基于人工智能的微分散基础研究

doi:10.11949/0438-1157.20231406

摘要/Abstract

摘要：

微分散是微化工技术的重要组成部分，其装备和过程的复杂性使得相关研究受到诸多限制，传统思想指导下微分散基础研究以“设计-实验-模型”为思路，进展缓慢。近年来，人工智能方法因其强大的识别和回归能力在化工领域倍受关注，人工智能辅助的微分散基础研究有利于形成微化工过程认识的新范式，促进微化工技术发展。本文介绍了人工智能方法的一般思路及其针对微分散研究的适用性，综述了人工智能技术在显微图像识别、液滴和气泡分散尺寸预测以及微分散过程控制和优化中的研究进展，对未来基于人工智能的微化工研究进行了展望。

关键词: 微流体学, 分散, 模型, 算法, 人工智能

Abstract:

Microdispersion is an important part of micro chemical engineering technology. The complexity of equipment and processes imposes many limitations on its related research. Under the guidance of traditional thinking, the basic research on microdispersion follows the approach of "design - experiment - modeling", resulting in slow progress. In recent years, artificial intelligence (AI) methods have received increasing attention in chemical engineering due to their powerful recognition and regression capabilities. AI-assisted basic research on microdispersion is conducive to forming a new paradigm for understanding micro chemical engineering processes, promoting the development of micro chemical engineering technology. This article introduces the general idea of AI methods and their applicability to microdispersion investigation; reviews the developments of AI technology in microscopic image recognition, droplet and bubble dispersion size prediction, and microdispersion process control and optimization; and offers insights into the prospective directions of micro chemical engineering based on AI technology.

Key words: microfluidics, dispersion, model, algorithm, artificial intelligence

中图分类号:

TQ 021.1

刘梦绮, 王凯, 骆广生. 基于人工智能的微分散基础研究[J]. 化工学报, DOI: 10.11949/0438-1157.20231406.

Mengqi LIU, Kai WANG, Guangsheng LUO. Fundamental research on microdispersion based on artificial intelligence[J]. CIESC Journal, DOI: 10.11949/0438-1157.20231406.

图/表 5

参考文献 55

1	Kim B, Jeon T Y, Oh Y K, et al. Microfluidic production of semipermeable microcapsules by polymerization-induced phase separation[J]. Langmuir, 2015, 31(22): 6027-6034.
2	Porta R, Benaglia M, Puglisi A. Flow chemistry: recent developments in the synthesis of pharmaceutical products[J]. Organic Process Research & Development, 2016, 20(1): 2-25.
3	Matuła K, Rivello F, Huck W T S. Single-cell analysis using droplet microfluidics[J]. Advanced Biosystems, 2020, 4(1): e1900188.
4	Wang K, Luo G. Microflow extraction: a review of recent development[J]. Chemical Engineering Science, 2017, 169: 18-33.
5	Zhu P G, Wang L Q. Passive and active droplet generation with microfluidics: a review[J]. Lab on a Chip, 2017, 17(1): 34-75.
6	Thorsen T, Roberts R W, Arnold F H, et al. Dynamic pattern formation in a vesicle-generating microfluidic device[J]. Physical Review Letters, 2001, 86(18): 4163-4166.
7	Anna S L, Bontoux N, Stone H A. Formation of dispersions using ``flow focusing'' in microchannels[J]. Applied Physics Letters, 2003, 82(3): 364.
8	Cramer C, Fischer P, Windhab E J. Drop formation in a co-flowing ambient fluid[J]. Chemical Engineering Science, 2004, 59(15): 3045-3058.
9	Wang K, Lu Y C, Xu J H, et al. Droplet generation in micro-sieve dispersion device[J]. Microfluidics and Nanofluidics, 2011, 10(5): 1087-1095.
10	Wang K, Lu Y C, Shao H W, et al. Heat-transfer performance of a Liquid-Liquid microdispersed system[J]. Industrial & Engineering Chemistry Research, 2008, 47(23): 9754-9758.
11	Sheng L, Wang K, Deng J, et al. Gas-liquid microdispersion and microflow for carbon dioxide absorption and utilization: a review[J]. Current Opinion in Chemical Engineering, 2023, 40: 100917.
12	Lashkaripour A, Rodriguez C, Ortiz L, et al. Performance tuning of microfluidic flow-focusing droplet generators[J]. Lab on a Chip, 2019, 19(6): 1041-1053.
13	Faustino V, Catarino S O, Lima R, et al. Biomedical microfluidic devices by using low-cost fabrication techniques: a review[J]. Journal of Biomechanics, 2016, 49(11): 2280-2292.
14	Battat S, Weitz D A, Whitesides G M. An outlook on microfluidics: the promise and the challenge[J]. Lab on a Chip, 2022, 22(3): 530-536.
15	Kashid M N., Renken A, Kiwi-Minsker L. Influence of flow regime on mass transfer in different types of microchannels[J]. Industrial & Engineering Chemistry Research, 2011, 50(11): 6906-6914.
16	Garstecki P, Fuerstman M J, Stone H A, et al. Formation of droplets and bubbles in a microfluidic T-junction—scaling and mechanism of break-up[J]. Lab on a Chip, 2006, 6(3): 437-446.
17	Korczyk P M, van Steijn V, Blonski S, et al. Accounting for corner flow unifies the understanding of droplet formation in microfluidic channels[J]. Nature Communications, 2019, 10: 2528.
18	陈宇超, 崔永晋, 王凯, 等. 阶梯式T型微通道内液滴、气泡分散规律[J]. 化工学报, 2020, 71(1): 265-273.
	Chen Y C, Cui Y J, Wang K, et al. Droplet and bubble dispersion in step T-junction microchannel[J]. CIESC Journal, 2020, 71(1): 265-273.
19	Shen C, Zheng Q B, Shang M J, et al. Using deep learning to recognize liquid–liquid flow patterns in microchannels[J]. AIChE Journal, 2020, 66(8): e16260.
20	Nunes J K, Tsai S S H, Wan J, et al. Dripping and jetting in microfluidic multiphase flows applied to particle and fiber synthesis[J]. Journal of Physics D: Applied Physics, 2013, 46(11): 114002.
21	Christopher G F, Anna S L. Microfluidic methods for generating continuous droplet streams[J]. Journal of Physics D: Applied Physics, 2007, 40(19): R319-R336.
22	Asadi-Saghandi H, Karimi-Sabet J, Ghorbanian S, et al. Dimensionless analysis on liquid–liquid two-phase flow patterns in a numbered-up microfluidic device[J]. Chemical Engineering Journal, 2022, 429: 132428.
23	Liu Z M, Zhao J, Pang Y, et al. Generation of droplets in the T-junction with a constriction microchannel[J]. Microfluidics and Nanofluidics, 2018, 22(11): 124.
24	Odera T, Hirama H, Kuroda J, et al. Droplet formation behavior in a microfluidic device fabricated by hydrogel molding[J]. Microfluidics and Nanofluidics, 2014, 17(3): 469-476.
25	Xu J H, Li S W, Tan J, et al. Correlations of droplet formation in T-junction microfluidic devices: from squeezing to dripping[J]. Microfluidics and Nanofluidics, 2008, 5(6): 711-717.
26	Li Y K, Liu G T, Xu J H, et al. A microdevice for producing monodispersed droplets under a jetting flow[J]. RSC Advances, 2015, 5(35): 27356-27364.
27	Fu T T, Ma Y G, Li H Z. Breakup dynamics of slender droplet formation in shear-thinning fluids in flow-focusing devices[J]. Chemical Engineering Science, 2016, 144: 75-86.
28	Lu Y T, Fu T T, Zhu C Y, et al. Scaling of the bubble formation in a flow-focusing device: role of the liquid viscosity[J]. Chemical Engineering Science, 2014, 105: 213-219.
29	Riordon J, Sovilj D, Sanner S, et al. Deep learning with microfluidics for biotechnology[J]. Trends in Biotechnology, 2019, 37(3): 310-324.
30	Srikanth S, Dubey S K, Javed A, et al. Droplet based microfluidics integrated with machine learning[J]. Sensors and Actuators A: Physical, 2021, 332: 113096.
31	McIntyre D, Lashkaripour A, Fordyce P, et al. Machine learning for microfluidic design and control[J]. Lab on a Chip, 2022, 22(16): 2925-2937.
32	Goda K, Lu H, Fei P, et al. Revolutionizing microfluidics with artificial intelligence: a new dawn for lab-on-a-chip technologies[J]. Lab on a Chip, 2023, 23(17): 3737-3740.
33	Isozaki A, Harmon J, Zhou Y Q, et al. AI on a chip[J]. Lab on a Chip, 2020, 20(17): 3074-3090.
34	Dedeloudi A, Weaver E, Lamprou D A. Machine learning in additive manufacturing & Microfluidics for smarter and safer drug delivery systems[J]. International Journal of Pharmaceutics, 2023, 636: 122818.
35	Zhang S Y, Li H R, Wang K, et al. Accelerating intelligent microfluidic image processing with transfer deep learning: a microchannel droplet/bubble breakup case study[J]. Separation and Purification Technology, 2023, 315: 123703.
36	Chu A, Nguyen D, Talathi S S, et al. Automated detection and sorting of microencapsulation via machine learning[J]. Lab on a Chip, 2019, 19(10): 1808-1817.
37	Zhang S Y, Qin K, Huang X Y, et al. Insight into microdispersion flows with a novel video deep learning method[J]. Advanced Intelligent Systems, 2022, 4(11): 2200098.
38	Durve M, Bonaccorso F, Montessori A, et al. A fast and efficient deep learning procedure for tracking droplet motion in dense microfluidic emulsions[J]. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2021, 379(2208): 20200400.
39	Mahdi Y, Daoud K. Microdroplet size prediction in microfluidic systems via artificial neural network modeling for water-in-oil emulsion formulation[J]. Journal of Dispersion Science and Technology, 2017, 38(10): 1501-1508.
40	Mottaghi S, Nazari M, Fattahi S M, et al. Droplet size prediction in a microfluidic flow focusing device using an adaptive network based fuzzy inference system[J]. Biomedical Microdevices, 2020, 22(3): 61.
41	Lashkaripour A, Goharimanesh M, Abouei Mehrizi A, et al. An adaptive neural-fuzzy approach for microfluidic droplet size prediction[J]. Microelectronics Journal, 2018, 78: 73-80.
42	Damiati S A, Rossi D, Joensson H N, et al. Artificial intelligence application for rapid fabrication of size-tunable PLGA microparticles in microfluidics[J]. Scientific Reports, 2020, 10: 19517.
43	Lashkaripour A, Rodriguez C, Mehdipour N, et al. Machine learning enables design automation of microfluidic flow-focusing droplet generation[J]. Nature Communications, 2021, 12: 25.
44	Chang Y, Sheng L, Wang J J, et al. A general neural network model co-driven by mechanism and data for the reliable design of gas–liquid T-junction microdevices[J]. Lab on a Chip, 2023, 23(22): 4888-4900.
45	Chagot L, Quilodrán-Casas C, Kalli M, et al. Surfactant-laden droplet size prediction in a flow-focusing microchannel: a data-driven approach[J]. Lab on a Chip, 2022, 22(20): 3848-3859.
46	Wang J X, Yu W, Wu Z, et al. Physics-based statistical learning perspectives on droplet formation characteristics in microfluidic cross-junctions[J]. Applied Physics Letters, 2022, 120(20): 204101.
47	Ali H S M, Blagden N, York P, et al. Artificial neural networks modelling the prednisolone nanoprecipitation in microfluidic reactors[J]. European Journal of Pharmaceutical Sciences, 2009, 37(3): 514-522.
48	Khor J W, Jean N, Luxenberg E S, et al. Using machine learning to discover shape descriptors for predicting emulsion stability in a microfluidic channel[J]. Soft Matter, 2019, 15(6): 1361-1372.
49	Han S, Kim T, Kim D, et al. Use of deep learning for characterization of microfluidic soft sensors[J]. IEEE Robotics and Automation Letters, 2018, 3(2): 873-880.
50	Dressler O J, Howes P D, Choo J, et al. Reinforcement learning for dynamic microfluidic control[J]. ACS Omega, 2018, 3(8): 10084-10091.
51	Gyimah N, Scheler O, Rang T, et al. Deep reinforcement learning-based digital twin for droplet microfluidics control[J]. Physics of Fluids, 2023, 35(8): 082020.
52	Fukada K, Seyama M. Microfluidic devices controlled by machine learning with failure experiments[J]. Analytical Chemistry, 2022, 94(19): 7060-7065.
53	Siemenn A E, Shaulsky E, Beveridge M, et al. A machine learning and computer vision approach to rapidly optimize multiscale droplet generation[J]. ACS Applied Materials & Interfaces, 2022, 14(3): 4668-4679.
54	Mekki-Berrada F, Ren Z K, Huang T, et al. Two-step machine learning enables optimized nanoparticle synthesis[J]. npj Computational Materials, 2021, 7: 55.
55	Probst P Boulesteix A L, Bischl B. Tunability: importance of hyperparameters of machine learning algorithms The journal of machine learning research, 2019, 20(1): 1934–1965.

几何结构	半经验关联式^a	适用范围	参考文献
T型微通道	d/W =2.7(Ca_d/Ca_c)^0.19	0.0003<Ca_c<0.1	[24]
毛细管嵌入T型微通道	d/W = 0.75(Q_d/Q_c)^1/3Ca_c^–1/5	0.002<Ca_c<0.01	[25]
毛细管嵌入阶梯T型微通道	d/W = 1.10(Q_d/Q_c)^0.49(μ_d/μ_c)^0.02	Ca_c>0.07	[26]
聚焦流微通道	d/W = 0.88(Q_d/Q_c)^0.18Ca_d^–0.15	0.002<Ca_c<0.223	[27]
聚焦流微通道	d/W = 1.29(Q_d/Q_c)^0.17Ca_c^–0.1	0.00065<Ca_c<0.2	[28]

几何结构	半经验关联式^a	适用范围	参考文献
T型微通道	d/W =2.7(Ca_d/Ca_c)^0.19	0.0003<Ca_c<0.1	[24]
毛细管嵌入T型微通道	d/W = 0.75(Q_d/Q_c)^1/3Ca_c^–1/5	0.002<Ca_c<0.01	[25]
毛细管嵌入阶梯T型微通道	d/W = 1.10(Q_d/Q_c)^0.49(μ_d/μ_c)^0.02	Ca_c>0.07	[26]
聚焦流微通道	d/W = 0.88(Q_d/Q_c)^0.18Ca_d^–0.15	0.002<Ca_c<0.223	[27]
聚焦流微通道	d/W = 1.29(Q_d/Q_c)^0.17Ca_c^–0.1	0.00065<Ca_c<0.2	[28]

模型类型	输入特征	输出目标	数据量	尺寸预测准确性	文献
ANN	Re_d、Re_c、Ca_c、Ca_d	液滴尺寸	742	R²≈1	[39]
ANFIS	Ca、μ_d/μ_c、Q_d/Q_c	液滴尺寸	36	R²=0.92	[40]
ANFIS	狭缝宽度Or、Q_c、Q_d、μ_c、μ_d、ρ_c	液滴尺寸	51	R²=0.96	[41]
ANN	聚合物浓度、Q_c、Q_d、微通道组合形式	液滴或颗粒尺寸	223	R²=0.971	[42]
ANN	Ca、Q_d/Q_c、连续相通道宽度W_c、分散相通道宽度W_d、Or、缩口长度L、下游通道宽度W_m、通道深度H	液滴尺寸生成频率	998	R²=0.893（滴状流） R²=0.966（射流）	[43]
DNN	We_d、We_c、Ca、Q_d/Q_c、W_d、W_c、W_m、α_lm、α_gm	气泡尺寸	854	MRE=2.02%	[44]
BRANN XGBoost	Q_c、Q_d、γ、浓度比ϕ/ϕ_CMC	液滴尺寸	468	MAPE=3.9%	[45]
DNN Deep-FM	Ca、通道夹角、Q_d/Q_c、μ_d/μ_c	液滴尺寸	298	MRE=4.09%	[46]
ANN	饱和度、Q_c、Q_d、通道夹角、W	颗粒尺寸	71	R²=0.89	[47]

模型类型	输入特征	输出目标	数据量	尺寸预测准确性	文献
ANN	Re_d、Re_c、Ca_c、Ca_d	液滴尺寸	742	R²≈1	[39]
ANFIS	Ca、μ_d/μ_c、Q_d/Q_c	液滴尺寸	36	R²=0.92	[40]
ANFIS	狭缝宽度Or、Q_c、Q_d、μ_c、μ_d、ρ_c	液滴尺寸	51	R²=0.96	[41]
ANN	聚合物浓度、Q_c、Q_d、微通道组合形式	液滴或颗粒尺寸	223	R²=0.971	[42]
ANN	Ca、Q_d/Q_c、连续相通道宽度W_c、分散相通道宽度W_d、Or、缩口长度L、下游通道宽度W_m、通道深度H	液滴尺寸生成频率	998	R²=0.893（滴状流） R²=0.966（射流）	[43]
DNN	We_d、We_c、Ca、Q_d/Q_c、W_d、W_c、W_m、α_lm、α_gm	气泡尺寸	854	MRE=2.02%	[44]
BRANN XGBoost	Q_c、Q_d、γ、浓度比ϕ/ϕ_CMC	液滴尺寸	468	MAPE=3.9%	[45]
DNN Deep-FM	Ca、通道夹角、Q_d/Q_c、μ_d/μ_c	液滴尺寸	298	MRE=4.09%	[46]
ANN	饱和度、Q_c、Q_d、通道夹角、W	颗粒尺寸	71	R²=0.89	[47]

[1]	麻雪怡, 刘克勤, 胡激江, 姚臻. POE溶液聚合反应器内混合与反应过程的CFD研究[J]. 化工学报, 2024, 75(1): 322-337.
[2]	温唯谷, 袁志宏, 王凯, 骆广生. 微分散液滴的光纤检测研究[J]. 化工学报, 2024, 75(1): 211-220.
[3]	钟东霖, 介素云, 杜淼, 潘鹏举, 单国荣. 聚苯基甲基硅氧烷分子量-折射率模型研究[J]. 化工学报, 2024, 75(1): 190-196.
[4]	尹刚, 钱中友, 曹文琦, 全鹏程, 许亨权, 颜非亚, 王民, 向禹, 向冬梅, 卢剑, 左玉海, 何文, 卢润廷. 基于Adaboost-PSO-SVM的铝电解槽健康状态诊断方法研究[J]. 化工学报, 2024, 75(1): 354-365.
[5]	宋嘉豪, 王文. 斯特林发动机与高温热管耦合运行特性研究[J]. 化工学报, 2023, 74(S1): 287-294.
[6]	连梦雅, 谈莹莹, 王林, 陈枫, 曹艺飞. 地下水预热新风一体化热泵空调系统制热性能研究[J]. 化工学报, 2023, 74(S1): 311-319.
[7]	金正浩, 封立杰, 李舒宏. 氨水溶液交叉型再吸收式热泵的能量及分析[J]. 化工学报, 2023, 74(S1): 53-63.
[8]	李科, 文键, 忻碧平. 耦合蒸气冷却屏的真空多层绝热结构对液氢储罐自增压过程的影响机制研究[J]. 化工学报, 2023, 74(9): 3786-3796.
[9]	王浩, 王振雷. 基于自适应谱方法的裂解炉烧焦模型化简策略[J]. 化工学报, 2023, 74(9): 3855-3864.
[10]	曹跃, 余冲, 李智, 杨明磊. 工业数据驱动的加氢裂化装置多工况切换过渡状态检测[J]. 化工学报, 2023, 74(9): 3841-3854.
[11]	尹刚, 李伊惠, 何飞, 曹文琦, 王民, 颜非亚, 向禹, 卢剑, 罗斌, 卢润廷. 基于KPCA和SVM的铝电解槽漏槽事故预警方法[J]. 化工学报, 2023, 74(8): 3419-3428.
[12]	李锦潼, 邱顺, 孙文寿. 煤浆法烟气脱硫中草酸和紫外线强化煤砷浸出过程[J]. 化工学报, 2023, 74(8): 3522-3532.
[13]	于旭东, 李琪, 陈念粗, 杜理, 任思颖, 曾英. 三元体系KCl + CaCl₂ + H₂O 298.2、323.2及348.2 K相平衡研究及计算[J]. 化工学报, 2023, 74(8): 3256-3265.
[14]	诸程瑛, 王振雷. 基于改进深度强化学习的乙烯裂解炉操作优化[J]. 化工学报, 2023, 74(8): 3429-3437.
[15]	闫琳琦, 王振雷. 基于STA-BiLSTM-LightGBM组合模型的多步预测软测量建模[J]. 化工学报, 2023, 74(8): 3407-3418.