大数据驱动建模平台技术在加氢裂化工艺预测上的应用

摘要/Abstract

摘要： 随着炼化装置传感器与集散控制系统的普及，基于数据驱动建模技术对生产大数据的分析、模拟，进而指导生产优化、监测预警，日益成为研究应用热点。总结了数据驱动建模方法，包括多元统计过程控制(MSPC)、机器学习(ML)、深度学习(DL) 等在炼化流程建模优化过程中的应用；介绍了中海油惠州石化有限公司构建的炼化大数据驱动建模平台，以及基于堆叠自编码器(SAEs)-混合高斯模型(GMM)对多模态工况下加氢裂化生产喷气燃料收率的预测。结果表明：炼化大数据驱动模型对加氢裂化过程预测的准确性优异；与传统数据驱动建模相比，该炼化大数据驱动建模平台实现了无代码建模工作流，建模耗时由传统的7 d大幅缩短为2 h左右，显著提升了建模效率。

关键词: 数据驱动建模, 炼化大数据, 建模平台, 加氢裂化, 多模态工况

Abstract: With the popularization of sensors and distributed control systems in refining and chemical plants, the analysis and simulation of production using big data based on data-driven modeling technology can guide production optimization, process monitoring and early warning, which has become a hot spot of research and application. Data-driven modeling methods, including the application of multivariate statistical process control (MSPC), machine learning (ML) and deep learning (DL) in the process of refining and chemical process modeling, were summarized. The big data-driven modeling platform built by CNOOC Huizhou Petrochemical Company for refining and petrochemical industry was introduced, and based on the big data-driven modeling platform, the stacked auto-encoders-Gaussian mixture model (SAEs-GMM) algorithm was proposed to predict the jet fuel yield of hydrocracking unit under multi-modes. The results showed that the prediction accuracy of the refinery and petrochemical big data-driven model was excellent, and compared with the traditional data-driven modeling methods, the refinery and petrochemical big data-driven modeling platform realized the code-less modeling workflow, the time of modeling was shortened from the traditional 7 days to about 2 hours, and the efficiency of modeling was greatly improved.

Key words: data-driven modeling, petrochemical big data, modeling platform, hydrocracking, multi-modes

王晨陆鹏飞阮冰杨纪方友曹晓红. 大数据驱动建模平台技术在加氢裂化工艺预测上的应用[J]. 石油炼制与化工, 2023, 54(3): 120-126.

参考文献

[1]张梦轩，刘洪辰，王敏，等. 化工过程的智能混合建模方法及应用[J]. 化工进展, 2021, 40(4): 1765-1775.
[2]王晨, 杨杰, 李保良. Petro-SIM模拟软件在中压加氢裂化流程模拟中的应用[J]. 炼油技术与工程, 2018, 48(5): 42-46
[3]孟凡辉，纪传佳，杨纪. 惠州石化有限公司连续重整装置工艺流程模拟与优化[J]. 化工进展, 2017,36(7): 2724.
[4]王晨. BP神经网络在中压加氢裂化装置多方面预测中的应用研究[J]. 石油炼制与化工, 2018, 49(7): 95-97.
[5] 闫乃锋，王晨. BP神经网络在加氢裂化装置航煤性质软测量中的应用[J]. 工业催化，2020, 28(8): 65-68.
[6] 杨帆，周敏，金继民，等. 智能优化算法及人工神经网络在催化裂化模型分析中的应用进展[J]. 石油学报（石油加工），2020，36（7）：879-888.
[7] 赵春晖，王福利. 工业过程运行状态智能监控：数据驱动方法[M]. 北京：化学工业出版社, 2019: 7-8.
[8] GE Zhiqiang. Review on data-driven modeling and monitoring for plant-wide industrial processes[J]. Chemometrics and Intelligent Laboratory Systems, 2017, 171: 16-25.
[9] SHANG Chao, YANG Fan, HUANG Dexian, et al. Data-driven soft sensor development based on deep learning technique[J]. Journal of Process Control, 2014, 24(3): 223-233.
[10] Diogo A.C. Narciso, F.G. Martins. Application of machine learning tools for energy efficiency in industry: A review[J]. Energy Reports, 2020, 6: 1181-1199.
[11] Marcos Qui?ones-Grueiro, Alberto Prieto-Moreno, Cristina Verde, et al. Data-driven monitoring of multimode continuous processes: A review. Chemometrics and Intelligent Laboratory Systems, 2019, 189(15): 56-71.
[12] GE Zhiqiang, SONG Zhihuan, GAO Furong. Review of recent research on data-based process monitoring[J]. Industrial & Engineering Chemistry Research, 2013, 52(10): 3543–3562.
[13] GE Zhiqiang, SONG Zhihuan, DING STEVEN X., et al. Data Mining and Analytics in the Process Industry: The Role of Machine Learning[J]. IEEE Access, 2017, 5: 20590-20616.
[14] SUN Qingqiang, GE Zhiqiang. A Survey on Deep Learning for Data-Driven Soft Sensors[J], IEEE Transaction on Industrial Informatics, 2021, 19(9): 5853-5866.
[15] Sylvain Verrona, Jing Li, Teodor Tiplica. Fault detection and isolation of faults in a multivariate process with Bayesian network[J]. Journal of Process Control, 2010, 20: 902-911.
[16] ZHANG Xi, YAN Weiwu, ZHAO Xu, etc. Nonlinear Real-Time Process Monitoring and Fault Diagnosis Based on Principal Component Analysis and Kernel Fisher Discriminant Analysis[J]. Chem. Eng. Technol. 2007, 30, No. 9, 1203–1211.
[17] Bart De Ketelaere, Mia Hubert, Eric Schmitt. Overview of PCA-Based Statistical Process Monitoring Methods for Time-Dependent, High-Dimensional Data[J]. Journal of Quality Technology, 47:4, 318-335.
[18] LIU Xueqin, LI Kang, Marion McAfeeb, etc. Improved nonlinear PCA for process monitoring using support vector data description[J]. Journal of Process Control, 2011, 21: 1306-1317.
[19] Wold S, Esbensen K, Geladi P. Principle component analysis[J]. Chemometrics and Intelligent Laboratory System, 1987, 2(1-3): 37-52.
[20] Jong S D. SIMPLS: an alternative approach to partial least squares regression[J]. Chemometrics and Intelligent Laboratory System, 1993, 18(3): 251-263.
[21] Comon P. Independent component analysis, a new concept[J]. Signal Processing, 1994, 36(3): 287-314.
[22] Chiang L H, Kotanchek M E, Kordon A K. Fault diagnosis based on Fisher discriminant analysis and support vector machines[J]. Computers & Chemical Engineering, 2004, 28(8): 1389-1401.
[23] Jong-Min Lee, ChangKyoo Yoo, Sang Wook Choi, etc. Nonlinear process monitoring using kernel principal component analysis[J]. Chemical Engineering Science, 2004(59): 223-234.
[24] JIANG Qingchao, YAN Xuefeng. Weighted kernel principal component analysis based on probability density estimation and moving window and its application in nonlinear chemical process monitoring[J]. Chemometrics and Intelligent Laboratory Systems, 2013(127): 121–131.
[25] Wang, X.; Kruger, U.; Irwin, G. W. Process Monitoring Approach Using Fast Moving Window PCA[J]. Industrial and Engineering Chemistry Research, 2005, 44: 5691-5702.
[26] Jin, H. D.; Lee, Y. H.; Lee, G.; Han, C. H. Robust recursive principal component analysis modeling for adaptive monitoring[J]. Industrial and Engineering Chemistry Research, 2006, 45: 696-703.
[27] Liu, X. Q.; Kruger, U.; Littler, T. Moving window kernel PCA for adaptive monitoring of nonlinear processes[J]. Chemometrics and Intelligent Laboratory Systems, 2009, 96: 132-143.
[28] JIANG Qingchao, HUANG Biao, YAN Xuefeng. GMM and optimal principal components-based Bayesian method for multi-mode fault diagnosis[J]. Computers and Chemical Engineering, 2016, 84: 338-349.
[29] JIANG Xiaodong, ZHAO Haitao, JIN Bo. Multimode Process Monitoring Based on Sparse Principal Component Selection and Bayesian Inference-Based Probability[J]. Mathematical Problems in Engineering, 2015, https://doi.org/10.1155/2015/465372
[30] Yijun Wu, Diju Liu, Xiaofeng Yuan, et al. A just-in-time fine-tuning framework for deep learning of SAE in adaptive data-driven modeling of time-varying industrial processes[J]. IEEE Sensor Journal, 2021, 21(3): 3497-3505.
[31] GUO Fan, XIE Ruimin, HUANG Biao. A Deep Learning Just-in-Time Modeling Approach for Soft Sensor Based on Variational Autoencoder[J]. Chemometrics and Intelligent Laboratory Systems, 2020, 197, https://doi.org/10.1016/j.chemolab.2019.103922
[32] Aurélien Géron. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow(SECOND EDITION)[M]. Canada: O'Reilly, 2019: 23-31.
[33] Geoffrey E. Hinton, Simon Osindero, Yee-Whye Teh. A Fast Learning Algorithm for Deep Belief Nets. Neural Computation, 2006, 18: 1527-1554.
[34] Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton. ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems 25(NIPS), 2012.
[35] FAN Yajun, TAO Bo, ZHENG Ying,JANG Shishang. A Data-Driven Soft Sensor Based on Multilayer Perceptron Neural Network with a Double LASSO Approach[J]. DOI 10.1109/TIM.2019.2947126, IEEE
[36] ZHANG Zhanpeng, ZHAO Jinsong. A deep belief network based fault diagnosis model for complex chemical processes[J]. Computers and Chemical Engineering, http://dx.doi.org/10.1016/j.compchemeng.2017.02.041
[37] JIANG Qingchao, YAN Xuefeng. Learning Deep Correlated Representations for Nonlinear Process Monitoring[J]. DOI 10.1109/TII.2018.2886048, IEEE
[38] YUAN Xiaofeng, OU Chen, WANG Yalin, et al. A Layer-Wise Data Augmentation Strategy for Deep Learning Networks and Its Soft Sensor Application in an Industrial Hydrocracking Process[J]. IEEE Trans. Neural Networks Learn. Syst. 2021, 32(8): 3296-3305.
[39] WANG Yalin, YANG Haibing, YUAN Xiaofeng, et al. Deep learning for fault-relevant feature extraction and fault classification with stacked supervised auto-encoder[J]. Journal of Process Control, 2020, 92: 79-89.
[40] WANG Yalin, WU Dongzhe, YUAN Xiaofeng. A two-layer ensemble learning framework for data-driven soft sensor of the diesel attributes in an industrial hydrocracking process[J]. Journal of Chemometrics, 2019, 33(12).
[41] GUO Liang, LI Naipeng, JIA Feng, et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings[J]. Neurocomputing, 2017, 240: 98-109.
[42] KE Wensi, HUANG Dexian, YANG Fan, et al. Soft sensor development and applications based on LSTM in deep neural networks[C]. IEEE Symposium Series on Computational Intelligence, SSCI 2017.
[43] WANG Yongjian, LI Hongguang, HUANG Jingwen, et al. An Improved Bar-Shaped Sliding Window CNN Tailored to Industrial Process Historical Data with Applications in Chemical Operational Optimizations. Industrial & Engineering Chemistry Research, 2019, 58(47): 21219-21232.
[44] WANG Yongjian, ZHANG Yichi, LI Hongguang. Adapted Receptive Field Temporal Convolutional Networks with Bar-Shaped Structures Tailored to Industrial Process Operation Models. Industrial & Engineering Chemistry Research, 2020, 59(13): 5482-5490.
[45] GENG Zhiqiang, ZHANG Yanhui, LI Chengfei, et al. Energy optimization and prediction modeling of petrochemical industries: An improved convolutional neural network based on cross-feature. Energy, 2020, 194(1): 116851.
[46] YUAN Xiaofeng, LI Lin, WANG Yalin, et al. Deep learning for quality prediction of nonlinear dynamic process with variable attention-based long short-term memory network[J]. The Canadian Journal of Chemical Engineering, 2020, 98(6): 1377-1389.
[47] 王晨，赵晨曦，杨杰. Pearson相关系数-BP神经网络法用于加氢装置高压换热器结盐分析[C]. 2019年中国智慧炼化高峰论坛论文集，2019: 55-60.
[48] Pablo A. Estévez, Michel Tesmer, Claudio A. Perez, etc. Normalized Mutual Information Feature Selection[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2009, 20(2): 189-201.
[49] Sepp Hochreiter, J urgen Schmidhuber. Long Short-Term Memory[J]. Neural Computation, 1997, 9: 1735–1780.
[50] Ajaya Kumar Pani, Hare Krishna Mohanta. A Survey of Data Treatment Techniques for Soft Sensor Design[J]. Chemical Product and Process Modeling, 2011, 6(1), DOI: 10.2202/1934-2659.1536
[51] A. Di Bella, L. Fortuna, S. Grazianil, et al. A Comparative Analysis of the Influence of Methods for Outliers Detection on the Performance of Data Driven Models[C]. IEEE Instrumentation and Measurement Technology Conference, 2007.