRC011) and the Open Research Project of the State Key Laboratory of Industrial Control Technology, China (Grant No. ICT2024B70).

A systematic data-driven modelling framework for nonlinear distillation processes incorporating data intervals clustering and new integrated learning algorithm

Zhe Wang¹, Renchu He², Jian Long¹

1. Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
2. Department of Automation, China University of Petroleum, Beijing 102249, China

通讯作者: Renchu He,E-mail:rche@cup.edu.cn;Jian Long,E-mail:longjian@ecust.edu.cn
基金资助:
This work was supported by the National Key Research and Development Program of China (2023YFB3307801), the National Natural Science Foundation of China (62394343, 62373155,62073142), Major Science and Technology Project of Xinjiang (No. 2022A01006-4), the Programme of Introducing Talents of Discipline to Universities (the 111 Project) under Grant B17017, the Fundamental Research Funds for the Central Universities, Science Foundation of China University of Petroleum, Beijing (No. 2462024YJRC011) and the Open Research Project of the State Key Laboratory of Industrial Control Technology, China (Grant No. ICT2024B70).

Abstract

Abstract: The distillation process is an important chemical process, and the application of data-driven modelling approach has the potential to reduce model complexity compared to mechanistic modelling, thus improving the efficiency of process optimization or monitoring studies. However, the distillation process is highly nonlinear and has multiple uncertainty perturbation intervals, which brings challenges to accurate data-driven modelling of distillation processes. This paper proposes a systematic data-driven modelling framework to solve these problems. Firstly, data segment variance was introduced into the K-means algorithm to form K-means data interval (KMDI) clustering in order to cluster the data into perturbed and steady state intervals for steady-state data extraction. Secondly, maximal information coefficient (MIC) was employed to calculate the nonlinear correlation between variables for removing redundant features. Finally, extreme gradient boosting (XGBoost) was integrated as the basic learner into adaptive boosting (AdaBoost) with the error threshold (ET) set to improve weights update strategy to construct the new integrated learning algorithm, XGBoost-AdaBoost-ET. The superiority of the proposed framework is verified by applying this data-driven modelling framework to a real industrial process of propylene distillation.

Key words: Integrated learning algorithm, Data intervals clustering, Feature selection, Application of artificial intelligence in distillation industry, Data-driven modelling

摘要： The distillation process is an important chemical process, and the application of data-driven modelling approach has the potential to reduce model complexity compared to mechanistic modelling, thus improving the efficiency of process optimization or monitoring studies. However, the distillation process is highly nonlinear and has multiple uncertainty perturbation intervals, which brings challenges to accurate data-driven modelling of distillation processes. This paper proposes a systematic data-driven modelling framework to solve these problems. Firstly, data segment variance was introduced into the K-means algorithm to form K-means data interval (KMDI) clustering in order to cluster the data into perturbed and steady state intervals for steady-state data extraction. Secondly, maximal information coefficient (MIC) was employed to calculate the nonlinear correlation between variables for removing redundant features. Finally, extreme gradient boosting (XGBoost) was integrated as the basic learner into adaptive boosting (AdaBoost) with the error threshold (ET) set to improve weights update strategy to construct the new integrated learning algorithm, XGBoost-AdaBoost-ET. The superiority of the proposed framework is verified by applying this data-driven modelling framework to a real industrial process of propylene distillation.

关键词: Integrated learning algorithm, Data intervals clustering, Feature selection, Application of artificial intelligence in distillation industry, Data-driven modelling

Zhe Wang, Renchu He, Jian Long. A systematic data-driven modelling framework for nonlinear distillation processes incorporating data intervals clustering and new integrated learning algorithm[J]. Chinese Journal of Chemical Engineering, 2025, 81(5): 182-199.

References

[1] A. Tgarguifa, S. Abderafi, T. Bounahmidi, Energetic optimization of Moroccan distillery using simulation and response surface methodology, Renew. Sustain. Energy Rev. 75 (2017) 415-425.
[2] A.K. Jana, Advances in heat pump assisted distillation column: a review, Energy Convers. Manag. 77 (2014) 287-297.
[3] N. Asprion, Modeling, simulation, and optimization 4.0 for a distillation column, Chem. Ing. Tech. 92 (7) (2020) 879-889.
[4] M. Xu, F. Xin, X.F. Li, X.L. Huai, J.F. Guo, H. Liu, Equilibrium model and performances of an isopropanol-acetone-hydrogen chemical heat pump with a reactive distillation column, Ind. Eng. Chem. Res. 52 (11) (2013) 4040-4048.
[5] X.G. Liu, L. Cong, Y.X. Zhou, Nonlinear model predictive control based on wave model of high-purity internal thermally coupled distillation columns, Ind. Eng. Chem. Res. 52 (19) (2013) 6470-6479.
[6] S. Zendehboudi, N. Rezaei, A. Lohi, Applications of hybrid models in chemical, petroleum, and energy systems: a systematic review, Appl. Energy 228 (2018) 2539-2566.
[7] J.A. Ramirez, R.J. Brown, T.J. Rainey, Liquefaction biocrudes and their petroleum crude blends for processing in conventional distillation units, Fuel Process. Technol. 167 (2017) 674-683.
[8] M. Chen, P.L. Li, C. Hu, J.H. Luo, M. Wen, B. Yu, F. Jiang, Y.T. An, Y. Shi, J.F. Song, W.H. Luo, Dynamic simulation of a cryogenic batch distillation process for hydrogen isotopes separation, Int. J. Hydrog. Energy 47 (23) (2022) 11955-11961.
[9] W.W. Li, B. Song, X.P. Li, Y.Z. Liu, Modelling of vacuum distillation in a rotating packed bed by Aspen, Appl. Therm. Eng. 117 (2017) 322-329.
[10] Z.S. Hou, R.H. Chi, H.J. Gao, An overview of dynamic-linearization-based data-driven control and applications, IEEE Trans. Ind. Electron. 64 (5) (2017) 4076-4090.
[11] Z.Q. Ge, Review on data-driven modeling and monitoring for plant-wide industrial processes, Chemom. Intell. Lab. Syst. 171 (2017) 16-25.
[12] P. Behnam, M. Faegh, M. Khiadani, A review on state-of-the-art applications of data-driven methods in desalination systems, Desalination 532 (2022) 115744.
[13] Z.L. Chen, Z. O’Neill, J. Wen, O. Pradhan, T. Yang, X. Lu, G.J. Lin, S. Miyata, S. Lee, C. Shen, R. Chiosa, M.S. Piscitelli, A. Capozzoli, F. Hengel, A. Kuhrer, M. Pritoni, W. Liu, J. Clauss, Y.M. Chen, T. Herr, A review of data-driven fault detection and diagnostics for building HVAC systems, Appl. Energy 339 (2023) 121030.
[14] Z.T. Shao, Q.Z. Zhai, X.H. Guan, Physical-model-aided data-driven linear power flow model: an approach to address missing training data, IEEE Trans. Power Syst. 38 (3) (2023) 2970-2973.
[15] G. Peters, F. Crespo, P. Lingras, R. Weber, Soft clustering-Fuzzy and rough approaches and their extensions and derivatives, Int. J. Approx. Reason. 54 (2) (2013) 307-322.
[16] K. Jiang, M. Kheradmandi, C. Hu, S. Pal, F.J. Yan, Data-driven fault tolerant predictive control for temperature regulation in data center with rack-based cooling architecture, Mechatronics 79 (2021) 102633.
[17] T.H. Pan, S.S. Shieh, S.S. Jang, W.H. Tseng, C.W. Wu, J.J. Ou, Statistical multi-model approach for performance assessment of cooling tower, Energy Convers. Manag. 52 (2) (2011) 1377-1385.
[18] Y. Huang, G.P. Liu, W.S. Hu, Priori-guided and data-driven hybrid model for wind power forecasting, ISA Trans. 134 (2023) 380-395.
[19] S. Zeng, Z.Y. Wang, R. Huang, L. Chen, D. Feng, A study on multi-kernel intuitionistic fuzzy C-means clustering with multiple attributes, Neurocomputing 335 (2019) 59-71.
[20] R.Y. Zhang, Z. Chen, S. Chen, J.W. Zheng, O. Buyukozturk, H. Sun, Deep long short-term memory networks for nonlinear structural seismic response prediction, Comput. Struct. 220 (2019) 55-68.
[21] Y. Choi, N. An, S. Hong, H. Cho, J. Lim, I.S. Han, I. Moon, J. Kim, Time-series clustering approach for training data selection of a data-driven predictive model: Application to an industrial bio 2, 3-butanediol distillation process, Comput. Chem. Eng. 161 (2022) 107758.
[22] Q.F. Chen, M.C. Xia, Y. Zhou, H.M. Cai, J.Z. Wu, H.B. Zhang, Optimal planning for partially self-sufficient microgrid with limited annual electricity exchange with distribution grid, IEEE Access 7 (2019) 123505-123520.
[23] J.H. Wang, Q.X. Zhang, S.S. Li, X.X. Pan, K. Chen, C. Zhang, Z. Wang, M.S. Jia, Short-term power forecasting of fishing-solar complementary photovoltaic power station based on a data-driven model, Energy Rep. 10 (2023) 1851-1863.
[24] Y. Choi, B. Bhadriaju, H. Cho, J. Lim, I.S. Han, I. Moon, J.S. Kwon, J. Kim, Data-driven modeling of multimode chemical process: Validation with a real-world distillation column, Chem. Eng. J. 457 (2023) 141025.
[25] Y.W. Zhao, Z.P. Wang, Z.Y. Sun, P. Liu, D.S. Cui, J.J. Deng, Data-driven lithium-ion battery degradation evaluation under overcharge cycling conditions, IEEE Trans. Power Electron. 38 (8) (2023) 10138-10150.
[26] L.G. Maltais, L. Gosselin, Forecasting of short-term lighting and plug load electricity consumption in single residential units: Development and assessment of data-driven models for different horizons, Appl. Energy 307 (2022) 118229.
[27] Q.L. Wang, J.Q. Han, F. Chen, S. Hu, C. Yun, Z. Dou, T.J. Yan, G.A. Yang, Modeling risk characterization networks for chemical processes based on multi-variate data, Energy 293 (2024) 130689.
[28] J.Y. Ye, A. Bab-Hadiashar, R. Hoseinnezhad, N. Alam, A. Vargas-Uscategui, M. Patel, I. Cole, Predictions of in situ melt pool geometric signatures via machine learning techniques for laser metal deposition, int j comput integr manuf 36 (9) (2023) 1345-1361.
[29] J.E. Cote-Ballesteros, V.H. Grisales Palacios, J.E. Rodriguez-Castellanos, A hybrid approach variable selection algorithm based on mutual information for data-driven industrial soft-sensor applications, Cien.Ing.Neogranadina 32 (1) (2022) 59-70.
[30] Q.C. Jiang, X.F. Yan, Nonlinear plant-wide process monitoring using MI-spectral clustering and Bayesian inference-based multiblock KPCA, J. Process. Contr. 32 (2015) 38-50.
[31] J. Wang, X.F. Yan, Mutual information-weighted principle components identified from the depth features of stacked autoencoders and original variables for oil dry point soft sensor, IEEE Access 7 (2018) 1981-1990.
[32] W.J. Li, H.Y. Fang, G.X. Qin, X.Q. Tan, Z.W. Huang, F.T. Zeng, H.W. Du, S.P. Li, Concentration estimation of dissolved oxygen in Pearl River Basin using input variable selection and machine learning techniques, Sci. Total Environ. 731 (2020) 139099.
[33] L. Wang, Y.G. He, X.Y. Liu, L. Li, K.X. Shao, M2TNet: Multi-modal multi-task Transformer network for ultra-short-term wind power multi-step forecasting, Energy Rep. 8 (2022) 7628-7642.
[34] D.N. Reshef, Y.A. Reshef, H.K. Finucane, S.R. Grossman, G. McVean, P.J. Turnbaugh, E.S. Lander, M. Mitzenmacher, P.C. Sabeti, Detecting novel associations in large data sets, Science 334 (6062) (2011) 1518-1524.
[35] C.K. Li, D.F. Zhao, S.J. Mu, W.H. Zhang, N. Shi, L.N. Li, Fault diagnosis for distillation process based on CNN-DAE, Chin. J. Chem. Eng. 27 (3) (2019) 598-604.
[36] 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.
[37] M. Behnasr, H. Jazayeri-Rad, Robust data-driven soft sensor based on iteratively weighted least squares support vector regression optimized by the cuckoo optimization algorithm, J. Nat. Gas Sci. Eng. 22 (2015) 35-41.
[38] X. Qian, S.K. Jia, K.J. Huang, H.S. Chen, Y. Yuan, X.G. Yuan, L. Zhang, MPC-PI cascade control for the Kaibel dividing wall column integrated with data-driven soft sensor model, Chem. Eng. Sci. 231 (2021) 116240.
[39] Z. Yan, H. Chen, X. Dong, K. Zhou, Z. Xu, Research on prediction of multi-class theft crimes by an optimized decomposition and fusion method based on XGBoost, Expert Systems with Applications 207 (2022) 117943.
[40] D.C. Feng, Z.T. Liu, X.D. Wang, Y. Chen, J.Q. Chang, D.F. Wei, Z.M. Jiang, Machine learning-based compressive strength prediction for concrete: an adaptive boosting approach, Constr. Build. Mater. 230 (2020) 117000.
[41] Z. Olujic, L. Sun, A. de Rijke, P.J. Jansens, Conceptual design of an internally heat integrated propylene-propane splitter, Energy 31 (15) (2006) 3083-3096.
[42] B. Liao, Z.G. Lei, Z. Xu, R.Q. Zhou, Z.T. Duan, New process for separating propylene and propane by extractive distillation with aqueous acetonitrile, Chem. Eng. J. 84 (3) (2001) 581-586.
[43] R.P. Sheridan, W.M. Wang, A. Liaw, J. Ma, E.M. Gifford, Extreme gradient boosting as a method for quantitative structure-activity relationships, J. Chem. Inf. Model. 56 (12) (2016) 2353-2360.
[44] X. Fei, K.L. He, Y.X. Huang, J.P. Tian, X.J. Hu, Y. Liang, X.Q. Yi, L.L. Xie, D. Huang, The rapid determination of the fatty acid content of rice by combining hyperspectral imaging and integrated learning models, Vib. Spectrosc. 129 (2023) 103609.
[45] Z. Zhao, H.G. Nan, Z.H. Liu, Y.B. Yu, Multi-step interval prediction of ultra-short-term wind power based on CEEMDAN-FIG and CNN-BiLSTM, Environ. Sci. Pollut. Res. Int. 29 (38) (2022) 58097-58109.
[46] J. Tang, Y.J. Li, M.F. Ding, H. Liu, D.P. Yang, X.Q. Wu, An ionospheric TEC forecasting model based on a CNN-LSTM-attention mechanism neural network, Remote. Sens. 14 (10) (2022) 2433.
[47] X.D. Hu, P.L. Zhang, Q. Zhang, A Novel Framework of CNN Integrated with Adaboost for Remote Sensing Scene ClassificationIGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium. September 26-October 2, 2020. Waikoloa, HI, USA. IEEE, 2020.
[48] T.M. Zhang, G.W. Yan, R. Li, S.Y. Xiao, M.F. Ren, L. Cheng, An online transfer kernel recursive algorithm for soft sensor modeling with variable working conditions, Contr. Eng. Pract. 141 (2023) 105726.
[49] Y.W. Chen, X.M. Zhang, Z.H. Song, M. Kano, Multivariate deep reconstruction neural network for multi-step-ahead prediction of industrial process quality variables, IFAC-PapersOnLine 56 (2) (2023) 2852-2857.
[50] Y.F. Chen, A.L. Li, X.Y. Li, D. Xue, J. Long, Efficient JITL framework for nonlinear industrial chemical engineering soft sensing based on adaptive multi-branch variable scale integrated convolutional neural networks, Adv. Eng. Inform. 58 (2023) 102199.

A systematic data-driven modelling framework for nonlinear distillation processes incorporating data intervals clustering and new integrated learning algorithm

A systematic data-driven modelling framework for nonlinear distillation processes incorporating data intervals clustering and new integrated learning algorithm

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 1

Recommended Articles

Metrics

Comments