Geofluids最新文献

Hydraulic fracturing is the principal technical method about the generation of sophisticated fracture system in shale gas development; there are still significant challenges in enunciating the origination and extension of hydraulic fracture and optimizing design methods about hydraulic fracturing. The manuscript builds a small-scale shale reservoir model that containing randomly complex natural fractures and bedding by means of the continuous–discontinuous element method (CDEM); it provides detailed investigation of the effects including the shale reservoir burial depth along with the fracturing fluid viscosity, and flow rate on the generation and growth of fracture systems. The simulated findings suggest that hydraulic fractures are more likely to be captured by bedding planes and tend to extend along the bedding planes when the burial depth is shallow. The phenomenon of fractures penetrating bedding planes increases obviously under high flow rate conditions, resulting in more branching and interlacing in the vertical direction. The guiding effect of bedding planes on the propagation direction of hydro fractures gradually weakens as the viscosity of the fracturing fluid increases. Increasing the flow velocity of fracturing fluid enlarges outer boundary of hydraulic fracture system, and hydraulic fracture system’s propagation length and aperture grow substantially within the same period of injecting time. The high-viscosity fracturing fluid will lead to relatively single propagation of hydraulic fracture branches that is in favor of forming a single main fracture. This research puts forward a novel perspective for understanding the expansion characteristics of elaborate hydraulic fracture systems and persistent refinement of hydraulic fracturing operations in shale reservoirs.

In deep foundation pit engineering, leakage of underground diaphragm walls poses a serious threat to construction safety and the surrounding environment. This study employs finite element numerical simulation to investigate the detection of such leakage using the cross-well electrical method. The influence of acquisition device, electrode spacing, and well spacing on detection performance is systematically analyzed. Additionally, the accuracy of apparent resistivity imaging and leakage localization is assessed under various leakage conditions, including leakage point resistivity, leakage size, leakage depth, and the number of leakage points. The results demonstrate that the AM-BN dipole–dipole acquisition device offers superior localization accuracy and deeper penetration capability. Notably, reducing the electrode spacing below 2 m significantly enhances detection precision and effectively suppresses spurious anomaly interference. The cross-well electrical method exhibits robust performance in identifying a broad range of leakage scenarios. Moreover, a reliable localization approach by identifying the minimum value location of the apparent resistivity curve is proposed.

Based on the engineering background of the Daliuta mining area, the distribution of fractures and movement of water within weakly cemented overlying strata were studied with a physical simulation of liquid–solid coupling and a COMSOL numerical simulation test. The results showed that after the initial caving of the roof, all kinds of fractures rapidly developed in the overburden in the longitudinal direction, forming longitudinal fractures on opposite sides of the working face with angles of 81° and 78°; the space between the separated strata became the main channel for water flow. Under the action of water flow and the movement of the rock strata, mining-induced fractures in the overlying rock displayed cyclic changes in the form of expansion, penetration, and closure. When the working face was fully mined, the penetrating fractures in the overlying strata and the mining-induced fractures in the working face became the main passages for water flow. The results of the numerical simulation showed that the seepage rate of overburden water increased with the advancement of the working face. When the working face advanced to 120 m, the attenuation of pressure and the increase in the seepage velocity were significantly slowed down. These experimental results provide a reference for the layout and maintenance of underground reservoirs and water-conserved mining in Western China.

This study develops a helical path tortuosity model with formation-specific parameterization to address the critical limitation of existing tortuosity models that typically excel within narrow permeability ranges. The model decomposes total tortuosity into three physically meaningful components: geometric tortuosity arising from flow path winding, constriction effects from pore throat narrowing, and percolation factors reflecting connectivity limitations. We implemented formation-specific critical porosity thresholds (ϕ₀) calibrated to unique connectivity characteristics of different sandstone classes. The model was validated across an extensive dataset spanning five orders of magnitude in permeability (0.00006–36 μm²), encompassing tight German sandstones, clean Fontainebleau quartz arenites, and unconsolidated high-connectivity formations. Formation-specific parameterization significantly improved predictive performance compared to the global model, with R² increasing from 0.77 (global) to 0.89–0.97 (formation-specific) and MAPE decreasing from 10.11% to 2.46%–4.40%. Quantitative component contribution analysis revealed that while geometric tortuosity dominates across all formation types (70%–77%), the relative importance of constriction and percolation effects varies systematically with formation characteristics. The model establishes formation-specific parameter scaling relationships that provide deeper insights into the fundamental physics governing fluid transport across different sandstone classes. This unified approach bridges critical gaps in our ability to predict tortuosity across the full spectrum of sandstone formations encountered in geological systems.

Lost circulation and gas invasion during cementing operations pose significant threats to safety and efficiency, making accurate prediction and timely early warning a critical concern in the industry. This paper proposes an intelligent early warning model based on ant colony optimization (ACO) and an enhanced version of XGBoost (SWXGBoost). The model extracts both raw features and slope-based variations of key parameters—flow rate, pressure, and density—to enhance representation, incorporates a sliding window and time decay mechanism to capture dynamic patterns, and leverages ACO to optimize hyperparameters for improved predictive performance. Experimental results based on 1800 field samples show that ACO-SWXGBoost achieves superior performance compared to mainstream baseline models, with a micro-F1 of 0.955, precision of 0.949, and recall of 0.961. On average, the model outperforms XGBoost, LightGBM, random forest, and decision tree by 5.55%, 7.28%, and 6.48% on the three respective metrics. Furthermore, SHAP analysis confirms a strong alignment between model predictions and field knowledge, highlighting the critical role of pressure, flow rate, and density in anomaly identification. The proposed approach is readily deployable within real-time monitoring systems, offering a reliable and interpretable solution for intelligent risk detection and early warning in cementing operations.

Deep coalbed methane (CBM) is a crucial resource for ensuring energy security. Despite some successful localized deep CBM developments, the unclear understanding of gas content and gas occurrence state remains a key obstacle to the comprehensive development of deep CBM. This study utilizes data from pressure-preserved coring and wireline coring gas content tests, isothermal adsorption tests, and well test temperature and pressure data to establish a methodological model. This model corrects the gas content obtained from wireline coring and determines the gas occurrence state. The gas content, including adsorbed and free gas content, and gas/water saturation were calculated, and the controlling factors were analyzed. The results reveal that the high values of total gas content and adsorbed gas content are concentrated in the southwestern part of the study area. The adsorption capacity of the coal, influenced by its degree of metamorphism, is identified as the primary factor affecting the total gas content and adsorbed gas content. Furthermore, the high values of free gas content are primarily concentrated at the northwestern edge of the study area. The main factors affecting the porosity difference of free gas are coal metamorphism type and inertinite content. Areas affected by magmatic thermal metamorphism and those with high inertinite content tend to have higher porosity. Additionally, pressure, rather than temperature, is identified as the main factor determining the density of free gas. These findings provide a relatively simple indirect method for obtaining deep CBM content and occurrence state, particularly for studying the free gas content in deep coal seams. This approach is aimed at offering theoretical support for the development of deep CBM in the middle Linxing block.

This research evaluates the performance of artificial neural networks (ANNs) and adaptive neuro-fuzzy inference systems (ANFIS) in selecting well candidates for hydraulic fracturing (HF) in the Bach Ho oilfield, Vietnam. Traditional well selection often depends on expert judgment and deterministic criteria, which may be limited in uncertain and data-constrained reservoir environments. To address this limitation, machine learning models are applied to improve decision-making accuracy. A dataset of 41 wells was analyzed using permeability, porosity, skin factor, reservoir pressure, water cut, and reservoir thickness to predict post-HF daily production rates. Both models were trained and evaluated using RMSE, MSE, MAE, and R². The ANFIS model demonstrated superior accuracy, achieving an RMSE of 4.24, R² of 0.93, and MAE of 4.24 on the training set. On the testing set, ANFIS achieved an RMSE of 40.44, R² of 0.81, and MAE of 30.33, outperforming the ANN model, which recorded an RMSE of 40.43, R² of 0.59, and MAE of 31.86. These results suggest that ANFIS is more effective in capturing nonlinear relationships and handling input uncertainties. The study presents a practical, interpretable tool for supporting petroleum engineers in prioritizing HF candidates, ultimately enhancing oil recovery and resource allocation in complex reservoir settings.

Due to the influence of high crustal stress, weak surrounding rock, and other factors, deep soft rock roadways are often in complex stress states. This study takes the III4104 roadway as the research object to reveal the deformation mechanisms of deep soft rock roadways, aiming to effectively address the support challenges faced by such roadways. Firstly, a field survey summarized the failure characteristics of deep soft rock roadway and analyzed the causes of failure. Secondly, a mechanical model of floor heave was established, and the maximum depth of floor damage was calculated. Subsequently, based on the UDEC Trigon method, a discrete element numerical model that meets engineering scale was constructed, and a correction method for model parameters was proposed. Through simulation research, the deformation characteristics of surrounding rock, displacement vector distribution, crack distribution, and plastic zone extent were explored. Finally, a combined control technology of “roof corner anchor cable + rib anchor cable + concrete inverted arch + floor anchor cable” was proposed, along with specific parameters, and successfully applied in the III4104 roadway. The application results showed that, after 50 days of monitoring, the surrounding rock deformation stabilized, with the overall deformation rate controlled within 5%, indicating good control performance. The research results provide technical assistance for the stable control of surrounding rock deformation in deep soft rock roadways.

To investigate the control factors of mine disasters induced by high ground pressure, this paper takes a longwall face of Pingdingshan No. 10 coal mine as the engineering background. By employing the analytical model and numerical simulation methods, this study quantified the movement characteristics of the hard roof strata and the evolution of mining-induced stress. The control mechanisms of various geological factors on the distribution of the abutment pressure were evaluated. The results indicate that (1) the PFC-FLAC coupled numerical model can effectively simulate and predict the characteristics of roof collapse and the distribution characteristics of abutment pressure. (2) The first weighting distance, the peak of the abutment pressure, and the influence range of the abutment pressure can be measured by the proposed model, whereas the periodic weighting barely can be observed. (3) The tensile strength and thickness of the hard roof directly control the peak value of the abutment pressure and the first weighting distance. As the strength and thickness of the hard roof increase, both the peak value and influence distance of the abutment pressure significantly increase. When the tensile strength of the hard roof reaches a high value, the development of fractures in the hard roof shows a clear periodicity pattern. The vertical stress significantly affects the movement and fracturing of the overburden, while the horizontal stress has a limited impact on the movement and fracturing of the overburden. The proposed methods provide technical guidance and theoretical support for controlling ground pressure in deep-buried longwall faces.

The study used an integrated method of water quality indices (WQIs) and multivariate statistical analysis to evaluate the hydrogeochemical characteristics of groundwater samples and their appropriateness for drinking and agricultural applications in Kombolcha City. The hydrogeochemical parameters from 17 water samples were examined using standard methods. The WQI, irrigation indices, and geographical information systems (GISs) were used to determine groundwater suitability for various purposes and the spatial distribution of major ions. The study also utilized multivariate statistical methods, including principal component analysis (PCA) and cluster analysis (CA), to evaluate complex groundwater quality datasets. The results revealed that all hydrogeochemical characteristics were in line with Ethiopian drinking water standards and WHO drinking water guidelines. The WQI indicates that 94% of the samples are in excellent condition, whereas about 6% are in acceptable drinking water conditions in the study area’s southern and central regions with considerable agricultural and industrial activities. The Ca²⁺-HCO₃⁻ facies comprise about 53% of the shallow aquifer, and the remaining samples are found in the Ca²⁺-Mg²⁺-Cl⁻ and Ca²⁺-Na⁺-HCO₃⁻–type facies. SAR, Na%, PI, and MAR indicators showed that most water samples are very good to moderately suitable for agricultural use. CA uses dendrogram plots to group groundwater characteristics and sample locations into three groups based on common groundwater features. The study concluded that the combination of WQI, multivariate statistical, and GIS is a viable approach for prioritizing groundwater mitigation and monitoring efforts in Ethiopia’s semiarid regions of the Awash River basin.