Pub Date : 2026-02-12DOI: 10.1016/j.uncres.2026.100349
Hao Wu , Yan Zhang , Xinrui Lyu , Zhelin Wang , Feng Qiu
Understanding the influence of water adsorption on gas transport properties is critical for optimizing coalbed methane (CBM) extraction. This study investigates the influence of adsorbed water on methane diffusion and apparent permeability in three coals representing sub-bituminous, bituminous, and anthracite ranks. The samples were selected as geologically representative end-members from major coalbed methane basins in China. Adsorbed water was introduced via humidity equilibration at 97% relative humidity, simulating residual moisture conditions typical of partially dewatered reservoirs. Pressure decay experiments show that under dry conditions, apparent permeability exhibits a non-monotonic variation with coal rank, with the medium-rank coal displaying the highest value among the three tested samples. Upon exposure to adsorbed water, apparent permeability decreases by 39% to 70% across all ranks, with greater suppression observed in low-rank and high-rank coals compared to the medium-rank sample. This rank-dependent response is attributed to differences in pore structure and water distribution associated with coalification history. While the findings are based on single samples per rank and reflect a simplified moisture condition, they provide mechanistic insight into how coal maturity modulates the sensitivity of gas transport to adsorbed water, offering implications for permeability modeling during the dewatering phase of coalbed methane recovery.
{"title":"Influence of adsorbed water on diffusion coefficient and permeability of coal","authors":"Hao Wu , Yan Zhang , Xinrui Lyu , Zhelin Wang , Feng Qiu","doi":"10.1016/j.uncres.2026.100349","DOIUrl":"10.1016/j.uncres.2026.100349","url":null,"abstract":"<div><div>Understanding the influence of water adsorption on gas transport properties is critical for optimizing coalbed methane (CBM) extraction. This study investigates the influence of adsorbed water on methane diffusion and apparent permeability in three coals representing sub-bituminous, bituminous, and anthracite ranks. The samples were selected as geologically representative end-members from major coalbed methane basins in China. Adsorbed water was introduced via humidity equilibration at 97% relative humidity, simulating residual moisture conditions typical of partially dewatered reservoirs. Pressure decay experiments show that under dry conditions, apparent permeability exhibits a non-monotonic variation with coal rank, with the medium-rank coal displaying the highest value among the three tested samples. Upon exposure to adsorbed water, apparent permeability decreases by 39% to 70% across all ranks, with greater suppression observed in low-rank and high-rank coals compared to the medium-rank sample. This rank-dependent response is attributed to differences in pore structure and water distribution associated with coalification history. While the findings are based on single samples per rank and reflect a simplified moisture condition, they provide mechanistic insight into how coal maturity modulates the sensitivity of gas transport to adsorbed water, offering implications for permeability modeling during the dewatering phase of coalbed methane recovery.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"11 ","pages":"Article 100349"},"PeriodicalIF":4.6,"publicationDate":"2026-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-12DOI: 10.1016/j.uncres.2026.100348
Rabih Murr , Tarek Ibrahim , Nicolas Youssef , Samer Ali , Wassim Salameh , Jalal Faraj , Mahmoud Khaled
A significant portion of building energy consumption is attributed to space heating, while a significant amount of power generator waste heat is left unused. At the same time, solar energy's sporadic nature restricts its use as a stand-alone heating source. The objective of this research is to design and evaluate a hybrid heat pump system that combines solar air heating with power generator exhaust gas recovery in order to minimize greenhouse gas emissions, improve energy efficiency, and consume less fuel in cold-climate operations. In contrast to traditional exhaust-recovery or solar-assisted systems, the suggested setup integrates waste-heat and dual renewable sources in various configurations to find the best thermodynamic performance. Through experimental validation and computational modelling, the unique hybrid design for sustainable heating is established by integrating solar and exhaust heat recovery with a heat pump cycle. Six system configurations were examined covering all possible integration sequences of the heat pump, exhaust gas recovery, and solar air heating components. Field data from a 45 kVA generator and 12 m-long solar air tubes were integrated with thermodynamic modelling in a combined simulation experimental framework. At −5 °C ambient conditions, the configuration combining the heat pump, solar air heater, and exhaust gas recovery produced the highest coefficient of performance, with a boost of up to 4770%. The system achieved significant monthly energy and cost savings, as well as up to 98% reductions in carbon dioxide emissions. While the integration of these distinct dynamic heat sources present control challenges, the modular nature of the system affirms feasible application across different scales of buildings. For next-generation building heating systems, this hybrid heat recovery technology provides an extremely effective and environmentally friendly option. This study illustrates the capabilities of hybrid renewable–waste heat recovery systems for the sustainable heating of buildings, paving the road for future experimental work and thermal energy storage integration.
{"title":"Hybrid solar–exhaust heat recovery heat pump system: A combined experimental and numerical study for high-efficiency sustainable heating","authors":"Rabih Murr , Tarek Ibrahim , Nicolas Youssef , Samer Ali , Wassim Salameh , Jalal Faraj , Mahmoud Khaled","doi":"10.1016/j.uncres.2026.100348","DOIUrl":"10.1016/j.uncres.2026.100348","url":null,"abstract":"<div><div>A significant portion of building energy consumption is attributed to space heating, while a significant amount of power generator waste heat is left unused. At the same time, solar energy's sporadic nature restricts its use as a stand-alone heating source. The objective of this research is to design and evaluate a hybrid heat pump system that combines solar air heating with power generator exhaust gas recovery in order to minimize greenhouse gas emissions, improve energy efficiency, and consume less fuel in cold-climate operations. In contrast to traditional exhaust-recovery or solar-assisted systems, the suggested setup integrates waste-heat and dual renewable sources in various configurations to find the best thermodynamic performance. Through experimental validation and computational modelling, the unique hybrid design for sustainable heating is established by integrating solar and exhaust heat recovery with a heat pump cycle. Six system configurations were examined covering all possible integration sequences of the heat pump, exhaust gas recovery, and solar air heating components. Field data from a 45 kVA generator and 12 m-long solar air tubes were integrated with thermodynamic modelling in a combined simulation experimental framework. At −5 °C ambient conditions, the configuration combining the heat pump, solar air heater, and exhaust gas recovery produced the highest coefficient of performance, with a boost of up to 4770%. The system achieved significant monthly energy and cost savings, as well as up to 98% reductions in carbon dioxide emissions. While the integration of these distinct dynamic heat sources present control challenges, the modular nature of the system affirms feasible application across different scales of buildings. For next-generation building heating systems, this hybrid heat recovery technology provides an extremely effective and environmentally friendly option. This study illustrates the capabilities of hybrid renewable–waste heat recovery systems for the sustainable heating of buildings, paving the road for future experimental work and thermal energy storage integration.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"11 ","pages":"Article 100348"},"PeriodicalIF":4.6,"publicationDate":"2026-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146161746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-09DOI: 10.1016/j.uncres.2026.100334
Nandito Davy , Ammar El-Husseiny , Manzar Fawad , Umair bin Waheed , Korhan Ayranci , Nicholas B. Harris
Accurate prospectivity assessment for unconventional reservoir requires defining sweet spots, a methodology that delineates prospective areas based on quality factors such as organic richness, fracability, and maturity-related reservoir characteristics. In shale gas systems, critical parameters are typically defined by total organic carbon (TOC), brittleness index (BI), and gas saturation . This study compares sweet spot delineation using conventional rock-physics templates (RPTs) with machine learning (ML) algorithms, specifically Artificial Neural Networks (ANNs), Extreme Gradient Boosting (XGB), and Logistic Regression (LR). The RPT-based approach utilizes cross plots between different elastic parameters and employs a statistically supported objective function with two thresholding techniques—box-shaped and mathematical-function thresholds—where the former demonstrates superior performance. The ML approach leverages elastic parameters as features and applies a randomized search for hyperparameter optimization. Results show that the optimized ML-based approach is superior to the RPT-based one, achieving an -score (obtained from precision and recall metrics) of 0.801 against 0.746. The analysis reveals that , , and consistently rank as the most impactful elastic parameters based on validation performance across multiple feature combinations and ML algorithms for delineating prospective zones. Though less powerful, the RPT-based approach offers simplicity and may be optimized further or combined with the ML technique. Our findings underline the practicality and reliability of the proposed ML-based methodologies for unconventional reservoir assessment to accurately delineate sweet spots and improve reservoir evaluation practices.
{"title":"Comparative study of optimized rock-physics templates (RPTs) and machine learning (ML) approaches for sweet spot delineation in shale gas reservoir","authors":"Nandito Davy , Ammar El-Husseiny , Manzar Fawad , Umair bin Waheed , Korhan Ayranci , Nicholas B. Harris","doi":"10.1016/j.uncres.2026.100334","DOIUrl":"10.1016/j.uncres.2026.100334","url":null,"abstract":"<div><div>Accurate prospectivity assessment for unconventional reservoir requires defining sweet spots, a methodology that delineates prospective areas based on quality factors such as organic richness, fracability, and maturity-related reservoir characteristics. In shale gas systems, critical parameters are typically defined by total organic carbon (TOC), brittleness index (BI), and gas saturation <span><math><mrow><msub><mi>S</mi><mi>g</mi></msub></mrow></math></span>. This study compares sweet spot delineation using conventional rock-physics templates (RPTs) with machine learning (ML) algorithms, specifically Artificial Neural Networks (ANNs), Extreme Gradient Boosting (XGB), and Logistic Regression (LR). The RPT-based approach utilizes cross plots between different elastic parameters and employs a statistically supported objective function with two thresholding techniques—box-shaped and mathematical-function thresholds—where the former demonstrates superior performance. The ML approach leverages elastic parameters as features and applies a randomized search for hyperparameter optimization. Results show that the optimized ML-based approach is superior to the RPT-based one, achieving an <span><math><mrow><mi>F</mi><mn>1</mn></mrow></math></span>-score (obtained from precision and recall metrics) of 0.801 against 0.746. The analysis reveals that <span><math><mrow><msub><mi>I</mi><mi>p</mi></msub></mrow></math></span>, <span><math><mrow><msub><mi>I</mi><mi>s</mi></msub></mrow></math></span>, and <span><math><mrow><msub><mi>V</mi><mi>p</mi></msub><mo>/</mo><msub><mi>V</mi><mi>s</mi></msub></mrow></math></span> consistently rank as the most impactful elastic parameters based on validation performance across multiple feature combinations and ML algorithms for delineating prospective zones. Though less powerful, the RPT-based approach offers simplicity and may be optimized further or combined with the ML technique. Our findings underline the practicality and reliability of the proposed ML-based methodologies for unconventional reservoir assessment to accurately delineate sweet spots and improve reservoir evaluation practices.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"11 ","pages":"Article 100334"},"PeriodicalIF":4.6,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-09DOI: 10.1016/j.uncres.2026.100335
Yeonpyeong Jo , Rasoul B. Sorkhabi , Palash Panja , Milind Deo
Despite the tremendous growth of oil production from tight reservoirs in the US, characterizing and forecasting these tight formations faces significant uncertainties and risks. Machine learning (ML) techniques have emerged as insightful complementary methods to physics-based unconventional reservoir simulations for predicting oil production due to complex reservoir properties that make traditional modeling challenging. This study applies ML approaches to predict oil production from the Bakken formation in the Williston basin and employs normalized production indices (NPIs) by normalizing cumulative oil production with time and completion parameters. Key parameters include surface coordinates, lateral perforated length, total base water volume and total proppant used in hydraulic fracturing, number of stimulated fracture stages, and water volume and proppant per stimulated fracture stage. Five scenarios were analyzed: raw cumulative oil and four NPIs. Feature importance analysis was conducted using SHAP, followed by prediction using random forest, XGBoost, and Multilayer perceptron. Results revealed that well location ranked among the top features, demonstrating superior production potential at the eastern Bakken formation. Total base water and water volume per fracture stage showed a contrasting trend, indicating that the water distribution strategy is critical for stage-level performance. Proppant variables displayed complex non-monotonic relationships with multiple optimal ranges. XGBoost outperformed other algorithms. Stage-normalized NPI2 achieved optimal prediction accuracy with 37% improvement in R2 compared to raw cumulative oil, while spacing-normalized NPI3 offered the most practical implementation. These findings demonstrate that completion optimization strategies should align with specific production objectives, providing quantitative guidance for improved capital efficiency in unconventional oil development.
{"title":"Machine learning for forecasting production in tight oil reservoirs: Application to the Bakken formation of Williston basin","authors":"Yeonpyeong Jo , Rasoul B. Sorkhabi , Palash Panja , Milind Deo","doi":"10.1016/j.uncres.2026.100335","DOIUrl":"10.1016/j.uncres.2026.100335","url":null,"abstract":"<div><div>Despite the tremendous growth of oil production from tight reservoirs in the US, characterizing and forecasting these tight formations faces significant uncertainties and risks. Machine learning (ML) techniques have emerged as insightful complementary methods to physics-based unconventional reservoir simulations for predicting oil production due to complex reservoir properties that make traditional modeling challenging. This study applies ML approaches to predict oil production from the Bakken formation in the Williston basin and employs normalized production indices (NPIs) by normalizing cumulative oil production with time and completion parameters. Key parameters include surface coordinates, lateral perforated length, total base water volume and total proppant used in hydraulic fracturing, number of stimulated fracture stages, and water volume and proppant per stimulated fracture stage. Five scenarios were analyzed: raw cumulative oil and four NPIs. Feature importance analysis was conducted using SHAP, followed by prediction using random forest, XGBoost, and Multilayer perceptron. Results revealed that well location ranked among the top features, demonstrating superior production potential at the eastern Bakken formation. Total base water and water volume per fracture stage showed a contrasting trend, indicating that the water distribution strategy is critical for stage-level performance. Proppant variables displayed complex non-monotonic relationships with multiple optimal ranges. XGBoost outperformed other algorithms. Stage-normalized NPI<sub>2</sub> achieved optimal prediction accuracy with 37% improvement in R<sup>2</sup> compared to raw cumulative oil, while spacing-normalized NPI<sub>3</sub> offered the most practical implementation. These findings demonstrate that completion optimization strategies should align with specific production objectives, providing quantitative guidance for improved capital efficiency in unconventional oil development.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"11 ","pages":"Article 100335"},"PeriodicalIF":4.6,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146161747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-06DOI: 10.1016/j.uncres.2026.100333
Ali Mahmoud, Rahul Gajbhiye
Lost circulation is one of the most persistent and costly challenges in drilling operations, particularly under high-pressure and high-temperature conditions and in fractured carbonate reservoirs. Despite decades of research, no universal solution exists, and severe fluid losses continue to jeopardize well construction, increase non-productive time, and compromise safety. This review delivers a comprehensive synthesis of mechanisms, materials, experimental evaluations, and field practices, spanning petroleum, geothermal, and emerging energy-transition wells. Mechanistic pathways of loss initiation are critically examined across porous, fractured, and cavernous formations, as well as severe lost circulation scenarios, highlighting the limitations of existing predictive models. Lost circulation materials, ranging from conventional particulates and fibers to advanced nano-enabled and biodegradable systems, are assessed in terms of bridging efficiency, survivability under high-pressure and high-temperature conditions, and sustainability. Experimental and modeling approaches, including fracture-slot tests, dynamic high-pressure and high-temperature flow loops, and computational tools such as computational fluid dynamics, discrete element modeling, and artificial intelligence and machine learning, are evaluated to expose the gap between laboratory results and field reliability. Field strategies, including wellbore strengthening, cement squeezes, and managed pressure drilling, are reviewed to underline their largely reactive nature. Finally, a forward-looking roadmap is presented, identifying research needs such as standardized high-pressure and high-temperature validation protocols, chemically compatible and durable materials for carbon dioxide and hydrogen wells, and the integration of digital twins with artificial intelligence-driven predictive diagnostics.
{"title":"Lost circulation in drilling: Mechanisms, materials, and future directions for HPHT and energy-transition wells","authors":"Ali Mahmoud, Rahul Gajbhiye","doi":"10.1016/j.uncres.2026.100333","DOIUrl":"10.1016/j.uncres.2026.100333","url":null,"abstract":"<div><div>Lost circulation is one of the most persistent and costly challenges in drilling operations, particularly under high-pressure and high-temperature conditions and in fractured carbonate reservoirs. Despite decades of research, no universal solution exists, and severe fluid losses continue to jeopardize well construction, increase non-productive time, and compromise safety. This review delivers a comprehensive synthesis of mechanisms, materials, experimental evaluations, and field practices, spanning petroleum, geothermal, and emerging energy-transition wells. Mechanistic pathways of loss initiation are critically examined across porous, fractured, and cavernous formations, as well as severe lost circulation scenarios, highlighting the limitations of existing predictive models. Lost circulation materials, ranging from conventional particulates and fibers to advanced nano-enabled and biodegradable systems, are assessed in terms of bridging efficiency, survivability under high-pressure and high-temperature conditions, and sustainability. Experimental and modeling approaches, including fracture-slot tests, dynamic high-pressure and high-temperature flow loops, and computational tools such as computational fluid dynamics, discrete element modeling, and artificial intelligence and machine learning, are evaluated to expose the gap between laboratory results and field reliability. Field strategies, including wellbore strengthening, cement squeezes, and managed pressure drilling, are reviewed to underline their largely reactive nature. Finally, a forward-looking roadmap is presented, identifying research needs such as standardized high-pressure and high-temperature validation protocols, chemically compatible and durable materials for carbon dioxide and hydrogen wells, and the integration of digital twins with artificial intelligence-driven predictive diagnostics.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"10 ","pages":"Article 100333"},"PeriodicalIF":4.6,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146189671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Interfacial tension (IFT) between displacing fluids and reservoir hydrocarbons is vital in enhanced oil recovery (EOR) as it affects fluid displacement efficiency and the mobilization of trapped oil. Lower IFT increases the capillary number and enhances fluid mobility, improving oil displacement in porous media. In this study, advanced machine learning (ML) techniques, including adaptive boosting decision tree (AdaBoost-DT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and random forest (RF) were utilized to model the IFT of n-alkanes and aqueous systems containing surfactants and nanoparticles (NPs), using a collection of 708 experimental data points. The results demonstrated that the LightGBM model outperformed the others, achieving average absolute relative errors (AARE) of 2.02%, 3.27%, and 2.27% for the training, testing, and total datasets, respectively, along with the highest overall determination coefficient (R2) value of 0.9967. Moreover, sensitivity and trend analyses highlighted that the phase inversion temperature (PIT) of surfactants and the NPs concentration significantly affect IFT, showing the strongest negative effects. The input variables were ranked by impact, with PIT, NPs concentration, surfactant concentration, hydrophilic-lipophilic balance (HLB), molecular weight (Mw) of n-alkanes, average NPs diameter, and temperature. The Mw of n-alkanes and the average NPs diameter positively influenced IFT, while the other factors negatively affected it. Finally, the leverage technique applied to the LightGBM model indicated that over 95% of the data fell within the acceptable validation zone, verifying the model's statistical robustness and the reliability of the experimental data collected. The models developed in this study are data-driven and demonstrate reliable performance within the reported data ranges. To ensure their broader applicability, these models should be validated using entirely unseen datasets. Future research efforts could focus on expanding the dataset, exploring alternative input variables, and examining the effects of various surfactants and NPs on the IFT behavior of hydrocarbons and aqueous mixtures.
{"title":"Modeling interfacial tension between n-alkanes and aqueous systems containing surfactants and nanoparticles","authors":"Behnam Amiri-Ramsheh , Seyyed-Mohammad-Mehdi Hosseini , Amir-Ehsan Avazzadeh , Mohammad-Reza Mohammadi , Saeid Atashrouz , Dragutin Nedeljkovic , Mehdi Ostadhassan , Abdolhossein Hemmati-Sarapardeh , Ahmad Mohaddespour","doi":"10.1016/j.uncres.2026.100332","DOIUrl":"10.1016/j.uncres.2026.100332","url":null,"abstract":"<div><div>Interfacial tension (IFT) between displacing fluids and reservoir hydrocarbons is vital in enhanced oil recovery (EOR) as it affects fluid displacement efficiency and the mobilization of trapped oil. Lower IFT increases the capillary number and enhances fluid mobility, improving oil displacement in porous media. In this study, advanced machine learning (ML) techniques, including adaptive boosting decision tree (AdaBoost-DT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and random forest (RF) were utilized to model the IFT of n-alkanes and aqueous systems containing surfactants and nanoparticles (NPs), using a collection of 708 experimental data points. The results demonstrated that the LightGBM model outperformed the others, achieving average absolute relative errors (AARE) of 2.02%, 3.27%, and 2.27% for the training, testing, and total datasets, respectively, along with the highest overall determination coefficient (R<sup>2</sup>) value of 0.9967. Moreover, sensitivity and trend analyses highlighted that the phase inversion temperature (PIT) of surfactants and the NPs concentration significantly affect IFT, showing the strongest negative effects. The input variables were ranked by impact, with PIT, NPs concentration, surfactant concentration, hydrophilic-lipophilic balance (HLB), molecular weight (Mw) of n-alkanes, average NPs diameter, and temperature. The Mw of n-alkanes and the average NPs diameter positively influenced IFT, while the other factors negatively affected it. Finally, the leverage technique applied to the LightGBM model indicated that over 95% of the data fell within the acceptable validation zone, verifying the model's statistical robustness and the reliability of the experimental data collected. The models developed in this study are data-driven and demonstrate reliable performance within the reported data ranges. To ensure their broader applicability, these models should be validated using entirely unseen datasets. Future research efforts could focus on expanding the dataset, exploring alternative input variables, and examining the effects of various surfactants and NPs on the IFT behavior of hydrocarbons and aqueous mixtures.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"10 ","pages":"Article 100332"},"PeriodicalIF":4.6,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146189670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-05DOI: 10.1016/j.uncres.2026.100331
Deivid Campos , Bruno da Silva Macêdo , Oscar Ikechukwu Ogali , Matteo Bodini , Dmitriy A. Martyushev , Farouk Abduh Kamil Al-Fahaidy , Camila Martins Saporetti , Leonardo Goliatt
Accurately predicting Flowing Bottom-Hole Pressure (FBHP) is critical for optimizing oil and gas production. Existing predictive methods often rely on oversimplified or complex, yet computationally expensive, models that fail to capture the intrinsic nonlinearities of well dynamics, leading to inaccurate predictions and potential economic losses. This paper introduces a three-layer heterogeneous stacking ensemble model to address the latter challenge. In particular, the key novelty of the developed work is a hierarchical architecture that integrates five distinct Machine Learning (ML) base learners, two meta-learners, and a final super-learner, i.e., an additional meta-model that combines the outputs of the meta-learners to capture complex, non-linear relationships in the data. When evaluated on a field dataset (total dataset samples ; test set samples ), the proposed Super Learner Stacking model (ST-S) demonstrated superior predictive performance on the independent test set, achieving R-squared () = and Root Mean Squared Error (RMSE) = . In addition, the ST-S model outperformed all individual models and simpler stacking ensembles reported in the article. As a result, the developed ST-S model provides a robust, data-driven tool for FBHP prediction, achieving high predictive accuracy without resorting to computationally expensive methods, thereby supporting improved well management and production optimization.
{"title":"Heterogeneous stacking strategy for modeling flowing bottom-hole pressure of oil wells","authors":"Deivid Campos , Bruno da Silva Macêdo , Oscar Ikechukwu Ogali , Matteo Bodini , Dmitriy A. Martyushev , Farouk Abduh Kamil Al-Fahaidy , Camila Martins Saporetti , Leonardo Goliatt","doi":"10.1016/j.uncres.2026.100331","DOIUrl":"10.1016/j.uncres.2026.100331","url":null,"abstract":"<div><div>Accurately predicting Flowing Bottom-Hole Pressure (FBHP) is critical for optimizing oil and gas production. Existing predictive methods often rely on oversimplified or complex, yet computationally expensive, models that fail to capture the intrinsic nonlinearities of well dynamics, leading to inaccurate predictions and potential economic losses. This paper introduces a three-layer heterogeneous stacking ensemble model to address the latter challenge. In particular, the key novelty of the developed work is a hierarchical architecture that integrates five distinct Machine Learning (ML) base learners, two meta-learners, and a final super-learner, <em>i.e.</em>, an additional meta-model that combines the outputs of the meta-learners to capture complex, non-linear relationships in the data. When evaluated on a field dataset (total dataset samples <span><math><mrow><mi>N</mi><mo>=</mo><mn>795</mn></mrow></math></span>; test set samples <span><math><mrow><mi>N</mi><mo>=</mo><mn>199</mn></mrow></math></span>), the proposed Super Learner Stacking model (ST-S) demonstrated superior predictive performance on the independent test set, achieving R-squared (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>) = <span><math><mrow><mn>0</mn><mo>.</mo><mn>857</mn><mo>±</mo><mn>0</mn><mo>.</mo><mn>006</mn></mrow></math></span> and Root Mean Squared Error (RMSE) = <span><math><mrow><mn>146</mn><mo>.</mo><mn>382</mn><mo>±</mo><mn>2</mn><mo>.</mo><mn>806</mn></mrow></math></span>. In addition, the ST-S model outperformed all individual models and simpler stacking ensembles reported in the article. As a result, the developed ST-S model provides a robust, data-driven tool for FBHP prediction, achieving high predictive accuracy without resorting to computationally expensive methods, thereby supporting improved well management and production optimization.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"10 ","pages":"Article 100331"},"PeriodicalIF":4.6,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146189669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-31DOI: 10.1016/j.uncres.2026.100330
Idriss Dagal
Africa remains the world's least electrified continent, with approximately 600 million people lacking access to electricity and persistent urban-rural disparities. Despite recent advances, progress toward Sustainable Development Goal 7 remains uneven, with fewer than one in six African countries currently on track to achieve universal access by 2030. This study provides a comprehensive assessment of Africa's electrification trajectory by synthesizing evidence on the structural, institutional, and socio-spatial determinants shaping energy access outcomes. Using an integrated mixed-methods approach, the analysis draws on a Preferred Reporting Items for Systematic Reviews and Meta-Analyses-based systematic review of 120 studies (2015–2025), complemented by geospatial and policy analysis, to evaluate patterns of access, reliability, and inclusion across the continent. The study introduces an Energy Justice Index to capture multidimensional inequities and develops an investment prioritization framework to support more equitable electrification strategies. Results indicate that decentralized renewable energy systems can reduce electrification costs by up to 40% in low-density areas, yet service quality remains a critical challenge, with a substantial share of connected households receiving limited daily supply. Overall, the findings suggest that achieving universal access will require a shift from grid-centric expansion toward coordinated deployment of decentralized solutions, strengthened regulatory environments, and inclusion-focused financing mechanisms. These insights offer policy-relevant guidance for accelerating equitable and sustainable progress toward SDG 7 in Africa.
{"title":"Decoding Africa's energy divide: A systematic review of SDG 7 progress, structural determinants, and pathways to inclusive electrification","authors":"Idriss Dagal","doi":"10.1016/j.uncres.2026.100330","DOIUrl":"10.1016/j.uncres.2026.100330","url":null,"abstract":"<div><div>Africa remains the world's least electrified continent, with approximately 600 million people lacking access to electricity and persistent urban-rural disparities. Despite recent advances, progress toward Sustainable Development Goal 7 remains uneven, with fewer than one in six African countries currently on track to achieve universal access by 2030. This study provides a comprehensive assessment of Africa's electrification trajectory by synthesizing evidence on the structural, institutional, and socio-spatial determinants shaping energy access outcomes. Using an integrated mixed-methods approach, the analysis draws on a Preferred Reporting Items for Systematic Reviews and Meta-Analyses-based systematic review of 120 studies (2015–2025), complemented by geospatial and policy analysis, to evaluate patterns of access, reliability, and inclusion across the continent. The study introduces an Energy Justice Index to capture multidimensional inequities and develops an investment prioritization framework to support more equitable electrification strategies. Results indicate that decentralized renewable energy systems can reduce electrification costs by up to 40% in low-density areas, yet service quality remains a critical challenge, with a substantial share of connected households receiving limited daily supply. Overall, the findings suggest that achieving universal access will require a shift from grid-centric expansion toward coordinated deployment of decentralized solutions, strengthened regulatory environments, and inclusion-focused financing mechanisms. These insights offer policy-relevant guidance for accelerating equitable and sustainable progress toward SDG 7 in Africa.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"10 ","pages":"Article 100330"},"PeriodicalIF":4.6,"publicationDate":"2026-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146189672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-30DOI: 10.1016/j.uncres.2026.100329
Anupam Krishnan, Abdulkareem Sh. Mahdi Al-Obaidi, Lee Ching Hao
Small wind turbines operate under low Reynolds number conditions, where improving aerodynamic efficiency becomes crucial due to the adverse inflow characteristics. This study systematically investigates the isolated effect of a broad range of inflow turbulence intensities on the aerodynamic performance of a NACA2414 airfoil at a Reynolds number of 105 using two-dimensional finite volume Unsteady Reynolds-Averaged Navier-Stokes simulations. The computational domain was discretized as a structured grid and a Realizable k-ɛ model was utilized as the closure model in ANSYS Fluent. Three turbulence intensities (1 %, 5 %, and 10 %) were examined over angles of attack spanning from 0° to 30°. Significant reduction in lift-to-drag ratio was observed pre-stall with a contrasting enhancement post-stall with increasing turbulence intensities. In addition, onset of stall was delayed. Statistical analysis, employing ANOVA, based on 800 design points from 0.1 % to 20 % confirmed turbulence intensity as a significant parameter governing lift-to-drag ratio explaining 42.5 % of the associated variance. The effect was substantial at lower angles of attack and diminished at higher angles as post-stall conditions dominated. The present work demonstrates a non-linear and regime-dependent influence of turbulence intensity over NACA2414 airfoil performance at a transitional flow regime, directly relevant to small-scale wind turbine operation. Rotor-level analysis using a validated blade element momentum model further indicates a reduction in mechanical power output with increasing turbulence intensity. The findings establish turbulence intensity as a critical design parameter for low-Reynolds number wind turbine airfoils.
{"title":"Reestablishing turbulence intensity as a critical parameter for NACA2414 airfoil performance at low Reynolds number: A computational study","authors":"Anupam Krishnan, Abdulkareem Sh. Mahdi Al-Obaidi, Lee Ching Hao","doi":"10.1016/j.uncres.2026.100329","DOIUrl":"10.1016/j.uncres.2026.100329","url":null,"abstract":"<div><div>Small wind turbines operate under low Reynolds number conditions, where improving aerodynamic efficiency becomes crucial due to the adverse inflow characteristics. This study systematically investigates the isolated effect of a broad range of inflow turbulence intensities on the aerodynamic performance of a NACA2414 airfoil at a Reynolds number of 10<sup>5</sup> using two-dimensional finite volume Unsteady Reynolds-Averaged Navier-Stokes simulations. The computational domain was discretized as a structured grid and a Realizable <em>k-ɛ</em> model was utilized as the closure model in ANSYS Fluent. Three turbulence intensities (1 %, 5 %, and 10 %) were examined over angles of attack spanning from 0° to 30°. Significant reduction in lift-to-drag ratio was observed pre-stall with a contrasting enhancement post-stall with increasing turbulence intensities. In addition, onset of stall was delayed. Statistical analysis, employing ANOVA, based on 800 design points from 0.1 % to 20 % confirmed turbulence intensity as a significant parameter governing lift-to-drag ratio explaining 42.5 % of the associated variance. The effect was substantial at lower angles of attack and diminished at higher angles as post-stall conditions dominated. The present work demonstrates a non-linear and regime-dependent influence of turbulence intensity over NACA2414 airfoil performance at a transitional flow regime, directly relevant to small-scale wind turbine operation. Rotor-level analysis using a validated blade element momentum model further indicates a reduction in mechanical power output with increasing turbulence intensity. The findings establish turbulence intensity as a critical design parameter for low-Reynolds number wind turbine airfoils.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"10 ","pages":"Article 100329"},"PeriodicalIF":4.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146189667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-30DOI: 10.1016/j.uncres.2026.100328
Yingyan Li , Chenlin Hu , Jie Zeng , Wenfeng Wang , Shiqian Xu , Yingfang Zhou , Fanhua Zeng , Jingpeng Wang
Graded proppant injection into complex fractures is frequently used to prop connected secondary fractures in tight unconventional reservoirs. A comprehensive conductivity model incorporating creep, decreasing proppant size distribution, proppant embedment and deformation, and unpropped fracture surface deformation is established to ascertain partially propped fracture network conductivity. The propped fracture width variation is described by creep deformation, proppant embedment, and proppant particle deformation. The corresponding fracture permeability is depicted by the Carman-Kozeny equation where the dynamic proppant pack porosity is calculated via proppant size and the number of proppant layers. For unpropped areas, the width is controlled by effective stress, and the permeability is a function of fracture aperture. The hydraulic–electric analogies concept is use to integrate the local conductivity of different areas and characterize the overall fracture network conductivity. The model is verified against long-term conductivity measurement data. Results show that the fracture width variation is mainly caused by rock creep and proppant embedment. Larger Kelvin shear modulus and Maxwell viscosity slow down the conductivity decline rate. The conductivity becomes stable after 4 days when the Kelvin shear modulus is increased to 5.4 × 108 Pa. The Maxwell shear modulus has the slightest influence on conductivity. Larger-size proppants offer higher overall conductivity and better maintain the conductivity. The fracture network conductivity is significantly larger than the conductivity of the main fracture fully supported by the graded proppants and that of the fracture branches. The three-dimensional (3D) conductivity diagram and two-dimensional (2D) conductivity maps are generated to better demonstrate time-dependent conductivity evolution.
{"title":"A comprehensive fracture network conductivity model for tight unconventional reservoirs considering various proppant size, creep deformation, and proppant compaction and embedment","authors":"Yingyan Li , Chenlin Hu , Jie Zeng , Wenfeng Wang , Shiqian Xu , Yingfang Zhou , Fanhua Zeng , Jingpeng Wang","doi":"10.1016/j.uncres.2026.100328","DOIUrl":"10.1016/j.uncres.2026.100328","url":null,"abstract":"<div><div>Graded proppant injection into complex fractures is frequently used to prop connected secondary fractures in tight unconventional reservoirs. A comprehensive conductivity model incorporating creep, decreasing proppant size distribution, proppant embedment and deformation, and unpropped fracture surface deformation is established to ascertain partially propped fracture network conductivity. The propped fracture width variation is described by creep deformation, proppant embedment, and proppant particle deformation. The corresponding fracture permeability is depicted by the Carman-Kozeny equation where the dynamic proppant pack porosity is calculated via proppant size and the number of proppant layers. For unpropped areas, the width is controlled by effective stress, and the permeability is a function of fracture aperture. The hydraulic–electric analogies concept is use to integrate the local conductivity of different areas and characterize the overall fracture network conductivity. The model is verified against long-term conductivity measurement data. Results show that the fracture width variation is mainly caused by rock creep and proppant embedment. Larger Kelvin shear modulus and Maxwell viscosity slow down the conductivity decline rate. The conductivity becomes stable after 4 days when the Kelvin shear modulus is increased to 5.4 × 10<sup>8</sup> Pa. The Maxwell shear modulus has the slightest influence on conductivity. Larger-size proppants offer higher overall conductivity and better maintain the conductivity. The fracture network conductivity is significantly larger than the conductivity of the main fracture fully supported by the graded proppants and that of the fracture branches. The three-dimensional (3D) conductivity diagram and two-dimensional (2D) conductivity maps are generated to better demonstrate time-dependent conductivity evolution.</div></div>","PeriodicalId":101263,"journal":{"name":"Unconventional Resources","volume":"10 ","pages":"Article 100328"},"PeriodicalIF":4.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146189668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号