Machine learning and metaheuristics in microfluidic transport characterization and optimization: CFD and experimental study integrated with predictive modelling

IF 1.6 4区 工程技术 Q3 ENGINEERING, CHEMICAL
Afshin Kouhkord, Moheb Amirmahani, Faridoddin Hassani, Naser Naserifar
{"title":"Machine learning and metaheuristics in microfluidic transport characterization and optimization: CFD and experimental study integrated with predictive modelling","authors":"Afshin Kouhkord,&nbsp;Moheb Amirmahani,&nbsp;Faridoddin Hassani,&nbsp;Naser Naserifar","doi":"10.1002/cjce.25430","DOIUrl":null,"url":null,"abstract":"<p>This study presents a comprehensive numerical and experimental analysis on microfluidic cell lysis through computational fluid dynamics (CFD), data-driven modelling, and multi-objective optimization. The proposed intelligent framework integrates artificial intelligence and CFD for data generation and extraction, alongside machine learning analysis and experimental studies for transport phenomena characterization in the cell lysis process. The framework explores compound effects of various inflow Reynolds numbers and geometrical parameters, including obstacle configurations and microchannel thickness. It shows substantial effects on flow patterns and mixing in varied microfluidic designs. A surrogate model, developed via central composite design, exhibits high accuracy in assessing system functionality (<span></span><math>\n <mrow>\n <msup>\n <mi>R</mi>\n <mn>2</mn>\n </msup>\n <mo>&gt;</mo>\n <mn>0.92</mn>\n </mrow></math>). The height of the implemented baffles from its lower value to the upper bound resulted in more than 42% and 14% increase in the mixing index at low and high Reynolds numbers, respectively, with minimal impact on pressure drop. The framework introduces data-driven modelling coupled with multi-objective optimization by desirability function (DF), non-dominated sorting genetic algorithm (NSGA-II), and differential evolution (DE). In the optimization of microfluidic processes, machine learning algorithms outperform desirability-based methods, and the DE algorithm surpasses the NSGA-II. An optimum micromixing reducing the mixing length by over 50% and mixing index above 97% achieved, fabricated, and experimental investigations conducted to validate numerical process. Through the precise control of microfluidic variables and the exploitation of microtransfer phenomena, it is possible to enhance the efficiency and selectivity of cell lysis. This not only improves the accuracy of diagnostic information but also opens up new avenues for personalized medicine and therapeutic development.</p>","PeriodicalId":9400,"journal":{"name":"Canadian Journal of Chemical Engineering","volume":"103 3","pages":"1400-1418"},"PeriodicalIF":1.6000,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Canadian Journal of Chemical Engineering","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/cjce.25430","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
引用次数: 0

Abstract

This study presents a comprehensive numerical and experimental analysis on microfluidic cell lysis through computational fluid dynamics (CFD), data-driven modelling, and multi-objective optimization. The proposed intelligent framework integrates artificial intelligence and CFD for data generation and extraction, alongside machine learning analysis and experimental studies for transport phenomena characterization in the cell lysis process. The framework explores compound effects of various inflow Reynolds numbers and geometrical parameters, including obstacle configurations and microchannel thickness. It shows substantial effects on flow patterns and mixing in varied microfluidic designs. A surrogate model, developed via central composite design, exhibits high accuracy in assessing system functionality ( R 2 > 0.92 ). The height of the implemented baffles from its lower value to the upper bound resulted in more than 42% and 14% increase in the mixing index at low and high Reynolds numbers, respectively, with minimal impact on pressure drop. The framework introduces data-driven modelling coupled with multi-objective optimization by desirability function (DF), non-dominated sorting genetic algorithm (NSGA-II), and differential evolution (DE). In the optimization of microfluidic processes, machine learning algorithms outperform desirability-based methods, and the DE algorithm surpasses the NSGA-II. An optimum micromixing reducing the mixing length by over 50% and mixing index above 97% achieved, fabricated, and experimental investigations conducted to validate numerical process. Through the precise control of microfluidic variables and the exploitation of microtransfer phenomena, it is possible to enhance the efficiency and selectivity of cell lysis. This not only improves the accuracy of diagnostic information but also opens up new avenues for personalized medicine and therapeutic development.

微流体传输特征描述和优化中的机器学习和元搜索:结合预测建模的 CFD 和实验研究
本研究通过计算流体动力学(CFD)、数据驱动建模和多目标优化,对微流体细胞裂解进行了全面的数值和实验分析。所提出的智能框架将人工智能与用于数据生成和提取的 CFD 相结合,并通过机器学习分析和实验研究来表征细胞裂解过程中的传输现象。该框架探索了各种流入雷诺数和几何参数(包括障碍物配置和微通道厚度)的复合效应。它显示了不同微流体设计对流动模式和混合的实质性影响。通过中心复合设计开发的代用模型在评估系统功能方面具有很高的准确性()。实施的挡板高度从下限值增加到上限后,在低雷诺数和高雷诺数条件下,混合指数分别增加了 42% 和 14%,而对压降的影响却微乎其微。该框架引入了数据驱动建模,并通过可取函数(DF)、非支配排序遗传算法(NSGA-II)和微分进化(DE)进行多目标优化。在微流控过程的优化中,机器学习算法优于基于可取性的方法,而差分进化算法则超过了 NSGA-II。实现了最佳微混合,将混合长度减少了 50%以上,混合指数超过 97%,并制作和进行了实验研究,验证了数值过程。通过精确控制微流控变量和利用微转移现象,可以提高细胞裂解的效率和选择性。这不仅提高了诊断信息的准确性,还为个性化医疗和治疗开发开辟了新途径。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Canadian Journal of Chemical Engineering
Canadian Journal of Chemical Engineering 工程技术-工程:化工
CiteScore
3.60
自引率
14.30%
发文量
448
审稿时长
3.2 months
期刊介绍: The Canadian Journal of Chemical Engineering (CJChE) publishes original research articles, new theoretical interpretation or experimental findings and critical reviews in the science or industrial practice of chemical and biochemical processes. Preference is given to papers having a clearly indicated scope and applicability in any of the following areas: Fluid mechanics, heat and mass transfer, multiphase flows, separations processes, thermodynamics, process systems engineering, reactors and reaction kinetics, catalysis, interfacial phenomena, electrochemical phenomena, bioengineering, minerals processing and natural products and environmental and energy engineering. Papers that merely describe or present a conventional or routine analysis of existing processes will not be considered.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信