基于BP神经网络模型优化Fe1-x O基氨合成催化剂

doi:10.16085/j.issn.1000-6613.2023-0433

摘要/Abstract

摘要：

运用BP神经网络建立了助催化剂含量与催化剂活性之间的预测模型，对Fe_1-_x O基氨合成催化剂的助催化剂进行优化。首先将前期实验数据整理归纳为含有3、4、5、6和7个助催化剂等5类催化剂，以助催化剂含量（体积分数）为输入变量，以425℃反应器出口氨浓度（活性）为输出变量，对助催化剂进行优化。结果表明，BP神经网络预测模型拟合值均方误差最高为0.2784，预测值均方误差最高为0.1592，构建的BP神经网络模型准确度较高。在该模型的基础上，运用多种群遗传算法进行极值寻优，求解最优的催化剂配方，并进行实验验证。结果表明，根据优化结果制备5个样品的实验测定值与预测值的相对误差最高为2.88%，优化结果较为准确；含有7个助催化剂的催化剂活性最高为18.83%，比原样本的统计平均活性值（17.52%）高1.31%，相对提高7.48%，助催化剂含量优化取得满意的结果。

关键词: Fe_1-_x O, 催化剂, 助催化剂, 神经网络, 遗传算法, 优化

Abstract:

A prediction model between the content of promoter and the activity of catalyst was established by BP neural network, with which the promoter of Fe_1-_x O ammonia synthesis catalyst was optimized. Firstly, the preliminary experimental data were summarized into five types of catalysts including three, four, five, six and seven promoters. With the content of the promoters (volume fraction) as the input model variable and the ammonia concentration (reactivity) at the outlet of the reactor at 425℃ as the output one, the formula of the promoter was optimized. The results showed that maximum mean square error of fitting values of BP neural network prediction model was 0.2784, while that of the predicted values was 0.1592, indicating the accuracy of the BP neural network model was high. On the basis of this model, multiple population genetic algorithm was used to search the extreme value, and the optimal catalyst formula was obtained and verified by experiments. The maximum relative error between the measured values of 5 samples prepared according to the optimized formula and the predicted ones was 2.88%. The highest activity was 18.83% for the catalyst containing seven promoters, 1.31% higher than the average reactivity value of the original sample (17.52%), and a relative increase of 7.48%.

Key words: Fe_1-_x O, catalyst, promoters, neural networks, genetic algorithm, optimization

中图分类号:

O643.36

张书铭, 刘化章. 基于BP神经网络模型优化Fe_1-_x O基氨合成催化剂[J]. 化工进展, 2024, 43(3): 1302-1308.

ZHANG Shuming, LIU Huazhang. Optimization of Fe_1-_x O ammonia synthesis catalyst by BP neural network model[J]. Chemical Industry and Engineering Progress, 2024, 43(3): 1302-1308.

图/表 10

参考文献 27

1	LIU Huazhang, LI Xiaonian. The precursor phase composition of iron catalyst and discovery of FeO based catalyst for ammonia synthesis[J]. Science China-Chemistry, 1995, 38: 529-537.
2	LIU Huazhang. Ammonia synthesis catalyst 100 years: Practice, enlightenment and challenge[J]. Chinese Journal of Catalysis, 2014, 35(10): 1619-1640.
3	刘化章, 李小年, 胡樟能, 等. Fe_1- _x O基氨合成催化剂的制备化学[J]. 高等学校化学学报, 2002, 23(1): 87-91.
	LIU Huazhang, LI Xiaonian, HU Zhangneng, et al. Preparation chemistry of Fe_1- _x O based catalyst for ammonia synthesis[J]. Chemical Research in Chinese Universities, 2002, 23(1): 87-91.
4	LIU Huazhang, LI Xiaonian. Study on the composition of parent phase of iron Catalyst and Discovery of FeO based ammonia synthesis catalyst[J]. Science in China, 1995, 25(1): 1-6.
5	陈志军, 严海宇, 蓝国钧，等. 助催化剂对Fe_1- _x O基氨合成催化剂性能的影响[J]. 工业催化, 2013, 21(5): 41-45.
	CHEN Zhijun, YAN Haiyu, LAN Guojun, et al. Influence of promoters on the performance of Fe_1- _x O based catalyst for ammonia synthesis[J]. Industrial Catalysis, 2013, 21(5): 41-45.
6	李小年, 刘化章, 陈诵英. 助催化剂对Fe_1- _x O基氨合成催化剂活性的影响[J]. 催化学报, 1998, 19(3): 201-205.
	LI Xiaonian, LIU Huazhang, CHEN Songying. Studies on the effect of promoters on the activity of Fe_1- _x O based ammonia synthesis[J]. Chinese Journal of Catalysis, 1998, 19(3): 201-205.
7	陈志军. Fe_1- _x O基氨合成催化剂助催化剂的选择与优化[D]. 杭州: 浙江工业大学, 2013.
	CHEN Zhijun. Exprolation and optimization of promoters of Fe_1- _x O based catalysts for ammonia synthesis[D]. Hangzhou: Zhejiang University of Technology, 2013.
8	李林辉, 刘化章, 韩文锋, 等. SrO助催化剂对Fe_1- _x O基熔铁氨合成催化剂性能的影响[J]. 化工进展, 2019, 38(3): 1371-1376.
	LI Linhui, LIU Huazhang, HAN Wenfeng, et al. Effect of SrO promoter on the activity and thermal-stability of wüstite based catalyst for ammonia synthesis[J]. Chemical Industry and Engineering Progress, 2019, 38(3): 1371-1376.
9	孙珍珍, 刘化章, 叶攀, 等. Fe_1- _x O基氨合成催化剂助催化剂的优选[J]. 化工进展, 2022,41(4): 1886-1893.
	SUN Zhenzhen, LIU Huazhang, YE Pan, et al. Optimization of promoters for Fe_1- _x O-based ammonia synthesis catalyst[J]. Chemical Industry and Engineering Progress, 2022,41(4): 1886-1893.
10	HAN Wenfeng, HUANG Shiliang, CHENG Tianhong, et al. Promotion of Nb₂O₅ on the wüstite-based iron catalyst for ammonia synthesis[J]. Applied Surface Science, 2015, 353: 17-23.
11	LIU Huazhang, LI Xiaonian. Precursor of iron catalyst for ammonia synthesis: Fe₃O₄, Fe_1- _x O, Fe₂O₃ or their mixture?[J]. Studies in Surface Science and Catalysis, 2000, 130: 2207-2212.
12	郑启富. 计算智能在氨合成催化剂设计与工艺优化中的应用[D]. 杭州: 浙江工业大学, 2012.
	ZHENG Qifu. Application of computational intelligence in catalyst design and process optimization of ammonia synthesis[D]. Hangzhou: Zhejiang University of Technology, 2012.
13	AGHAREZAEI P, SAHU T, SHOCK J, et al. Designing catalysts via evolutionary-based optimization techniques[J]. Computational Materials Science, 2023, 216: 111833.
14	刘方, 徐龙, 马晓迅. BP神经网络的发展及其在化学化工中的应用[J]. 化工进展, 2019, 38(6): 2559-2573.
	LIU Fang, XU Long, MA Xiaoxun. Development of BP neural network and its application in chemistry and chemical engineering[J]. Chemical Industry and Engineering Progress, 2019, 38(6): 2559-2573.
15	GARONA H A, CAVALCANTI F M, DE ABREU T F, et al. Evaluation of Fischer-Tropsch synthesis to light olefins over Co- and Fe-based catalysts using artificial neural network[J]. Journal of Cleaner Production, 2021, 321: 129003.
16	温正. 精通MATLAB智能算法[M]. 北京: 清华大学出版社, 2015.
	WEN Zheng. Proficient in MATLAB intelligent algorithm[M]. Beijing: Tsinghua University Press, 2015.
17	雷英杰, 张善文. MATLAB遗传算法工具箱及应用[M]. 2版. 西安: 西安电子科技大学出版社, 2014.
	LEI Yingjie, ZHANG Shanwen. MATLAB genetic algorithm toolbox and its application[M]. 2nd ed. Xi’an: Xidian University Press, 2014.
18	刘鹏程, 李新利. 基于多种群遗传算法的含分布式电源的配电网故障区段定位算法[J]. 电力系统保护与控制, 2016, 44(2): 36-41.
	LIU Pengcheng, LI Xinli. Fault-section location of distribution network containing distributed generation based on the multiple-population genetic algorithm[J]. Power System Protection and Control, 2016, 44(2): 36-41.
19	石磊, 高卫红, 吕莉莉, 等. 基于BP人工神经网络和遗传算法的葛根总黄酮提取工艺优化研究[J]. 中国中医急症, 2018, 27(2): 198-201.
	SHI Lei, GAO Weihong, Lili LYU, et al. Research on extraction process optimization for total flavones in radix puerariae based on back propagation artificial neural network and genetic algorithm[J]. Journal of Emergency in Traditional Chinese Medicine, 2018, 27(2): 198-201.
20	MESBAH M, SOLTANALI S, BAHRANIFARD Z, et al. Production of thymol from alkylation of m-cresol with isopropanol over ZSM-5 catalysts: Artificial neural network (ANN) modelling[J]. Journal of the Indian Chemical Society, 2023, 100(2): 100882.
21	PAJAK M, BUCHANIEC S, KIMIJIMA S, et al. A multiobjective optimization of a catalyst distribution in a methane/steam reforming reactor using a genetic algorithm[J]. International Journal of Hydrogen Energy, 2021, 46(38): 20183-20197.
22	ZAFARI M, KUMAR D, UMER M, et al. Machine learning-based high throughput screening for nitrogen fixation on boron-doped single atom catalysts[J]. Journal of Materials Chemistry A, 2020, 8(10): 5209-5216.
23	CHANDANA K S, KARKA S, GUJRAL M K, et al. Machine learning aided catalyst activity modelling and design for direct conversion of CO₂ to lower olefins[J]. Journal of Environmental Chemical Engineering, 2023, 11(2): 109555.
24	QIU Huixian, XIA Xuewen, LI Yuanxiang, et al. A dynamic multipopulation genetic algorithm for multiobjective workflow scheduling based on the longest common sequence[J]. Swarm and Evolutionary Computation, 2023, 78: 101291.
25	YE Jiahao, PENG Qingguo. Improved emissions conversion of diesel oxidation catalyst using multifactor impact analysis and neural network[J]. Energy, 2023, 271: 127048.
26	FAN Haonan, YANG Xiaodie, MA Jing, et al. Computer-assisted design of asymmetric PNP ligands for ethylene tri-/tetramerization: A combined DFT and artificial neural network approach[J]. Journal of Catalysis, 2023, 418: 121-129.
27	刘科研, 盛万兴, 马晓晨, 等. 基于多种群遗传算法的分布式光伏接入配电网规划研究[J]. 太阳能学报, 2021, 42(6): 146-155.
	LIU Keyan, SHENG Wanxing, MA Xiaochen, et al. Planning research of distributed photovoltaic source access distribution network based on multi-population genetic algorithm[J]. Acta Energiae Solaris Sinica, 2021, 42(6): 146-155.

助催化剂数目	助催化剂种类	样本数	出口氨浓度（活性）平均值/%
3	Al₂O₃、K₂O、CaO	10	15.74
4	Al₂O₃、K₂O、CaO、A	11	16.09
5	Al₂O₃、K₂O、CaO、A、B	14	16.14
6	Al₂O₃、K₂O、CaO、A、B、C	30	17.40
7	Al₂O₃、K₂O、CaO、A、B、C、D	30	17.52

助催化剂数目	助催化剂种类	样本数	出口氨浓度（活性）平均值/%
3	Al₂O₃、K₂O、CaO	10	15.74
4	Al₂O₃、K₂O、CaO、A	11	16.09
5	Al₂O₃、K₂O、CaO、A、B	14	16.14
6	Al₂O₃、K₂O、CaO、A、B、C	30	17.40
7	Al₂O₃、K₂O、CaO、A、B、C、D	30	17.52

助催化剂数目	节点数
助催化剂数目	输入层	隐含层	输出层
3	3	7	1
4	4	9	1
5	5	8	1
6	6	6	1
7	7	5	1

助催化剂数目	节点数
助催化剂数目	输入层	隐含层	输出层
3	3	7	1
4	4	9	1
5	5	8	1
6	6	6	1
7	7	5	1

操作参数	助催化剂数目
操作参数	3	4	5	6	7
种群数量G	5	5	5	10	10
种群规模N	20	20	30	40	40
交叉概率P_c	0.6~0.9	0.6~0.9	0.6~0.9	0.6~0.9	0.6~0.9
变异概率P_m	0.1~0.4	0.1~0.4	0.1~0.4	0.1~0.4	0.1~0.4
最大繁衍代数	100	100	100	100	100