Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes

Lin Guocong; Ni Qingxu; Liu Xuehai; Geng Zhiqiang; Pan Feng; Tian Jun; Han Yongming

doi:null

Your Location：

Home >

Browse articles >

Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes

Updated：2026-05-14

- Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes
- Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes
- 中国化学工程学报（英文） 2026年90卷第2期页码：348-360
- Affiliations：
  
  College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China
  Engineering Research Center of Intelligent PSE, Ministry of Education of China, Beijing, 100029, China
  Sinopec Sales Co., Ltd., North China Branch, Tianjin 300384, China
  School of Economics and Management, Beijing University of Chemical Technology, Beijing 100029, China
- Author bio：
  
  These authors contributed equally to this work.
  Corresponding authors. E-mail addresses: gengzhiqiang@mail.buct.edu.cn (Z. Geng)
  Corresponding authors. E-mail addresses: 2010500027@buct.edu.cn (J. Tian)
  Corresponding authors. E-mail addresses: hanym@mail.buct.edu.cn (Y. Han).
- Funds：
  
  the Xinjiang Uygur Autonomous Region Key R&D Project, China(2023B01031-3);the National Natural Science Foundation of China(62422303;62373035)
- DOI：
  中图分类号：
- 收稿：2025-06-04，
  
  修回：2025-08-26，
  
  录用：2025-09-10，
  
  网络首发：2025-11-26，
  
  纸质出版：2026-02
- Accepted：
Scan QR Code
Lin Guocong, Ni Qingxu, Liu Xuehai, 等. Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes[J]. 中国化学工程学报（英文）, 2026,90(2):348-360.

Lin Guocong, Ni Qingxu, Liu Xuehai, et al. Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes[J]. Chinese Journal of Chemical Engineering, 2026, 90(2): 348-360.
Lin Guocong, Ni Qingxu, Liu Xuehai, 等. Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes[J]. 中国化学工程学报（英文）, 2026,90(2):348-360. DOI：

Lin Guocong, Ni Qingxu, Liu Xuehai, et al. Improved neural network automation design method for energy saving and carbon emission reduction of petrochemical production processes[J]. Chinese Journal of Chemical Engineering, 2026, 90(2): 348-360. DOI：

摘要

Abstract

Ethylene yield serves as a key metric in petrochemical production

where optimizing its energy efficiency remains a critical challenge for sustainable production. Meanwhile

manual hyperparameter tuning of deep learning based yield prediction models often results in suboptimal configurations

which reduces prediction accuracy and reliability due to extreme operating conditions generating outliers in ethylene production processes. Therefore

a novel neural network automatic design method (NNADM) is proposed

which incorporates the neural network parameters automatic optimization and loss function adaptive construction. An innovative adaptive loss formulation is proposed to strategically integrate the complementary strengths of the mean squared error (MSE) and the Log-Cosh functions

featuring dynamic outlier resistance through self-adjusting weight coefficients. Then

the Bayesian optimization search algorithm is utilized to discover optimal hyperparameters of the neural network

including hidden layer unit

epoch

batch size

and the loss function. Finally

the NNADM is integrated with several classical neural networks for ethylene yield prediction. Experimental results show that several classical neural networks realize an average of 16.86% decrease in MSE index after integrating the NNADM. In addition

the proposed model offers direction and development blueprints for ethylene production facilities that have low energy efficiency. By implementing this model

it is possible to cut approximately 8376.4 tons of carbon emissions and simultaneously secure an extra 499 tons of ethylene output.

关键词

Keywords

references

S.X. Gong, Multi-scale energy efficiency recognition and diagnosis scheme for ethylene production based on a hierarchical multi-indicator system, Energy 267 (2023)126478.

Z.Q. Geng, W.K. Kong, X.T. W ang, L. Wang, Y.M. Han, Adaptive search based Grey Wolf optimization algorithm for multi-objective optimization of ethylene cracking furnace, Swarm Evol. Comput. 92(2025)101810.

J.S. Samuel, A.H. Muggeridge, Fast modelling of gas reservoir performance with proper orthogonal decomposition based autoencoder and radial basis function non-intrusive reduced order models, J. Petrol. Sci. Eng. 211(2022)110011.

N. Liu, J. Wang, S.L. Sun, C.K. Li, W.D. Tian, Optimized principal component analysis and multi-state Bayesian network integrated method for chemical process monitoring and variable state prediction, Chem. Eng. J. 430 (2022) 132617.

M.I. Jahirul, M.G. Rasul, R.J. Brown, W. Senadeera, M.A. Hosen, R. Haque, S.C. Saha, T.M.I. Mahlia, Investigation of correlation between chemical composition and properties of biodiesel using principal component analysis (PCA) and artificial neural network (ANN), Renew. Energy 168 (2021) 632—646.

H.L. Chen, I. Ahmadianfar, G.X. Liang, A.A. Heidari, Robust kernel extreme learning machines with weighted mean of vectors and variational mode decomposition for forecasting total dissolved solids, Eng. Appl. Artif. Intell. 133 (2024) 108587.

S. Afzal, A. Shokri, B.M. Ziapour, H. Shakibi, B. Sobhani, Building energy consumption prediction and optimization using different neural networkassisted models; comparison of different networks and optimization algorithms, Eng. Appl. Artif. Intell. 127 (2024) 107356.

Y.Z. Li, R. Gupta, W.L. Li, Y. Fang, J. Toney, S.M. You, Machine learning-assisted life cycle assessment of biochar soil application, J. Clean. Prod. 498(2025)145109.

D.J. Godwin, E.G. Varuvel, M.L. Jesu Martin, A. Jasmine R, F. Josephin Js, A comparative analysis of advanced machine learning models for the prediction of combustion, emission and performance characteristics using endoscopic combustion flame image of a pine oil—gasoline fuelled spark ignition engine, J. Clean. Prod. 484 (2024) 144284.

F. Jani, S. Hosseini, A. Sepahi, S.K. Afzali, F. Torabi, R. Ghorbani, S. Houshmandmoayed, Enhancing quality control of polyethylene in industrial polymerization plants through predictive multivariate data-driven soft sensors, Can. J. Chem. Eng. 103 (5) (2025) 2234—2250.

J. Taghinezhad, S. Sheidaei, Prediction of operating parameters and output power of ducted wind turbine using artificial neural networks, Energy Rep. 8 (2022)3085—3095.

G. Işık, H. Öğüt, M. Mutlu, Deep learning based electricity demand forecasting to minimize the cost of energy imbalance: a real case application with some fortune 500 companies in Türkiye, Eng. Appl. Artif. Intell. 118(2023)105664.

X. Ma, J. Wang, J. Huang, Y. Ke, Energy consumption and carbon emission modeling and forecasting study with novel d eep learning methods, Expert Syst. Appl. 290(2025)128314.

B. Metcalfe, J.C. Acosta-Pavas, C.E. Robles-Rodriguez, G.K. Georgakilas, T. Dalamagas, C.A. Aceves-Lara, F. Daboussi, J.J. Koehorst, D.C. Corrales, Towards a machine learning operations (MLOps) soft sensor for real-time predictions in industrial-scale fed-batch fermentation, Comput. Chem. Eng. 194 (2025) 108991.

Y.N. Sun, Z.E. Qu, Z.D. Liu, X.Y. Li, Hierarchical multi-scale decomposition and deep learning ensemble framework for enhanced carbon emission prediction, Mathematics 13 (12) (2025) 1924.

X.Y. Zhou, E.W. Song, M.M. Wang, E.Q. Wang, Dynamic temperature control of dividing wall batch distillation with middle vessel based on neural network soft-sensor and fuzzy control, Chin. J. Chem. Eng. 79(2025)200—211.

S.W. Gao, P.X. Yun, W.B. Yang, J. Yan, Dynamic soft sensor modelling based on data imputation and spatiotemporal attention, Can. J. Chem. Eng. 103 (10) (2025)4751—4773.

Y. Li, H.T. Cao, X. Wang, Z.N. Yang, N. Li, W.F. Shen, A new correlationsimilarity conjoint algorithm for developing encoder-decoder based deep learning multi-step prediction model of chemical process, Chem. Eng. Sci. 288 (2024) 119748.

Z. Li, Y.C. Xia, J. Long, C.S. Liu, L.F. Zhang, Multi-scale feature fused stacked autoencoder and its application for soft sensor modeling, Chin. J. Chem. Eng. 81 (2025)241—254.

R.Z. Homod, B.S. Munahi, H.I. Mohammed, M.A.A. Albadr, A. Abderrahmane, J. M. Mahdi, M.B. Ben Hamida, B.N. Alhasnawi, A.S. Albahri, H. Togun, U.F. Alqsair, Z.M. Yaseen, Deep clustering of reinforcement learning based on the bang-bang principle to optimize the energy in multi-boiler for intelligent buildings, Appl. Energy 356(2024)122357.

G.Y. Xu, J. Shi, J.M. Wu, C.B. Lu, C.Y. Wu, D. Wang, Z. Han, An optimal solutions-guided deep reinforcement learning approach for online energy storage control, Appl. Energy 361(2024) 122915.

X.T. Deng, Y. Zhang, Y. Jiang, Y. Zhang, H. Qi, A novel operation method for renewable building by combining distributed DC energy system and deep reinforcement learning, Appl. Energy 353(2024)122188.

K. Li, Y.C. Mu, F. Yang, H.Y. Wang, Y. Yan, C.H. Zhang, Joint forecasting of source-load-price for integrated energy system based on multi-task learning and hybrid attention mechanism, Appl. Energy 360(2024)122821.

C.R. Song, H.D. Yang, J.Y. Cai, P. Yang, H. Bao, K.K. Xu, X.B. Meng, Multi-energy load forecasting via hierarchical multi-task learning and spatiotemporal attention, Appl. Energy 373(2024)123788.

C. Brozos, J.G. Rittig, E. Akanny, S. Bhattacharya, C. Kohlmann, A. Mitsos, Predicting the temperature-dependent CMC of surfactant mixtures with graph neural networks, Comput. Chem. Eng. 198(2025)109085.

Y.M. Han, X.H. Liu, C. Guo, H. Wu, M. Liu, Z.Q. Geng, Improved pearson correlation coefficient-based graph neural network for dynamic soft sensor of polypropylene industries, Ind. Eng. Chem. Res. 64 (1) (2025) 551—563.

W.L. Guo, Y.Z. Wen, M.H. Liu, H.W. Chen, L.Q. Wen, Real-time, reagent-free total phosphorus soft sensor based on frequency-enhanced decomposed transformer model, Measurement 253(2025)117509.

G.H. Ban, Y. Chen, Z.H. Xiong, Y.X. Zhuo, K. Huang, The univariate model for long-term wind speed forecasting based on wavelet soft threshold denoising and improved autoformer, Energy 290(2024)130225.

Y.M. Han, C.K. Zeng, Q.X. Ni, J.Q. Wang, Z.Y. Chu, X.X. Zhang, Z.Q. Geng, L. Tan, Y.D. Liu, Time series prediction of anaerobic digestion yield and carbon emissions from food waste based on iTransformer model, Chem. Eng. J. 513 (2025)163064.

X.P. Guo, S.L. Pan, Y. Li, Global dynamic features and information of adjacent hidden layer enhancement based on autoencoder for industrial process soft sensor application, Can. J. Chem. Eng. 103(5)(2025)2218—2233.

L.J. Ge, Y.L. Li, J. Yan, Y.Q. Wang, N. Zhang, Short-term load prediction of integrated energy system with wavelet neural network model based on improved particle swarm optimization and chaos optimization algorithm, J. Mod. Power Syst. Clean Energy 9(6)(2021)1490—1499.

K. Nakamura, S. Soatto, B.W. Hong, Stochastic batch size for adaptive regularization in deep network optimization, Pattern Recogn. 129 (2022) 108776.

R. Liang, T. Le-Hung, T. Nguyen-Thoi, Energy consumption prediction of airconditioning systems in eco-buildings using hunger games search optimizationbased artificial neural network model, J. Build. Eng. 59 (2022) 105087.

M.S. Nawaz, B. Shoaib, M.A. Ashraf, Intelligent cardiovascular disease prediction empowered with gradient descent optimization, Heliyon 7 (5) (2021) e06948.

H.B. Huang, X.R. Huang, W.P. Ding, S.W. Zhang, J. Pang, Optimization of electric vehicle sound package based on LSTM with an adaptive learning rate forest and multiple-level multiple-object method, Mech. Syst. Signal Process. 187(2023)109932.

M. Zulfiqar, M. Kamran, M.B. Rasheed, T. Alquthami, A.H. Milyani, Hyperparameter optimization of support vector machine using adaptive differential evolution for electricity load forecasting, Energy Rep. 8 (2022) 13333—13352.

Z.H. Chang, W. Yuan, K. Huang, Remaining useful life prediction for rolling bearings using multi-layer grid search and LSTM, Comput. Electr. Eng. 101 (2022)108083.

S. Narmadha, N.V. Balaji, Improved network anomaly detection system using optimized autoencoder- LSTM, Expert Syst. Appl. 273(2025)126854.

M. Tunçel, A. Duran, Effectiveness of grid and random approaches for a model parameter vector optimization, J. Comput. Sci. 67(2023)101960.

P. Sing Babu, A. Goswami, V.S. Pandit, Optimization of beam line parameters for space charge dominated multi-species beam using random search method, Phys. Lett. 376(45)(2012)3192—3198.

S. Ansari, A. Ayob, M.S. Hossain Lipu, A. Hussain, M.H.M. Saad, Particle swarm optimized data-driven model for remaining useful life prediction of lithium-ion batteries by systematic sampling, J. Energy Storage 56 (2022) 106050.

F. El Jamiy ElSaid, J. Higgins, B. Wild, T. Desell, Optimizing long short-term memory recurrent neural networks using ant colony optimization to predict turbine engine vibration, Appl. Soft Comput. 73 (2018) 969—991.

H. Kedia, A. Pandey, A. Kumar, A. Majumdar, J. De, N. Ghosh, Optimization of process parameters in machining of inconel 718 super alloy on HMT(NH 2 2) capstan lathe using genetic algorithm subject to minimization of surface roughness, Mater. Today Proc. 66(2022)3910—3915.

S. Wang, L.H. Li, H.S. Zhang, X.F. Liu, N.Q. Li, Q.H. Wang, A local semisupervised soft sensor modeling method based on SAE neural networks for spatiotemporal dynamic chemical process, Ind. Eng. Chem. Res. 63 (50) (2024) 21645—21661.

Q.H. Wang, L.H. Li, Research on semi-supervised soft sensor modeling method for sulfur recovery unit based on ISSA-VMD-ESN, Chem. Eng. Sci. 298 (2024)120397.

B. Liu, Y.W. Liu, Corrosion life prediction of glass fiber reinforced plastics by optimized BP neural network, Inorg. Chem. Commun. 144(2022)109846.

M. Ghasemi, S.K. Mohammadi, M. Zare, S. Mirjalili, M. Gil, R. Hemmati, A new firefly algorithm with improved global exploration and convergence with application to engineering optimization, Decis. Anal. J. 5(2022)100125.

H.R. Nejati Mahmoodzadeh, M. Mohammadi, H. Hashim Ibrahim, S. Rashidi, T. Ahmed Rashid, Forecasting tunnel boring machine penetration rate using LSTM deep neural network optimized by grey wolf optimization algorithm, Expert Syst. Appl. 209(2022)118303.

V. Gkonis, I. Tsakalos, A hybrid optimized deep learning model via the golden jackal optimizer for accurate stock price forecasting, Expert Syst. Appl. 278 (2025)127287.

J. Long, K. Deng, R.C. He, Closed-loop scheduling optimization strategy based on particle swarm optimization with niche technology and soft sensor method of attributes-applied to gasoline blending process, Chin. J. Chem. Eng. 61(2023)43—57.

Y.N. Qian, Q.G. Li, W.X. Li, Q.T. Hu, B.Z. Yan, C. J. Yu, S.Y. Peng, Evaluation of hydraulic fracturing in coal seam based on time difference-particle swarm optimization algorithm, Energy Fuel. 38(12)(2024)10955—10971.

Z. Sun, E. Caetano, S. Pereira, C. Moutinho, Employing histogram of oriented gradient to enhance concrete crack detection performance with classification algorithm and Bayesian optimization, Eng. Fail. Anal. 150(2023)107351.

J.L. Li, R.X. Chen, X.Z. Huang, Y.Z. Qu, Development of deep residual neural networks for gear pitting fault diagnosis using Bayesian optimization, IEEE Trans. Instrum. Meas. 71(2022) 2521715.

D.F. Chen, W.L. Wu, K.Y. Chang, Y.H. Li, P.C. Pei, X.M. Xu, Performance degradation prediction method of PEM fuel cells using bidirectional long short-term memory neural network based on Bayesian optimization, Energy 285(2023)129469.

D. Kim, K. Kwon, K. Pham, J.Y. Oh, H. Choi, Surface settlement prediction for urban tunneling using machine learning algorithms with Bayesian optimization, Autom. Constr. 140 (2022) 104331.

J.P. Xu, S.W. Gao, X.C. Dang, W.F. Zhao, Q.S. Zhang, S.L. Qiu, BO-MADRSN:bayesian optimized multi-attention residual shrinkage networks for industrial soft sensor modeling, Measurement 224 (2024) 113477.

R.M. Zou, J.X. Yang, Y. Wang, F. Liu, M. Essaaidi, D. Srinivasan, Wind turbine power curve modeling using an asymmetric error characteristic-based loss function and a hybrid intelligent optimizer, Appl. Energy 304 (2021) 117707.

Z.T. Zhao, K.T. Chong, J.Y. Jiang, K. Wilson, X.C. Zhang, F. Wang, Low-carbon roadmap of chemical production: a case study of ethylene in China, Renew. Sustain. Energy Rev. 97(2018)580—591.

浏览量

Downloads

CSCD

文章被引用时，请邮件提醒。

Submit

工具集

关联资源

Prediction of ionic liquid toxicity by interpretable machine learning

Estimating heat capacities of liquid organic compounds based on elements and chemical bonds contribution

Proportion integral-type active disturbance rejection generalized predictive control for distillation process based on grey wolf optimization parameter tuning

Multivariable identification of membrane fouling based on compacted cascade neural network

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