CIESC Journal ›› 2018, Vol. 69 ›› Issue (11): 4814-4822.doi: 10.11949/j.issn.0438-1157.20180534

Previous Articles     Next Articles

Self-adaptive iterative hybrid modeling and its application in acetylene hydrogenation process

GUO Jingjing, XU Jinjin, DU Wenli, YE Zhencheng   

  1. State Key Laboratory of Chemical Engineering, Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education, East China University of Science and Technology, Shanghai 200237, China
  • Received:2018-05-22 Revised:2018-07-03 Online:2018-11-05 Published:2018-07-10
  • Supported by:

    supported by the Major Program of National Natural Science Foundation of China (61590923), the Key Program of National Natural Science Foundation of China (61333010) and the National Natural Science Foundation for Distinguished Young Scholars(61725301).


The reaction mechanism of chemical process is complex. There is a modeling error between the mechanism model and the actual reaction system. At the same time, there are complex slow time-varying features, such as catalyst deactivation, fuel coking, etc. Thus there will be a mismatch between the process model and the actual process system. A self-adaptive iterative hybrid model (SAIHM) is established to reflect the dynamic characteristics of the process accurately over a long period. The mechanism model and the data-driven model are effectively combined to improve the prediction accuracy of the model; the data-driven model uses the deep recurrent neural network (DRNN) to fully exploit the timing relationship between adjacent conditions; the data-driven model is automatically updated based on the evaluation indicators to resolve the contradiction between accuracy and efficiency. The simulation and comparison results of the self-adaptive iterative hybrid model and the existing mechanism model established based on the historical operation data of an acetylene hydrogenation adiabatic reactor show that the self-adaptive iterative hybrid model can more effectively track the actual system.

Key words: self-adaptive, iterative, hybrid model, dynamic, acetylene hydrogenation

CLC Number: 

  • TQ221.242

[1] CHEN L, HONTOIR Y, HUANG D, et al. Combining first principles with black-box techniques for reaction systems[J]. Control Engineering Practice, 2004, 12(7):819-826.
[2] YANG A, MARTIN E, MORRIS J. Identification of semi-parametric hybrid process models[J]. Computers & Chemical Engineering, 2011, 35(1):63-70.
[3] LUO N, DU W L, YE Z C, et al. Development of a hybrid model for industrial ethylene oxide reactor[J]. Industrial & Engineering Chemistry Research, 2012, 51(19):6926-6932.
[4] ZHONG W M, JIANG C, PENG X, et al. Online quality prediction of industrial terephthalic acid hydropurification process using modified regularized slow feature analysis[J] Industrial & Engineering Chemistry Research, 2018, 57(29):9604-9614.
[5] AZARPOUR A, ALWI S R W, ZAHEDI G, et al. Catalytic activity evaluation of industrial Pd/C catalyst via gray-box dynamic modeling and simulation of hydropurification reactor[J]. Applied Catalysis A General, 2015, 489:262-271.
[6] MUKUL A. Combining neural and conventional paradigms for modelling, prediction and control[J]. International Journal of Systems Science, 1997, 28(1):65-81.
[7] KUMAR B S, VENKATESWARLU C. Estimating biofilm reaction kinetics using hybrid mechanistic-neural network rate function model[J]. Bioresource Technology, 2012, 103(1):300-308.
[8] SU H T, BHAT N, MINDERMAN P A, et al. Integrating neural networks with first principles models for dynamic modeling[J]. IFAC Proceedings Volumes, 1992, 25(5):327-332.
[9] THOMPSON M L, KRAMER M A. Modeling chemical processes using prior knowledge and neural networks[J]. AIChE Journal, 1994, 40(8):1328-1340.
[10] OLIVEIRA R. Combining first principles modelling and artificial neural networks:a general framework[J]. Computers & Chemical Engineering, 2004, 28(5):755-766.
[11] SU H T, MCAVOY T J. Integration of multilayer perceptron networks and linear dynamic models[J]. Industrial & Engineering Chemistry Research, 1993, 26(2):137-40.
[12] STOSCH M V, RUI O, PERES J, et al. Hybrid semi-parametric modeling in process systems engineering:past, present and future[J]. Computers & Chemical Engineering, 2014, 60(2):86-101.
[13] AL-MUSAYLH M S, DEO R C, LI Y, et al. Two-phase particle swarm optimized-support vector regression hybrid model integrated with improved empirical mode decomposition with adaptive noise for multiple-horizon electricity demand forecasting[J]. Applied Energy, 2018, 217:422-439.
[14] 张东平, 王功华. 乙炔加氢反应器的模拟与分析[J]. 石油化工, 2003, 32(5):414-418. ZHANG D P, WANG G H. Simulating and analysis of reactor for selective hydrogenation of acetylene[J]. Petrochemical Technology, 2003, 32(5):415-418.
[15] 罗雄麟, 刘建新, 许锋, 等. 乙炔加氢反应器二维非均相机理动态建模及分析[J]. 化工学报, 2008, 59(6):1454-1461. LUO X L, LIU J X, XU F, et al. Heterogeneous, two-dimensional dynamic modeling and analysis of acetylene hydrogenation reactor[J]. Journal of Chemical Industry and Engineering, 2008, 59(6):1454-1461.
[16] 田亮, 蒋达, 钱锋. 催化剂失活条件下的碳二加氢反应器模拟与优化[J]. 化工学报, 2012, 63(1):185-192. TIAN L, JIANG D, QIAN F. Simulation and optimization of acetylene converter with decreasing catalyst activity[J]. CIESC Journal, 2012, 63(1):185-192.
[17] AZIZI M, SHARAK A Z, MOUSAVI S A, et al. Study on the acetylene hydrogenation process for ethylene production:simulation, modification, and optimization[J]. Chemical Engineering Communications, 2013, 200(7):863-877.
[18] JIN Y, DATYE A K, RIGHTOR E, et al. The influence of catalyst restructuring on the selective hydrogenation of acetylene to ethylene[J]. Journal of Catalysis, 2001, 203(2):292-306.
[19] WU W, LI Y L, CHEN W S, et al. Kinetic studies and operating strategies for an industrial selective hydrogenation process[J]. Ind. Eng. Chem. Res., 2011, 50(3):1264-1271.
[20] BENAVIDEZ A D, BURTON P D, NOGALES J L, et al. Improved selectivity of carbon-supported palladium catalysts for the hydrogenation of acetylene in excess ethylene[J]. Applied Catalysis A General, 2014, 482(28):108-115.
[21] 田亮, 蒋达, 钱锋. 钯金属催化剂上的乙炔工业选择性加氢反应动力学比较[J]. 计算机与应用化学, 2012, 29(9):1031-1035. TIAN L, JIANG D, QIAN F. Reaction kinetic comparsions for industrial selective hydrogenation of acetylene on palladium catalyst[J]. Computers and Applied Chemistry, 2012, 29(9):1031-1035.
[22] HUANG W, MCCORMICK J R, LOBO R F, et al. Selective hydrogenation of acetylene in the presence of ethylene on zeolite-supported bimetallic catalysts[J]. Journal of Catalysis, 2007, 246(1):40-51.
[23] HOUZVICKA J, PESTMAN R, PONEC V. The role of carbonaceous deposits and support impurities in the selective hydrogenation of ethyne[J]. Catalysis Letters, 2012, 445/446(1/2/3/4):351-358.
[24] 张健, 黄邦印, 隋志军, 等. 碳二加氢失活Pd-Ag催化剂的表征[J]. 石油化工, 2017, 46(7):839-844. ZHANG J, HUANG B Y, SUI Z J, et al. The characterization of the deactivated C2 hydrogenation Pd-Ag/Al2O3 catalyst[J]. Petrochemical Technology, 2017, 46(7):839-844.
[25] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors.[J]. Nature, 1986, 323(6088):399-421.
[26] GOODFELLOW I, BENGIO Y, COURVILLE A, et al. Deep Learning[M]. Cambridge:MIT Press, 2016:379-382.
[27] JAEGER H. Discovering multiscale dynamical features with hierarchical Echo State Networks[J]. Vtls Inc., 2007, 35(2):277-284.
[28] GRAVES A. Generating sequences with recurrent neural networks[J]. Computer Science, arXiv preprint arXiv:1308.0850, 2013.
[29] PASCANU R, GULCEHRE C, CHO K, et al. How to construct deep recurrent neural networks[J]. Computer Science, arXiv preprint arXiv:1312.6026, 2013.
[30] KINGMA D P, BA J. Adam:a method for stochastic optimization[J]. Computer Science, arXiv preprint arXiv:1412.6980, 2014.

[1] KANG Yinhu, ZHANG Pengyuan, LIU Congcong, MA Jiangze, LU Xiaofeng. Extinction mechanism of ethylene opposed-flow diffusion flame using chemical explosive mode analysis method [J]. CIESC Journal, 2019, 70(4): 1644-1651.
[2] JIANG Aipeng, ZHANG Quannan, WANG Haokun, DING Qiang, XU Weifeng, WANG Jian. An improved dynamic real time optimization strategy for heat pump heating system [J]. CIESC Journal, 2019, 70(4): 1494-1504.
[3] YANG Shipin, HUANG Zhen, LI Lijuan, SONG Jianquan, YE Jing, WANG Hui. A deep dive diagnostic and correction algorithm for mismatched sub-models in complicated chemical processes [J]. CIESC Journal, 2019, 70(4): 1485-1493.
[4] LI Peng, HAN Zhonghe, JIA Xiaoqiang, MEI Zhongkai, HAN Xu. Influence of dynamic turbine efficiency on performance of organic Rankine cycle system [J]. CIESC Journal, 2019, 70(4): 1532-1541.
[5] MENG Xiangkai, JIANG Yingying, ZHAO Wenjing, PENG Xudong. Analysis of thermohydrodynamic lubrication performance of spiral-grooved liquid film seals [J]. CIESC Journal, 2019, 70(4): 1512-1521.
[6] LIU Wanqiang, LU Haixia, LIU Fengping, CHEN Guanfan, HU Tian, YUE Ming, QIU Minghua. Estimation of thermal conductivity of liquid alcohols using finite element solution based on principle of minimum potential energy [J]. CIESC Journal, 2019, 70(4): 1245-1254.
[7] CHEN Hongxia, SUN Yuan, GONG Yifei, HUANG Linbin. Visual measurement and data analysis of pool boiling on silicon surfaces [J]. CIESC Journal, 2019, 70(4): 1309-1317.
[8] LIANG Jie, LI Yuxing, LIU Cuiwei, ZHU Jianlu, WANG Shaoxiong, LIU Na. Experimental study on leakage characteristics of buried gas pipelines [J]. CIESC Journal, 2019, 70(4): 1635-1643.
[9] YU Xudong, HUANG Qin, WANG Lin, LI Maolan, ZHENG Hong, ZENG Ying. Measurements and simulation for ternary system KCl-PEG4000-H2O at 288, 298 and 308 K [J]. CIESC Journal, 2019, 70(3): 830-839.
[10] HUANG Xiuhui, WANG Jun, CUI Guomin. Dynamic simulation and analysis of control strategies of acetic acid dehydration tower in PTA plant [J]. CIESC Journal, 2019, 70(2): 625-633.
[11] YE Zhencheng, ZHOU Huanlan, RAO Debao. Hybrid modeling and optimization of acetylene hydrogenation process [J]. CIESC Journal, 2019, 70(2): 496-507.
[12] CHEN Haisheng, WANG Tengfei, HUANG Kejin, YUAN Yang, QIAN Xing, ZHANG Liang. Decentralized control system designs for reactive distillation columns with external recycle [J]. CIESC Journal, 2019, 70(2): 440-449.
[13] LI Wenjing, LI Meng, QIAO Junfei. Effluent BOD soft measurement based on mutual information and self-organizing RBF neural network [J]. CIESC Journal, 2019, 70(2): 687-695.
[14] ZHAI Kun, DU Wenxia, LYU Feng, XIN Tao, JU Xiyuan. Fault detect method based on improved dynamic kernel principal component analysis [J]. CIESC Journal, 2019, 70(2): 716-722.
[15] ZHU Yi, WANG Hao, CHEN Liping, GUO Zichao, HE Zhongqi, CHEN Wanghua. Calculate time to maximum rate under adiabatic condition by numerical calculation method [J]. CIESC Journal, 2019, 70(1): 379-387.
Full text



No Suggested Reading articles found!