Robust machine-learning model for prediction of carbon dioxide adsorption on metal-organic frameworks; Journal of Alloys and Compounds; Vol. 1010
| Parent link: | Journal of Alloys and Compounds.— .— Amsterdam: Elsevier Science Publishing Company Inc. Vol. 1010.— 2025.— Article number 177890, 18 p. |
|---|---|
| Andre forfattere: | , , , |
| Summary: | Title screen Carbon dioxide (CO2) capture is promising with economic and environmental benefits as an active climate-change mitigation approach. Nevertheless, the low adsorption capacity of traditional materials such as zeolites remains challenging. Metal-organic framework (MOF) has been proposed as a remedy, but its laboratory screening is costly and time-consuming. Hybrid machine learning and optimization models offer precise and rapid predictions of CO2 uptake, effectively addressing the issue of inadequate hyperparameter tuning often seen in standalone machine learning models. In this study, hybrid machine-learning models are employed to accurately predict the CO2 adsorption capacity of MOFs required for swift screening of MOF candidates, expediting the discovery of novel materials and fostering a more comprehensive understanding of CO2 adsorption mechanisms. Six models, including multi-layer perceptron neural network (MLPNN) and least square support vector machine (LSSVM) algorithms, along with their hybrid counterparts utilizing particle swarm optimization (PSO) and growth optimization (GO) algorithms, are evaluated based on experimental data from previous studies. The experimental data consists of 475 data points and six input variables: Brunauer-Emmett-Teller specific surface area (SBET), pressure (P), temperature (T), Langmuir-specific surface area (SL), metal center (MC), and total pore volume (VT). Results demonstrate that the LSSVM-GO model outperforms the others, with an overall root mean squared error of 0.7484 and an R2 of 0.9798. The LSSVM-GO model’s applicability domain is verified using a Williams plot, and a feature importance analysis conducted revealed that SBET has the greatest effect and VT has the least effect on its MOF CO2-uptake predictions. The novelty of this study lies in developing a machine-learning-based approach that provides an accurate and cost-effective prediction of adsorption capacity for different MOF design experiments. This approach has the potential to significantly reduce screening costs and time required for various MOF designs, making it a more viable carbon capture strategy Текстовый файл AM_Agreement |
| Sprog: | engelsk |
| Udgivet: |
2025
|
| Fag: | |
| Online adgang: | https://doi.org/10.1016/j.jallcom.2024.177890 |
| Format: | MixedMaterials Electronisk Book Chapter |
| KOHA link: | https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=678424 |
MARC
| LEADER | 00000naa0a2200000 4500 | ||
|---|---|---|---|
| 001 | 678424 | ||
| 005 | 20250204100333.0 | ||
| 090 | |a 678424 | ||
| 100 | |a 20250204d2025 k||y0rusy50 ba | ||
| 101 | 0 | |a eng | |
| 102 | |a NL | ||
| 135 | |a drcn ---uucaa | ||
| 181 | 0 | |a i |b e | |
| 182 | 0 | |a b | |
| 183 | 0 | |a cr |2 RDAcarrier | |
| 200 | 1 | |a Robust machine-learning model for prediction of carbon dioxide adsorption on metal-organic frameworks |f Promise O. Longe, Shadfar Davoodi, Mohammad Mehrad, David A. Wood | |
| 203 | |a Текст |b визуальный |c электронный | ||
| 283 | |a online_resource |2 RDAcarrier | ||
| 300 | |a Title screen | ||
| 320 | |a References: 71 tit | ||
| 330 | |a Carbon dioxide (CO2) capture is promising with economic and environmental benefits as an active climate-change mitigation approach. Nevertheless, the low adsorption capacity of traditional materials such as zeolites remains challenging. Metal-organic framework (MOF) has been proposed as a remedy, but its laboratory screening is costly and time-consuming. Hybrid machine learning and optimization models offer precise and rapid predictions of CO2 uptake, effectively addressing the issue of inadequate hyperparameter tuning often seen in standalone machine learning models. In this study, hybrid machine-learning models are employed to accurately predict the CO2 adsorption capacity of MOFs required for swift screening of MOF candidates, expediting the discovery of novel materials and fostering a more comprehensive understanding of CO2 adsorption mechanisms. Six models, including multi-layer perceptron neural network (MLPNN) and least square support vector machine (LSSVM) algorithms, along with their hybrid counterparts utilizing particle swarm optimization (PSO) and growth optimization (GO) algorithms, are evaluated based on experimental data from previous studies. The experimental data consists of 475 data points and six input variables: Brunauer-Emmett-Teller specific surface area (SBET), pressure (P), temperature (T), Langmuir-specific surface area (SL), metal center (MC), and total pore volume (VT). Results demonstrate that the LSSVM-GO model outperforms the others, with an overall root mean squared error of 0.7484 and an R2 of 0.9798. The LSSVM-GO model’s applicability domain is verified using a Williams plot, and a feature importance analysis conducted revealed that SBET has the greatest effect and VT has the least effect on its MOF CO2-uptake predictions. The novelty of this study lies in developing a machine-learning-based approach that provides an accurate and cost-effective prediction of adsorption capacity for different MOF design experiments. This approach has the potential to significantly reduce screening costs and time required for various MOF designs, making it a more viable carbon capture strategy | ||
| 336 | |a Текстовый файл | ||
| 371 | 0 | |a AM_Agreement | |
| 461 | 1 | |t Journal of Alloys and Compounds |c Amsterdam |n Elsevier Science Publishing Company Inc. | |
| 463 | 1 | |t Vol. 1010 |v Article number 177890, 18 p. |d 2025 | |
| 610 | 1 | |a электронный ресурс | |
| 610 | 1 | |a труды учёных ТПУ | |
| 610 | 1 | |a metal-organic frameworks | |
| 610 | 1 | |a CO2-adsorption prediction | |
| 610 | 1 | |a carbon capture and storage | |
| 610 | 1 | |a hybrid machine-learning-optimization model | |
| 610 | 1 | |a least-square support vector machine | |
| 701 | 1 | |a Longe |b P. O. |g Promise | |
| 701 | 1 | |a Davoodi |b Sh. |c specialist in the field of petroleum engineering |c Research Engineer of Tomsk Polytechnic University |f 1990- |g Shadfar |9 22200 | |
| 701 | 1 | |a Mehrad |b M. |g Mohammad | |
| 701 | 1 | |a Wood |b D. A. |g David | |
| 801 | 0 | |a RU |b 63413507 |c 20250204 | |
| 850 | |a 63413507 | ||
| 856 | 4 | |u https://doi.org/10.1016/j.jallcom.2024.177890 |z https://doi.org/10.1016/j.jallcom.2024.177890 | |
| 942 | |c CF | ||