Physics-informed and chemistry-aware framework to predict hydrogen-brine interfacial tension for underground hydrogen storage reservoir conditions; Geoenergy Science and Engineering; Vol. 266

Dades bibliogràfiques
Parent link:Geoenergy Science and Engineering.— .— Amsterdam: Elsevier Science Publishing Company Inc.
Vol. 266.— 2026.— Article number 214609, 23 p.
Autor principal: Davoodi Sh. Shadfar
Altres autors: Wood D. A. David
Sumari:Title screen
The distribution of hydrogen (H2) and cushion gas mixtures in underground hydrogen (gas) storage reservoirs (UHS) is partly controlled by capillary forces, which are determined by the gas-brine interfacial tension (IFT) in the pore space. Reliably and transparently predicting IFT from gas compositional and reservoir environmental conditions is useful for determining storage capabilities and gas-containment risks in UHS reservoirs. A 2649-data-record dataset is compiled for published laboratory tests of gas-brine mixtures at subsurface salinity, temperature, and pressure conditions relevant to UHS for predicting IFT. The gas compositions are divided into three classes: H2 only (Class I), binary gas combinations (H2+CO2, H2+CH4, or H2+N2; Class II), and the quaternary gas combination (H2+CO2+CH4+N2; Class III). A series of empirical IFT-prediction equations are developed based on the physics-guided, chemistry-aware-feature-influence (CAFI) method and the non-parametric, customized formulaic optimization (CFO) methods. Thirteen equations are tested on the data records of the three specific gas classes and compared with published IFT empirical equations. A further three equations develop general predictors applied to the entire dataset, predicting IFT for any of the gas compositions involved. These developed equations substantially outperform published IFT empirical relationships. The class-focused equations outperform the generic equations, although both provide credible IFT prediction performance. The CFO equations outperform the CAFI equations. The best performing equation for Class III (2240 records) achieves IFT prediction performance of predictions (R2 = 0.9602; RMSE = 2.27 mN/M). The best performing general gas-brine IFT prediction equation (2649 records) achieves reliable IFT prediction performance (R2 = 0.7878; RMSE = 5.25 mN/M). These results justify the use of a physics-guided and chemistry-aware IFT prediction approach using the CAFI and CFO methods
Текстовый файл
AM_Agreement
Idioma:anglès
Publicat: 2026
Matèries:
Accés en línia:https://doi.org/10.1016/j.geoen.2026.214609
Format: Electrònic Capítol de llibre
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=687177

MARC

LEADER 00000naa0a2200000 4500
001 687177
005 20260630115022.0
090 |a 687177 
100 |a 20260630d2026 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 Physics-informed and chemistry-aware framework to predict hydrogen-brine interfacial tension for underground hydrogen storage reservoir conditions  |f Shadfar Davoodi, David A. Wood  
203 |a Текст  |b визуальный  |c электронный 
283 |a online_resource  |2 RDAcarrier 
300 |a Title screen 
320 |a References: 51 tit 
330 |a The distribution of hydrogen (H2) and cushion gas mixtures in underground hydrogen (gas) storage reservoirs (UHS) is partly controlled by capillary forces, which are determined by the gas-brine interfacial tension (IFT) in the pore space. Reliably and transparently predicting IFT from gas compositional and reservoir environmental conditions is useful for determining storage capabilities and gas-containment risks in UHS reservoirs. A 2649-data-record dataset is compiled for published laboratory tests of gas-brine mixtures at subsurface salinity, temperature, and pressure conditions relevant to UHS for predicting IFT. The gas compositions are divided into three classes: H2 only (Class I), binary gas combinations (H2+CO2, H2+CH4, or H2+N2; Class II), and the quaternary gas combination (H2+CO2+CH4+N2; Class III). A series of empirical IFT-prediction equations are developed based on the physics-guided, chemistry-aware-feature-influence (CAFI) method and the non-parametric, customized formulaic optimization (CFO) methods. Thirteen equations are tested on the data records of the three specific gas classes and compared with published IFT empirical equations. A further three equations develop general predictors applied to the entire dataset, predicting IFT for any of the gas compositions involved. These developed equations substantially outperform published IFT empirical relationships. The class-focused equations outperform the generic equations, although both provide credible IFT prediction performance. The CFO equations outperform the CAFI equations. The best performing equation for Class III (2240 records) achieves IFT prediction performance of predictions (R2 = 0.9602; RMSE = 2.27 mN/M). The best performing general gas-brine IFT prediction equation (2649 records) achieves reliable IFT prediction performance (R2 = 0.7878; RMSE = 5.25 mN/M). These results justify the use of a physics-guided and chemistry-aware IFT prediction approach using the CAFI and CFO methods 
336 |a Текстовый файл 
371 0 |a AM_Agreement 
461 1 |t Geoenergy Science and Engineering  |c Amsterdam  |n Elsevier Science Publishing Company Inc. 
463 1 |t Vol. 266  |v Article number 214609, 23 p.  |d 2026 
610 1 |a электронный ресурс 
610 1 |a труды учёных ТПУ 
610 1 |a Multi-component gas mixtures 
610 1 |a Gas-brine interfacial tension 
610 1 |a Physics-guided empirical relationships 
610 1 |a Thermodynamic consistency 
610 1 |a Correlation-guided feature selection 
700 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 Wood  |b D. A.  |g David 
801 0 |a RU  |b 63413507  |c 20260630 
850 |a 63413507 
856 4 0 |u https://doi.org/10.1016/j.geoen.2026.214609  |z https://doi.org/10.1016/j.geoen.2026.214609 
942 |c CF