{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T22:18:06Z","timestamp":1780525086950,"version":"3.54.1"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,11,14]],"date-time":"2024-11-14T00:00:00Z","timestamp":1731542400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Computational analysis of countercurrent flows in packed absorption columns, often used in solvent-based post-combustion carbon capture systems (CCSs), is challenging. Typically, computational fluid dynamics (CFD) approaches are used to simulate the interactions between a solvent, gas, and column's packing geometry while accounting for the thermodynamics, kinetics, heat, and mass transfer effects of the absorption process. These simulations can then be used explain a column's hydrodynamic characteristics and evaluate its CO<jats:sub>2<\/jats:sub>-capture efficiency. However, these approaches are computationally expensive, making it difficult to evaluate numerous designs and operating conditions to improve efficiency at industrial scales. In this work, we comprehensively explore the application of statistical ML methods, convolutional neural networks (CNNs), and graph neural networks (GNNs) to aid and accelerate the scale-up and design optimization of solvent-based post-combustion CCSs. We apply these methods to CFD datasets of countercurrent flows in absorption columns with structured packings characterized by several geometric parameters. We train models to use these parameters, inlet velocity conditions, and other model-specific representations of the column to estimate key determinants of CO<jats:sub>2<\/jats:sub>-capture efficiency without having to simulate additional CFD datasets. We also evaluate the impact of different input types on the accuracy and generalizability of each model. We discuss the strengths and limitations of each approach to further elucidate the role of CNNs, GNNs, and other machine learning approaches for CO<jats:sub>2<\/jats:sub>-capture property prediction and design optimization.<\/jats:p>","DOI":"10.3389\/frai.2024.1441934","type":"journal-article","created":{"date-parts":[[2024,11,14]],"date-time":"2024-11-14T11:15:53Z","timestamp":1731582953000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Comparative study of machine learning techniques for post-combustion carbon capture systems"],"prefix":"10.3389","volume":"7","author":[{"given":"Yeping","family":"Hu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Lei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yash Girish","family":"Shah","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jose","family":"Cadena","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amar","family":"Saini","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Grigorios","family":"Panagakos","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Phan","family":"Nguyen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2024,11,14]]},"reference":[{"key":"B1","first-page":"724","year":"2011","journal-title":"Ansys Fluent Theory Guide"},{"key":"B2","doi-asserted-by":"publisher","first-page":"5388","DOI":"10.1021\/es5005347","article-title":"Drivers of the growth in global greenhouse gas emissions","volume":"48","author":"Arto","year":"2014","journal-title":"Environ. Sci. Technol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1002\/ceat.200500302","article-title":"Experimental and CFD simulation study for the wetting of a structured packing element with liquids","volume":"29","author":"Ataki","year":"2006","journal-title":"Chem. Eng. Technol"},{"key":"B4","article-title":"\u201cScientific computing algorithms to learn enhanced scalable surrogates for mesh physics,\u201d","author":"Bartoldson","year":"2023","journal-title":"ICLR 2023 Workshop on Physics for Machine Learning"},{"key":"B5","doi-asserted-by":"publisher","first-page":"525","DOI":"10.1007\/s00466-019-01740-0","article-title":"Prediction of aerodynamic flow fields using convolutional neural networks","volume":"64","author":"Bhatnagar","year":"2019","journal-title":"Comput. Mech"},{"key":"B6","doi-asserted-by":"publisher","first-page":"2047","DOI":"10.1080\/01496395.2019.1622566","article-title":"3D printed structures for optimized carbon capture technology in packed bed columns","volume":"54","author":"Bolton","year":"2019","journal-title":"Sep. Sci. Technol"},{"key":"B7","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1016\/0021-9991(92)90240-Y","article-title":"A continuum method for modeling surface tension","volume":"100","author":"Brackbill","year":"1992","journal-title":"J. Comput. Phys"},{"key":"B8","doi-asserted-by":"publisher","first-page":"750","DOI":"10.1145\/3292500.3330958","article-title":"\u201cGraph transformation policy network for chemical reaction prediction,\u201d","author":"Do","year":"2019","journal-title":"Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery"},{"key":"B9","article-title":"\u201cNeural message passing for quantum chemistry,\u201d","author":"Gilmer","year":"2017","journal-title":"Proceedings of the 34th International Conference on Machine Learning - Volume 70"},{"key":"B10","doi-asserted-by":"crossref","first-page":"4166","DOI":"10.1145\/3580305.3599835","article-title":"\u201cGraph learning in physical-informed mesh-reduced space for real-world dynamic systems,\u201d","volume-title":"Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","author":"Hu","year":"2023"},{"key":"B11","doi-asserted-by":"publisher","first-page":"16681","DOI":"10.1021\/acssuschemeng.2c05985","article-title":"Predicting CO2 absorption in ionic liquids with molecular descriptors and explainable graph neural networks","volume":"10","author":"Jian","year":"2022","journal-title":"ACS Sustain. Chem. Eng"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1190\/geo2022-0755.1","article-title":"Deep-learning-based 3D fault detection for carbon capture and storage","author":"Kaur","year":"2023","journal-title":"Geophysics"},{"key":"B13","doi-asserted-by":"publisher","first-page":"e2101784118","DOI":"10.1073\/pnas.2101784118","article-title":"Machine learning-accelerated computational fluid dynamics","volume":"118","author":"Kochkov","year":"2021","journal-title":"Proc. Nat. Acad. Sci"},{"key":"B14","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/j.supflu.2017.07.029","article-title":"The CO2 economy: review of CO2 capture and reuse technologies","volume":"132","author":"Koytsoumpa","year":"2018","journal-title":"J. Supercrit. Fluids"},{"key":"B15","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"B16","doi-asserted-by":"publisher","first-page":"6999","DOI":"10.1109\/TNNLS.2021.3084827","article-title":"A survey of convolutional neural networks: analysis, applications, and prospects","volume":"33","author":"Li","year":"2021","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B17","doi-asserted-by":"publisher","first-page":"2058","DOI":"10.1214\/aos\/1069362310","article-title":"On latin hypercube sampling","volume":"24","author":"Loh","year":"1996","journal-title":"Ann. Stat"},{"key":"B18","doi-asserted-by":"publisher","first-page":"847","DOI":"10.1016\/j.cej.2019.03.008","article-title":"Adsorption characteristics of supercritical CO2\/CH4 on different types of coal and a machine learning approach","volume":"368","author":"Meng","year":"2019","journal-title":"Chem. Eng. J"},{"key":"B19","doi-asserted-by":"publisher","first-page":"134","DOI":"10.1016\/j.ijggc.2013.01.023","article-title":"A simulation model of a CO2 absorption process with methyldiethanolamine solvent and piperazine as an activator","volume":"15","author":"Mudhasakul","year":"2013","journal-title":"Int. J. Greenh. Gas Control"},{"key":"B20","doi-asserted-by":"publisher","first-page":"2147","DOI":"10.1021\/acs.jcim.0c01318","article-title":"Predicting energetics materials' crystalline density from chemical structure by machine learning","volume":"61","author":"Nguyen","year":"2021","journal-title":"J. Chem. Inf. Model"},{"key":"B21","volume-title":"A computational investigation of the effect of packing structural features on the performance of carbon capture for solvent-based post-combustion applications","author":"Panagakos","year":"2023"},{"key":"B22","volume-title":"Computational Methods for Multiphase Flow","author":"Prosperetti","year":"2009"},{"key":"B23","doi-asserted-by":"publisher","first-page":"142","DOI":"10.1016\/j.ijggc.2013.02.011","article-title":"CFD simulation of CO2 capture from gas mixtures in nanoporous membranes by solution of 2-amino-2-methyl-1-propanol and piperazine","volume":"15","author":"Razavi","year":"2013","journal-title":"Int. J. Greenh. Gas Control"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1811.05660","article-title":"Mt-cgcnn: Integrating crystal graph convolutional neural network with multitask learning for material property prediction","author":"Sanyal","year":"2018","journal-title":"arXiv"},{"key":"B25","doi-asserted-by":"publisher","first-page":"119113","DOI":"10.1016\/j.energy.2020.119113","article-title":"A machine learning approach for modeling and optimization of a CO2 post-combustion capture unit","volume":"215","author":"Shalaby","year":"2021","journal-title":"Energy"},{"key":"B26","doi-asserted-by":"publisher","first-page":"138124","DOI":"10.1016\/j.cej.2022.138124","article-title":"Hydrodynamics of countercurrent flow in an additive-manufactured column with triply periodic minimal surfaces for carbon dioxide capture","volume":"450","author":"Singh","year":"2022","journal-title":"Chem. Eng. J"},{"key":"B27","doi-asserted-by":"publisher","first-page":"25","DOI":"10.2514\/1.J058291","article-title":"Deep learning methods for reynolds-averaged navier-stokes simulations of airfoil flows","volume":"58","author":"Thuerey","year":"2020","journal-title":"AIAA J"},{"key":"B28","doi-asserted-by":"publisher","first-page":"1173","DOI":"10.1002\/aic.12345","article-title":"A dimensionless model for predicting the mass-transfer area of structured packing","volume":"57","author":"Tsai","year":"2011","journal-title":"AIChE J"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1710.10903","article-title":"Graph attention networks","author":"Veli\u010dkovi\u0107","year":"2017","journal-title":"arXiv"},{"key":"B30","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1016\/j.jcou.2017.06.012","article-title":"Predicting CO2 capture of ionic liquids using machine learning","volume":"21","author":"Venkatraman","year":"2017","journal-title":"J. CO2 Util"},{"key":"B31","doi-asserted-by":"publisher","first-page":"650","DOI":"10.1016\/j.egypro.2017.03.1209","article-title":"A review of post-combustion CO2 capture technologies from coal-fired power plants","volume":"114","author":"Wang","year":"2017","journal-title":"Energy Procedia"},{"key":"B32","doi-asserted-by":"publisher","first-page":"145301","DOI":"10.1103\/PhysRevLett.120.145301","article-title":"Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties","volume":"120","author":"Xie","year":"2018","journal-title":"Phys. Rev. Lett"},{"key":"B33","doi-asserted-by":"publisher","author":"Xu","year":"2018","DOI":"10.48550\/arXiv.1810.00826"},{"key":"B34","doi-asserted-by":"publisher","first-page":"7029","DOI":"10.1039\/D3CP05490J","article-title":"Predicting the enthalpy of formation of energetic molecules via conventional machine learning and GNN","volume":"26","author":"Zhang","year":"2024","journal-title":"Phys. Chem. Chem. Phys"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40649-019-0069-y","article-title":"Graph convolutional networks: a comprehensive review","volume":"6","author":"Zhang","year":"2019","journal-title":"Comput. Soc. Netw"},{"key":"B36","doi-asserted-by":"publisher","first-page":"120399","DOI":"10.1016\/j.memsci.2022.120399","article-title":"Machine learning aided high-throughput prediction of ionic liquid@ mof composites for membrane-based CO2 capture","volume":"650","author":"Zhang","year":"2022","journal-title":"J. Memb. Sci"},{"key":"B37","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1016\/j.aiopen.2021.01.001","article-title":"Graph neural networks: a review of methods and applications","volume":"1","author":"Zhou","year":"2020","journal-title":"AI Open"},{"key":"B38","doi-asserted-by":"publisher","first-page":"102790","DOI":"10.1016\/j.ijggc.2019.102790","article-title":"A data-driven CO2 leakage detection using seismic data and spatial-temporal densely connected convolutional neural networks","volume":"90","author":"Zhou","year":"2019","journal-title":"Int. J. Greenh. Gas Control"},{"key":"B39","doi-asserted-by":"publisher","first-page":"119943","DOI":"10.1016\/j.ijheatmasstransfer.2020.119943","article-title":"Gas-liquid distribution and mass transfer of CO2 absorption into sodium glycinate aqueous solution in parallel multi-channel microreactor","volume":"157","author":"Zhu","year":"2020","journal-title":"Int. J. Heat Mass Transf"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2024.1441934\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,11,14]],"date-time":"2024-11-14T11:15:58Z","timestamp":1731582958000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2024.1441934\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,11,14]]},"references-count":39,"alternative-id":["10.3389\/frai.2024.1441934"],"URL":"https:\/\/doi.org\/10.3389\/frai.2024.1441934","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,11,14]]},"article-number":"1441934"}}