深入探索智能算法与反应网络研究的融合

doi:10.16085/j.issn.1000-6613.2021-1481

摘要/Abstract

摘要：

反应网络是化工过程机理在微观分子尺度上的表达方式，但网络的复杂性为深入认识生产过程提出了挑战。本文提出了探索智能算法与反应网络研究融合的思路，基于物质转化的“透明工程”的概念，深入剖析反应网络的结构统计指标、结构拓扑特征、节点性质特征、机理动态演化、建模应用性能等特点。随后阐述了使用“数据结构化、智能优化与分析、智能代理建模”三步结合的机理数值化反应网络研究方法，既实现了在微观层面的局部放大，又实现了在工业应用中的准确预测。文中指出，智能算法融合反应网络后可以对实际工业过程执行可视化、可解释性的建模、分析与优化，为相关工业生产提质增效提供决策依据，并进一步帮助人类突破认知的极限，更深入地理解反应过程，提取关键的反应规律，助力化学工业的智能化。

关键词: 智能, 算法, 反应, 网络, 机理数值化, 透明工程

Abstract:

Reaction network is a micromolecular-scale expression of the chemical process mechanism, but the complexity of the network poses a challenge for in-depth understanding of the production process. This paper propose the idea of exploring the integration of intelligent algorithm and reaction network, of which the structural statistical indicators, structural topological characteristics, node properties, mechanism dynamic evolution, modeling application performance are analyzed based on the conception of “transparency engineering”. Subsequently, a three-step mechanism-digital reaction network research method combining data structuring, intelligent optimization and analysis, and intelligent surrogate modeling is used, which provide a partial enlargement in micro scale and an accurate prediction in industrial application. The integration of the intelligent algorithms and reaction network implement a visible, explainable modeling, analysis and optimization of the industrial practice, which further provide decision-making recommendations for quality and efficiency enhancement. Additionally, these attempts could help humans to break through the limit of cognition and go deeper into the understanding of the reaction mechanisms, extraction of the key reactions, and intellectualization of the chemical industry.

Key words: intelligent, algorithm, reaction, network, mechanism-digitalization, transparency engineering

中图分类号:

TQ021

毕可鑫, 邱彤. 深入探索智能算法与反应网络研究的融合[J]. 化工进展, 2022, 41(6): 2818-2825.

BI Kexin, QIU Tong. Going deep into the integration of intelligent algorithms and reaction network research[J]. Chemical Industry and Engineering Progress, 2022, 41(6): 2818-2825.

图/表 7

图1 复杂反应网络的Petri-net形式转化

图2 乙烯裂解网络的节点度分布

图3 复杂反应网络的PageRank分析与可视化

图4 基于PageRank值排序结果对反应网络进行智能优化

图5 基于子网重构的关键反应信息智能提取算法

图6 PYD值和RPRV值随乙烷进料物质的量分数的变化趋势

图7 反应特征智能迁移学习算法结构框图CONV—图卷积层；pool—池化层

参考文献 25

1	袁晴棠, 殷瑞钰, 曹湘洪, 等. 面向2035的流程制造业智能化目标、特征和路径战略研究[J]. 中国工程科学, 2020, 22(3): 148-156.
	YUAN Qingtang, YIN Ruiyu, CAO Xianghong, et al. Strategic research on the goals, characteristics, and paths of intelligentization of process manufacturing industry for 2035[J]. Strategic Study of CAE, 2020, 22(3): 148-156.
2	覃伟中, 谢道雄, 赵劲松. 石油化工智能制造[M]. 北京: 化学工业出版社, 2019.
	QIN Weizhong, XIE Daoxiong, ZHAO Jinsong. Intelligent manufacturing of petrochemical industry[M]. Beijing: Chemical Industry Press, 2019.
3	吉旭, 党亚固, 周利, 等. 化学工业多尺度融合的智能制造模式——互联化工[J]. 化工进展, 2020, 39(8): 2927-2936.
	JI Xu, DANG Yagu, ZHOU Li, et al. Interconnected chemical engineering: an intelligent manufacture mode of chemical industry emphasizing the integration of multiscale[J]. Chemical Industry and Engineering Progress, 2020, 39(8): 2927-2936.
4	TURANYI T, TOMLIN A S, PILLING M J. On the error of the quasi-steady-state approximation[J]. The Journal of Physical Chemistry, 1993, 97(1): 163-172.
5	TURÁNYI T, BÉRCES T, VAJDA S. Reaction rate analysis of complex kinetic systems[J]. International Journal of Chemical Kinetics, 1989, 21(2): 83-99.
6	Tamás TURÁNYI. Applications of sensitivity analysis to combustion chemistry[J]. Reliability Engineering & System Safety, 1997, 57(1): 41-48.
7	LU Tianfeng, Chung K LAW. A directed relation graph method for mechanism reduction[J]. Proceedings of the Combustion Institute, 2005, 30(1): 1333-1341.
8	LAM S H, GOUSSIS D A. The CSP method for simplifying kinetics[J]. International Journal of Chemical Kinetics, 1994, 26(4): 461-486.
9	FANG Zhou, QIU Tong, CHEN Bingzhen. Improvement of ethylene cracking reaction network with network flow analysis algorithm[J]. Computers & Chemical Engineering, 2016, 91: 182-194.
10	HUA Feng, FANG Zhou, QIU Tong. Application of convolutional neural networks to large-scale naphtha pyrolysis kinetic modeling[J]. Chinese Journal of Chemical Engineering, 2018, 26(12): 2562-2572.
11	BI Kexin, ZHANG Chen, QIU Tong. An ingenious characterization of reaction network using sub-network reconstruction[J]. Computers & Chemical Engineering, 2020, 134: 106695.
12	KOCH Ina. Petri nets—A mathematical formalism to analyze chemical reaction networks[J]. Molecular Informatics, 2010, 29(12): 838-843.
13	GAO Connie W, ALLEN Joshua W, GREEN William H, et al. Reaction mechanism generator: automatic construction of chemical kinetic mechanisms[J]. Computer Physics Communications, 2016, 203: 212-225.
14	LIU Mengjie, DANA Alon Grinberg, JOHNSON Matthew S, et al. Reaction mechanism generator V3.0: advances in automatic mechanism generation[J]. Journal of Chemical Information and Modeling, 2021, 61(6): 2686-2696.
15	RANGARAJAN Srinivas, BHAN Aditya, DAOUTIDIS Prodromos. Language-oriented rule-based reaction network generation and analysis: description of RING[J]. Computers & Chemical Engineering, 2012, 45: 114-123.
16	RANGARAJAN Srinivas, BHAN Aditya, DAOUTIDIS Prodromos. Language-oriented rule-based reaction network generation and analysis: applications of RING[J]. Computers & Chemical Engineering, 2012, 46: 141-152.
17	TANGMUNARUNKIT Hongsuda, GOVINDAN Ramesh, JAMIN Sugih, et al. Network topology generators[J]. Computer Communication Review, 2002, 32(4): 147-159.
18	WATTS Duncan J, STROGATZ Steven H. Collective dynamics of ‘small-world’ networks[J]. Nature, 1998, 393(6684): 440-442.
19	PAGE L, BRIN S, MOTWANI R, et al. The pagerank citation ranking: bringing order to the web[J]. Stanford Digital Libraries Working Paper, 1998, DOI：10.1007/978-3-319-08789-4_10 .
20	CHEN Jie, SAAD Yousef. Dense subgraph extraction with application to community detection[J]. IEEE Transactions on Knowledge and Data Engineering, 2012, 24(7): 1216-1230.
21	许江, 马艳捷, 黄剑锋, 等. 乙烯裂解装置的节能增效技术分析[J]. 能源化工, 2017, 38(1): 17-21.
	XU Jiang, MA Yanjie, HUANG Jianfeng, et al. Analysis on technologies of energy saving and efficiency improvement in ethylene cracking plant[J]. Energy Chemical Industry, 2017, 38(1): 17-21.
22	纪晔, 王杭州, 蒋子龙, 等. 基于结构导向集总的分子级催化重整过程建模方法[J]. 计算机与应用化学, 2018, 35(9): 767-774.
	JI Ye, WANG Hangzhou, JIANG Zilong, et al. Molecular level modeling method of catalytic reforming based on structure-oriented lumping concept[J]. Computers and Applied Chemistry, 2018, 35(9): 767-774.
23	纪晔, 王杭州, 蒋子龙, 等. 基于结构导向集总的分子级炼油过程全厂模拟和优化方法[J]. 计算机与应用化学, 2020(3): 218-225.
	JI Ye, WANG Hangzhou, JIANG Zilong, et al. Modelling and optimization for the overall refinery operation based on the structure-oriented lumping concept[J]. Computers and Applied Chemistry, 2020(3): 218-225.
24	BILLA Triveni, HORTON Scott R, SAHASRABUDHE Mayuresh, et al. Enhancing the value of detailed kinetic models through the development of interrogative software applications[J]. Computers & Chemical Engineering, 2017, 106: 512-528.
25	BI Kexin, BEYKAL Burcu, AVRAAMIDOU Styliani, et al. Integrated modeling of transfer learning and intelligent heuristic optimization for steam cracking process[J]. Industrial & Engineering Chemistry Research, 2020, 59(37): 16357-16367.

[1]	许家珩, 李永胜, 罗春欢, 苏庆泉. 甲醇水蒸气重整工艺的优化[J]. 化工进展, 2023, 42(S1): 41-46.
[2]	孙玉玉, 蔡鑫磊, 汤吉海, 黄晶晶, 黄益平, 刘杰. 反应精馏合成甲基丙烯酸甲酯工艺优化及节能[J]. 化工进展, 2023, 42(S1): 56-63.
[3]	盛维武, 程永攀, 陈强, 李小婷, 魏嘉, 李琳鸽, 陈险峰. 微气泡和微液滴双强化脱硫反应器操作分析[J]. 化工进展, 2023, 42(S1): 142-147.
[4]	黄益平, 李婷, 郑龙云, 戚傲, 陈政霖, 史天昊, 张新宇, 郭凯, 胡猛, 倪泽雨, 刘辉, 夏苗, 主凯, 刘春江. 三级环流反应器中气液流动与传质规律[J]. 化工进展, 2023, 42(S1): 175-188.
[5]	徐晨阳, 都健, 张磊. 基于图神经网络的化学反应优劣评价[J]. 化工进展, 2023, 42(S1): 205-212.
[6]	林晓鹏, 肖友华, 管奕琛, 鲁晓东, 宗文杰, 傅深渊. 离子聚合物-金属复合材料（IPMC）柔性电极的研究进展[J]. 化工进展, 2023, 42(9): 4770-4782.
[7]	王鹏, 史会兵, 赵德明, 冯保林, 陈倩, 杨妲. 过渡金属催化氯代物的羰基化反应研究进展[J]. 化工进展, 2023, 42(9): 4649-4666.
[8]	程涛, 崔瑞利, 宋俊男, 张天琪, 张耘赫, 梁世杰, 朴实. 渣油加氢装置杂质沉积规律与压降升高机理分析[J]. 化工进展, 2023, 42(9): 4616-4627.
[9]	张帆, 陶少辉, 陈玉石, 项曙光. 基于改进恒热传输模型的精馏模拟初始化[J]. 化工进展, 2023, 42(9): 4550-4558.
[10]	罗成, 范晓勇, 朱永红, 田丰, 崔楼伟, 杜崇鹏, 王飞利, 李冬, 郑化安. 中低温煤焦油加氢反应器不同分配器中液体分布的CFD模拟[J]. 化工进展, 2023, 42(9): 4538-4549.
[11]	刘炫麟, 王驿凯, 戴苏洲, 殷勇高. 热泵中氨基甲酸铵分解反应特性及反应器结构优化[J]. 化工进展, 2023, 42(9): 4522-4530.
[12]	邓建, 王凯, 骆广生. 面向硝基化学品安全生产的绝热连续微反应技术发展及思考[J]. 化工进展, 2023, 42(8): 3923-3925.
[13]	吴正浩, 周天航, 蓝兴英, 徐春明. 人工智能驱动化学品创新设计的实践与展望[J]. 化工进展, 2023, 42(8): 3910-3916.
[14]	毛善俊, 王哲, 王勇. 基团辨识加氢：从概念到应用[J]. 化工进展, 2023, 42(8): 3917-3922.
[15]	吕程远, 张函, 杨明旺, 杜健军, 樊江莉. 生物成像用二氧杂环丁烷余辉发光体系的研究进展[J]. 化工进展, 2023, 42(8): 4108-4122.