Graph convolutional network for axial concentration profiles prediction in simulated moving bed

doi:10.1016/j.cjche.2024.05.029

中国化学工程学报 ›› 2024, Vol. 73 ›› Issue (9): 270-280.DOI: 10.1016/j.cjche.2024.05.029

Graph convolutional network for axial concentration profiles prediction in simulated moving bed

Can Ding, Minglei Yang, Yunmeng Zhao, Wenli Du

Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China

收稿日期:2023-10-12 修回日期:2024-05-25 接受日期:2024-05-26 出版日期:2024-11-21 发布日期:2024-07-16
通讯作者: Yunmeng Zhao,E-mail:yunmeng.zhao@ecust.edu.cn;Wenli Du,E-mail:wldu@ecust.edu.cn
基金资助:
This work was supported by the National Key Research and Development Program of China (2022YFB3305900), National Natural Science Foundation of China (62293501, 62394343), the Shanghai Committee of Science and Technology, China (22DZ1101500), Major Program of Qingyuan Innovation Laboratory (00122002), Fundamental Research Funds for the Central Universities (222202417006) and Shanghai AI Lab.

Graph convolutional network for axial concentration profiles prediction in simulated moving bed

Can Ding, Minglei Yang, Yunmeng Zhao, Wenli Du

Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China

Received:2023-10-12 Revised:2024-05-25 Accepted:2024-05-26 Online:2024-11-21 Published:2024-07-16
Contact: Yunmeng Zhao,E-mail:yunmeng.zhao@ecust.edu.cn;Wenli Du,E-mail:wldu@ecust.edu.cn
Supported by:
This work was supported by the National Key Research and Development Program of China (2022YFB3305900), National Natural Science Foundation of China (62293501, 62394343), the Shanghai Committee of Science and Technology, China (22DZ1101500), Major Program of Qingyuan Innovation Laboratory (00122002), Fundamental Research Funds for the Central Universities (222202417006) and Shanghai AI Lab.

摘要/Abstract

摘要： The simulated moving bed (SMB) chromatographic separation is a continuous compound separation process based on the differences in adsorption capacity exhibited by distinct constituents of a mixture on the fluid phase and stationary phase. The prediction of axial concentration profiles along the beds in a unit is crucial for the operating optimization of SMB. Though the correlation shared by operating variables of SMB has an enormous impact on the operational state of the device, these correlations have been long overlooked, especially by the data-driven models. This study proposes an operating variable-based graph convolutional network (OV-GCN) to enclose the underrepresented correlations and precisely predict axial concentration profiles prediction in SMB. The OV-GCN estimates operating variables with the Spearman correlation coefficient and incorporates them in the adjacency matrix of a graph convolutional network for information propagation and feature extraction. Compared with Random Forest, K-Nearest Neighbors, Support Vector Regression, and Backpropagation Neural Network, the values of the three performance evaluation metrics, namely MAE, RMSE, and R², indicate that OV-GCN has better prediction accuracy in predicting five essential aromatic compounds' axial concentration profiles of an SMB for separating p-xylene (PX). In addition, the OV-GCN method demonstrates a remarkable ability to provide high-precision and fast predictions in three industrial case studies. With the goal of simultaneously maximizing PX purity and yield, we employ the non-dominated sorting genetic algorithm-II optimization method to perform multi-objective optimization of the PX purity and yield. The outcome suggests a promising approach to extracting and representing correlations among operating variables in data-driven process modeling.

关键词: Chromatography, Prediction, Operating variables, Graph convolutional network, Optimization

Abstract: The simulated moving bed (SMB) chromatographic separation is a continuous compound separation process based on the differences in adsorption capacity exhibited by distinct constituents of a mixture on the fluid phase and stationary phase. The prediction of axial concentration profiles along the beds in a unit is crucial for the operating optimization of SMB. Though the correlation shared by operating variables of SMB has an enormous impact on the operational state of the device, these correlations have been long overlooked, especially by the data-driven models. This study proposes an operating variable-based graph convolutional network (OV-GCN) to enclose the underrepresented correlations and precisely predict axial concentration profiles prediction in SMB. The OV-GCN estimates operating variables with the Spearman correlation coefficient and incorporates them in the adjacency matrix of a graph convolutional network for information propagation and feature extraction. Compared with Random Forest, K-Nearest Neighbors, Support Vector Regression, and Backpropagation Neural Network, the values of the three performance evaluation metrics, namely MAE, RMSE, and R², indicate that OV-GCN has better prediction accuracy in predicting five essential aromatic compounds' axial concentration profiles of an SMB for separating p-xylene (PX). In addition, the OV-GCN method demonstrates a remarkable ability to provide high-precision and fast predictions in three industrial case studies. With the goal of simultaneously maximizing PX purity and yield, we employ the non-dominated sorting genetic algorithm-II optimization method to perform multi-objective optimization of the PX purity and yield. The outcome suggests a promising approach to extracting and representing correlations among operating variables in data-driven process modeling.

Key words: Chromatography, Prediction, Operating variables, Graph convolutional network, Optimization

Can Ding, Minglei Yang, Yunmeng Zhao, Wenli Du. Graph convolutional network for axial concentration profiles prediction in simulated moving bed[J]. 中国化学工程学报, 2024, 73(9): 270-280.

Can Ding, Minglei Yang, Yunmeng Zhao, Wenli Du. Graph convolutional network for axial concentration profiles prediction in simulated moving bed[J]. Chinese Journal of Chemical Engineering, 2024, 73(9): 270-280.

参考文献

[1] C. Migliorini, M. Mazzotti, M. Morbidelli, Continuous chromatographic separation through simulated moving beds under linear and nonlinear conditions, J. Chromatogr. A 827 (2) (1998) 161-173.
[2] M. Minceva, A.E. Rodrigues, Modeling and simulation of a simulated moving bed for the separation of p-xylene, Ind. Eng. Chem. Res. 41 (14) (2002) 3454-3461.
[3] M. Minceva, A.E. Rodrigues, Influence of the transfer line dead volume on the performance of an industrial scale simulated moving bed for p-xylene separation, Sep. Sci. Technol. 38 (7) (2003) 1463-1497.
[4] P. Suvarov, J.W. Lee, A. Vande Wouwer, A. Seidel-Morgenstern, A. Kienle, Online estimation of optimal operating conditions for simulated moving bed chromatographic processes, J. Chromatogr. A 1602 (2019) 266-272.
[5] J. Strube, U. Altenhoner, M. Meurer, H. Schmidt-Traub, M. Schulte, Dynamic simulation of simulated moving-bed chromatographic processes for the optimization of chiral separations, J. Chromatogr. A 769 (1) (1997) 81-92.
[6] M, Minceva, A.E. Rodrigues, Two-level optimization of an existing SMB for p-xylene separation, Comput. Chem. Eng., 29(10) 2005 2215-2228.
[7] C.Y. Yao, Z.W. Zheng, K. Chen, L.W. Wei, L. Shen, Y.H. Lu, E.G. Fan, Using a machine learning model for the optimal design of simulated moving bed processes and its application to separate rebaudioside A and stevioside, J. Chem. Technol. Biotechnol. 96 (9) (2021) 2558-2568.
[8] W.S. Lee, C.H. Lee, Dynamic modeling and machine learning of commercial-scale simulated moving bed chromatography for application to multi-component normal paraffin separation, Sep. Purif. Technol. 288 (2022) 120597.
[9] I.B.R. Nogueira, A.M. Ribeiro, R. Requiao, K.V. Pontes, H. Koivisto, A.E. Rodrigues, J.M. Loureiro, A quasi-virtual online analyser based on an artificial neural networks and offline measurements to predict purities of raffinate/extract in simulated moving bed processes, Appl. Soft Comput. 67 (2018) 29-47.
[10] C.Y. Wang, K.U. Klatt, G. Dunnebier, S. Engell, F. Hanisch, Neural network-based identification of SMB chromatographic processes, Contr. Eng. Pract. 11 (8) (2003) 949-959.
[11] P. Suvarov, A. Kienle, C. Nobre, G. De Weireld, A. Vande Wouwer, Cycle to cycle adaptive control of simulated moving bed chromatographic separation processes, J. Process. Contr. 24 (2) (2014) 357-367.
[12] Z.W. Zhang, P. Cui, W.W. Zhu, Deep learning on graphs: a survey, IEEE Trans. Knowl. Data Eng. 34 (1) (2022) 249-270.
[13] Z.H. Wu, S.R. Pan, F.W. Chen, G.D. Long, C.Q. Zhang, P.S. Yu, A comprehensive survey on graph neural networks, IEEE Trans. Neural Netw. Learn. Syst. 32 (1) (2021) 4-24.
[14] Z.W. Wang, M. Xia, M. Lu, L.L. Pan, J. Liu, Parameter identification in power transmission systems based on graph convolution network, IEEE Trans. Power Deliv. 37 (4) (2022) 3155-3163.
[15] Y. Ding, Z.L. Zhang, X.F. Zhao, D.F. Hong, W. Cai, C.G. Yu, N.J. Yang, W.W. Cai, Multi-feature fusion: Graph neural network and CNN combining for hyperspectral image classification, Neurocomputing 501 (2022) 246-257.
[16] L. Zhao, Y.J. Song, C. Zhang, Y. Liu, P. Wang, T. Lin, M. Deng, H.F. Li, T-GCN: a temporal graph convolutional network for traffic prediction, IEEE Trans. Intell. Transp. Syst. 21 (9) (2020) 3848-3858.
[17] Y. Huang, C. Zhang, J. Yella, S. Petrov, X.Y. Qian, Y.F. Tang, X.Q. Zhu, S. Bom, GraSSNet: graph soft sensing neural networks, 2021 IEEE International Conference on Big Data (Big Data). Orlando, FL, USA. IEEE, (2021) 746-756.
[18] M.W. Jia, Y. Dai, D.Y. Xu, T. Yang, Y. Yao, Y. Liu, Deep graph network for process soft sensor development, 2021 8th International Conference on Information, Cybernetics, and Computational Social Systems (ICCSS). Beijing, China. IEEE, (2021) 1-6.
[19] Y.H. Wang, P.F. Yan, M.H. Gai, Dynamic soft sensor for anaerobic digestion of kitchen waste based on SGSTGAT, IEEE Sens. J. 21 (17) (2021) 19198-19208.
[20] K. Zhu, C.H. Zhao, Dynamic graph-based adaptive learning for online industrial soft sensor with mutable spatial coupling relations, IEEE Trans. Ind. Electron. 70 (9) (2023) 9614-9622.
[21] J.X. Ye, J.J. Zhao, K.J. Ye, C.Z. Xu, Multi-graph convolutional network for relationship-driven stock movement prediction, 2020 25th International Conference on Pattern Recognition (ICPR). Milan, Italy. IEEE, (2021) 6702-6709.
[22] Z. Al Ani, A.M. Gujarathi, A.H. Al-Muhtaseb, A state of art review on applications of multi-objective evolutionary algorithms in chemicals production reactors, Artif. Intell. Rev. 56 (3) (2023) 2435-2496.
[23] Z.Q. Wang, L. Wang, Z.H. Yuan, B.Z. Chen, Data-driven optimal operation of the industrial methanol to olefin process based on relevance vector machine, Chin. J. Chem. Eng. 34 (2021) 106-115.
[24] M.L. Yang, R. Hu, J. Long, Y.J. Lv, Z.C. Ye, Z. Li, W.M. Zhong, Modeling and optimization of simulated moving bed for Paraxylene Purification, Pet. Chem. 61 (2) (2021) 214-219.
[25] L. Huang, L. Sun, N. Wang, X.M. Jin, Multiobjective optimization of simulated moving bed by tissue P system, Chin. J. Chem. Eng. 15 (5) (2007) 683-690.
[26] W.W. Jiang, J.Y. Luo, Graph neural network for traffic forecasting: a survey, Expert Syst. Appl. 207 (2022) 117921.
[27] T. Nguyen, G.T.T. Nguyen, T. Nguyen, D.H. Le, Graph convolutional networks for drug response prediction, IEEE/ACM Trans. Comput. Biol. Bioinform. 19 (1) (2022) 146-154.
[28] F.M. Bianchi, D. Grattarola, L. Livi, C. Alippi, Graph neural networks with convolutional ARMA filters, IEEE Trans. Pattern Anal. Mach. Intell. 44 (7) (2022) 3496-3507.
[29] R. Levie, F. Monti, X. Bresson, M.M. Bronstein, CayleyNets: graph convolutional neural networks with complex rational spectral filters, IEEE Trans. Signal Process. 67 (1) (2019) 97-109.
[30] M.Y. Dai, M.F. Demirel, Y.Y. Liang, J.M. Hu, Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials, NPJ Comput. Mater. 7 (2021) 103.
[31] P. Sa Gomes, A.E. Rodrigues, Simulated moving bed chromatography: from concept to proof-of-concept, Chem. Eng. Technol. 35 (1) (2012) 17-34.
[32] P.P. Li, H.J. Hao, Z. Zhang, X.G. Mao, J.J. Xu, Y.T. Lv, W.M. Chen, D.B. Ge, A field study to estimate heavy metal concentrations in a soil-rice system: Application of graph neural networks, Sci. Total Environ. 832 (2022) 155099.
[33] V.M. Ribeiro, G. Soutinho, I. Soares, Natural gas prices in the framework of European union’s energy transition: assessing evolution and drivers, Energies 16 (4) (2023) 2029.
[34] Y.H. Chen, R. Preece, M. Barnes, Identifying feasibility region boundaries in power systems with multiple VSCs, IEEE Trans. Power Syst. 38 (2) (2023) 1229-1241.
[35] S. Conti, V. Ovchinnikov, M. Karplus, Ppdx: Automated modeling of protein-protein interaction descriptors for use with machine learning, J. Comput. Chem. 43 (25) (2022) 1747-1757.
[36] Z. Tariq, A. BinGhanim, M.S. Aljawad, M.S. Kamal, M. Mahmoud, Z. AlYousef, AI-driven foam rheological model based on HPHT foam rheometer experiments, J. Petrol. Sci. Eng. 213 (2022) 110439.
[37] P. Schober, C. Boer, L.A. Schwarte, Correlation coefficients: appropriate use and interpretation, Anesth. Analg. 126 (5) (2018) 1763-1768.
[38] K. Deb, A. Pratap, S. Agarwal, T. Meyarivan, A fast and elitist multiobjective genetic algorithm: NSGA-II, IEEE Trans. Evol. Comput. 6 (2) (2002) 182-197.
[39] Z.Q. Ge, Z.H. Song, F.R. Gao, Review of recent research on data-based process monitoring, Ind. Eng. Chem. Res. 52 (10) (2013) 3543-3562.
[40] Y. Tian, W.L. Du, F. Qian, Fault detection and diagnosis for non-Gaussian processes with periodic disturbance based on AMRA-ICA, Ind. Eng. Chem. Res. 52 (34) (2013) 12082-12107.
[41] Z. Li, Y.H. Ying, M.L. Yang, L. Zhao, L. Zhao, W.L. Du, Monitoring and path optimization of catalytic reformer in a refinery: Principal component analysis and A^* algorithm application, Expert Syst. Appl. 209 (2022) 118358.

Graph convolutional network for axial concentration profiles prediction in simulated moving bed

Graph convolutional network for axial concentration profiles prediction in simulated moving bed

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	Zhongxiang Ding, Wei Song, Tong Zhou, Weihua Cui, Changsong Wang. Thiourea crystal growth kinetics, mechanism and process optimization during cooling crystallization[J]. 中国化学工程学报, 2024, 73(9): 62-69.
[2]	Jing Chen, Wenqi Zhong, Guanwen Zhou, Jinming Li, Shasha Ding. Desulfurization characteristics of slaked lime and regulation optimization of circulating fluidized bed flue gas desulfurization process—A combined experimental and numerical simulation study[J]. 中国化学工程学报, 2024, 73(9): 163-175.
[3]	Kai Ji, Zhencheng Ye, Feng Qian. Reaction network design and hybrid modeling of S Zorb[J]. 中国化学工程学报, 2024, 73(9): 301-310.
[4]	Pengcheng Lu, Yaoyao Li, Jianjun Zhang, Yuchao Zhao, Qingqiang Wang, Ying Chen, Nan Jin, Xiugang Yu. Continuous synthesis of N, N-dicyanoethylaniline in microreactors: Reaction kinetics and process intensification[J]. 中国化学工程学报, 2024, 72(8): 95-105.
[5]	Baowei Niu, Yanjie Yi, Yuwen Wei, Fuzhen Zhang, Lili Wang, Li Xia, Xiaoyan Sun, Shuguang Xiang. Phase equilibrium data prediction and process optimizationin butadiene extraction process[J]. 中国化学工程学报, 2024, 71(7): 1-12.
[6]	Chunliang Liu, Jianhui Zhong, Ranran Wei, Jiuxu Ruan, Kaicong Wang, Zhaoyou Zhu, Yinglong Wang, Limei Zhong. Process design and intensification of multicomponent azeotropes special distillation separation via molecular simulation and system optimization[J]. 中国化学工程学报, 2024, 71(7): 24-44.
[7]	Dandan Ren, Jiale Xie, Tianle Chen, Haibin Qu, Xingchu Gong. Design and optimization of a greener sinomenine hydrochloride preparation process considering variations among different batches of the medicinal herb[J]. 中国化学工程学报, 2024, 71(7): 77-90.
[8]	Zhiwei Zhu, Minglei Yang, Wangli He, Renchu He, Yunmeng Zhao, Feng Qian. A deep reinforcement learning approach to gasoline blending real-time optimization under uncertainty[J]. 中国化学工程学报, 2024, 71(7): 183-192.
[9]	Cheng Yang, Chao Jiang, Guo Yu, Jun Li, Cuimei Bo. Transferable adversarial slow feature extraction network for few-shot quality prediction in coal-to-ethylene glycol process[J]. 中国化学工程学报, 2024, 71(7): 258-271.
[10]	Xin Dai, Liang Zhao, Renchu He, Wenli Du, Weimin Zhong, Zhi Li, Feng Qian. Data-driven Wasserstein distributionally robust chance-constrained optimization for crude oil scheduling under uncertainty[J]. 中国化学工程学报, 2024, 69(5): 152-166.
[11]	Hebin Li, Zifei Meng, Dehua Wang, Ye Lu, Longlong Jiang, Le Zhang, Hanbin Wang, Xiaoxiong Wang. Application of different fiber structures and arrangements by electrospinning in triboelectric nanogenerators[J]. 中国化学工程学报, 2024, 69(5): 177-191.
[12]	Liuyu Song, Haibo Li, Pengkai Wang, Yu Shang, Yue Yang, Zhaoyu Wu. MXene: Promising materials for magnesium-ion batteries[J]. 中国化学工程学报, 2024, 69(5): 199-211.
[13]	Guangxia Jin, Yuxue Wu, Feng Wei. Simultaneous purification of minor components in natural products using twin-column recycling chromatography with a step solvent gradient[J]. 中国化学工程学报, 2024, 69(5): 212-219.
[14]	Qingping Qu, Daoyan Liu, Hao Lyu, Jinsheng Sun. Process synthesis for the separation of coal-to-ethanol products[J]. 中国化学工程学报, 2024, 69(5): 263-278.
[15]	Xuepu Cao, Shengkun Jia, Xing Qian, Yiqing Luo, Xigang Yuan. Cascade equilibrium stage relaxation method by introducing equilibrium efficiency parameter[J]. 中国化学工程学报, 2024, 66(2): 145-156.