Hybrid intelligent models for water activity prediction in complex ionic liquid systems; Journal of Environmental Chemical Engineering; Vol. 13, iss. 6
| Parent link: | Journal of Environmental Chemical Engineering.— .— Amsterdam: Elsevier Science Publishing Company Inc. Vol. 13, iss. 6.— 2025.— Article number 120358, 24 p. |
|---|---|
| Altres autors: | , , , , , |
| Sumari: | Title screen Precise prediction of water activity (WA) in ionic liquids (IL) systems is impeded by their complex, nonlinear interactions, and the limitations of traditional thermodynamic models. Machine learning offers an effective alternative for capturing complex property patterns and improving WA predictions. This study develops predictive models for WA using the multilayer extreme learning machine (MELM) and least squares support vector machine algorithms, as well as hybrid versions optimized with particle swarm optimization (PSO) and genetic algorithm, based on a large laboratory dataset. Following data cleaning, the feature selection analysis identified seven key parameters: temperature, pressure, molality of second composition, molality of IL, critical temperature of IL, acentric factor of IL, and critical pressure of IL. A portion of the training data was allocated as validation data using k-fold cross-validation, with the number of folds carefully selected to ensure optimal model performance and prevent overfitting. The results demonstrate that hybrid models in the prediction of the target variable, particularly MELM-PSO, exhibit superior reproducibility, lower error values across the training, validation, and testing phases, and reduced uncertainty and overfitting compared to other models. The proposed WA-prediction models can potentially be applied to various industrial sectors, including energy storage, gas purification, and process control Текстовый файл AM_Agreement |
| Idioma: | anglès |
| Publicat: |
2025
|
| Matèries: | |
| Accés en línia: | https://doi.org/10.1016/j.matlet.2025.140037 |
| Format: | Electrònic Capítol de llibre |
| KOHA link: | https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=687136 |
MARC
| LEADER | 00000naa0a2200000 4500 | ||
|---|---|---|---|
| 001 | 687136 | ||
| 005 | 20260629094259.0 | ||
| 090 | |a 687136 | ||
| 100 | |a 20260629d2025 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 Hybrid intelligent models for water activity prediction in complex ionic liquid systems |f Shadfar Davoodi, Hung Vo Thanh, David A. Wood [et al.] | |
| 203 | |a Текст |b визуальный |c электронный | ||
| 283 | |a online_resource |2 RDAcarrier | ||
| 300 | |a Title screen | ||
| 320 | |a References: 83 tit | ||
| 330 | |a Precise prediction of water activity (WA) in ionic liquids (IL) systems is impeded by their complex, nonlinear interactions, and the limitations of traditional thermodynamic models. Machine learning offers an effective alternative for capturing complex property patterns and improving WA predictions. This study develops predictive models for WA using the multilayer extreme learning machine (MELM) and least squares support vector machine algorithms, as well as hybrid versions optimized with particle swarm optimization (PSO) and genetic algorithm, based on a large laboratory dataset. Following data cleaning, the feature selection analysis identified seven key parameters: temperature, pressure, molality of second composition, molality of IL, critical temperature of IL, acentric factor of IL, and critical pressure of IL. A portion of the training data was allocated as validation data using k-fold cross-validation, with the number of folds carefully selected to ensure optimal model performance and prevent overfitting. The results demonstrate that hybrid models in the prediction of the target variable, particularly MELM-PSO, exhibit superior reproducibility, lower error values across the training, validation, and testing phases, and reduced uncertainty and overfitting compared to other models. The proposed WA-prediction models can potentially be applied to various industrial sectors, including energy storage, gas purification, and process control | ||
| 336 | |a Текстовый файл | ||
| 371 | 0 | |a AM_Agreement | |
| 461 | 1 | |t Journal of Environmental Chemical Engineering |c Amsterdam |n Elsevier Science Publishing Company Inc. | |
| 463 | 1 | |t Vol. 13, iss. 6 |v Article number 120358, 24 p. |d 2025 | |
| 610 | 1 | |a электронный ресурс | |
| 610 | 1 | |a труды учёных ТПУ | |
| 610 | 1 | |a Water activity prediction | |
| 610 | 1 | |a Ionic liquids | |
| 610 | 1 | |a Multilayer extreme learning machine | |
| 610 | 1 | |a Hybrid models | |
| 610 | 1 | |a Particle swarm optimization | |
| 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 | 0 | |a Thanh Hung Vo | |
| 701 | 1 | |a Wood |b D. A. |g David | |
| 701 | 1 | |a Makarov |b N. S. |g Nikita Sergeevich |f 1999- |c specialist in the field of petroleum engineering |c Engineer of Tomsk Polytechnic University |9 88868 | |
| 701 | 1 | |a Mehrad |b M. |g Mohammad | |
| 701 | 1 | |a Burnaev |b E. |g Evgeny | |
| 801 | 0 | |a RU |b 63413507 |c 20260629 | |
| 850 | |a 63413507 | ||
| 856 | 4 | 0 | |u https://doi.org/10.1016/j.matlet.2025.140037 |z https://doi.org/10.1016/j.matlet.2025.140037 |
| 942 | |c CF | ||