Machine-learning modeling of water, oil, and solids content in oil-based drilling muds: An alternative approach to retort test; Colloids and Surfaces A: Physicochemical and Engineering Aspects; Vol. 728, pt. 2
| Parent link: | Colloids and Surfaces A: Physicochemical and Engineering Aspects.— .— Amsterdam: Elsevier Science Publishing Company Inc. Vol. 728, pt. 2.— 2026.— Article number 138500, 21 p. |
|---|---|
| Autor principal: | |
| Altres autors: | , |
| Sumari: | Title screen Monitoring the content of water, oil, and solids in oil-based muds (OBMs) is crucial in maintaining a smooth and efficient well-drilling process. Nevertheless, the only measurement method available, the retort test, is time-consuming, preventing the drilling-mud crew from frequently measuring these three OBM parameters. To address this issue, the present study leverages a vast field dataset to build robust, novel machine-learning models that precisely predict the content of water, oil, and solids in OBMs using five frequently measured drilling fluid parameters. In this regard, following the removal of outliers and the selection of the most influential variables, four predictive models, namely, multi-layer extreme learning machine, extreme gradient boosting (XGB), and their hybrid forms with particle swarm optimization (PSO), were developed and precisely evaluated using multiple performance and uncertainty measurement analyses. Among the developed models, the XGB-PSO consistently outperformed others across the training, validation, and blind testing phases, achieving the lowest average absolute relative errors in predicting the target parameters. A comprehensive performance assessment revealed that the XGB-PSO model exhibited minimal systematic bias, strong resistance to noise, the lowest risk of overfitting, as indicated by stable learning curves, and high reliability, confirmed by the narrowest bootstrapped confidence intervals. Finally, Shapley additive explanations analysis performed on the best-performing predictive models revealed mud weight as the most influential feature in predicting the target parameters. In contrast, Marsh funnel viscosity and mud type showed relatively minor influences. During drilling operations, this intelligent approach can assist the drilling-mud crew in making frequent and credible determinations of the water, oil, and solids content in OBMs Текстовый файл AM_Agreement |
| Idioma: | anglès |
| Publicat: |
2026
|
| Matèries: | |
| Accés en línia: | https://doi.org/10.1016/j.colsurfa.2025.138500 |
| Format: | Electrònic Capítol de llibre |
| KOHA link: | https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=687164 |
MARC
| LEADER | 00000naa0a2200000 4500 | ||
|---|---|---|---|
| 001 | 687164 | ||
| 005 | 20260629151137.0 | ||
| 090 | |a 687164 | ||
| 100 | |a 20260629d2026 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 Machine-learning modeling of water, oil, and solids content in oil-based drilling muds: An alternative approach to retort test |f Shadfar Davoodi, Evgeny Burnaev, Amir H. Mohammadi | |
| 203 | |a Текст |b визуальный |c электронный | ||
| 283 | |a online_resource |2 RDAcarrier | ||
| 300 | |a Title screen | ||
| 320 | |a References: 49 tit | ||
| 330 | |a Monitoring the content of water, oil, and solids in oil-based muds (OBMs) is crucial in maintaining a smooth and efficient well-drilling process. Nevertheless, the only measurement method available, the retort test, is time-consuming, preventing the drilling-mud crew from frequently measuring these three OBM parameters. To address this issue, the present study leverages a vast field dataset to build robust, novel machine-learning models that precisely predict the content of water, oil, and solids in OBMs using five frequently measured drilling fluid parameters. In this regard, following the removal of outliers and the selection of the most influential variables, four predictive models, namely, multi-layer extreme learning machine, extreme gradient boosting (XGB), and their hybrid forms with particle swarm optimization (PSO), were developed and precisely evaluated using multiple performance and uncertainty measurement analyses. Among the developed models, the XGB-PSO consistently outperformed others across the training, validation, and blind testing phases, achieving the lowest average absolute relative errors in predicting the target parameters. A comprehensive performance assessment revealed that the XGB-PSO model exhibited minimal systematic bias, strong resistance to noise, the lowest risk of overfitting, as indicated by stable learning curves, and high reliability, confirmed by the narrowest bootstrapped confidence intervals. Finally, Shapley additive explanations analysis performed on the best-performing predictive models revealed mud weight as the most influential feature in predicting the target parameters. In contrast, Marsh funnel viscosity and mud type showed relatively minor influences. During drilling operations, this intelligent approach can assist the drilling-mud crew in making frequent and credible determinations of the water, oil, and solids content in OBMs | ||
| 336 | |a Текстовый файл | ||
| 371 | 0 | |a AM_Agreement | |
| 461 | 1 | |t Colloids and Surfaces A: Physicochemical and Engineering Aspects |c Amsterdam |n Elsevier Science Publishing Company Inc. | |
| 463 | 1 | |t Vol. 728, pt. 2 |v Article number 138500, 21 p. |d 2026 | |
| 610 | 1 | |a электронный ресурс | |
| 610 | 1 | |a труды учёных ТПУ | |
| 610 | 1 | |a Retort test | |
| 610 | 1 | |a Oil-based mud composition | |
| 610 | 1 | |a Hybrid machine learning | |
| 610 | 1 | |a Particle swarm optimization | |
| 610 | 1 | |a Extreme gradient boostinge | |
| 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 Burnaev |b E. |g Evgeny | |
| 701 | 1 | |a Mohammadi |b A. H. |g Amir | |
| 801 | 0 | |a RU |b 63413507 |c 20260629 | |
| 850 | |a 63413507 | ||
| 856 | 4 | 0 | |u https://doi.org/10.1016/j.colsurfa.2025.138500 |z https://doi.org/10.1016/j.colsurfa.2025.138500 |
| 942 | |c CF | ||