Mud leakage during workover can contaminate existing fractures and lead to significant deviations in refracturing treatment-pressure designs. This study aims to characterize mud leakage behaviour in fractured reservoirs and predict its impact on refracturing treatment pressure. A two-phase Darcy flow framework was established to simulate mud leakage and to obtain the spatial distribution of mud saturation. Based on the simulated mud-retention geometry in existing fractures, a treatment-pressure prediction model was developed by partitioning the fracture into mud-occupied and unleaked flow regions and incorporating permeability damage. The workflow was validated using field data from two mud-contaminated refracturing wells in the Tarim Basin, NW China (wells A and B). Field data indicate leaked mud volumes of 13.0 m(3) (Well A) and 9.6 m(3) (Well B), accompanied by substantial productivity degradation (the unrestricted flow rates decreased from 142.3 & times; 10(4) to 52.2 & times; 10(4) m(3)/day in Well A and from 38 & times; 10(4) to 1 & times; 10(4) m(3)/day in Well B) and residual permeability ratios of 0.366 and 0.026, respectively. A baseline pressure prediction that ignores mud impact underestimates the observed treatment-pressure window (110-120 MPa for Well A; 120-130 MPa for Well B) by 20.3-36.2 and 14.6-42.2 MPa, respectively. After incorporating mud retention and permeability damage, the predicted pressure ranges shift to 109.6-117.4 MPa (Well A) and 122.4-146.9 MPa (Well B), yielding a clear overlap with the measured pressure windows and substantially reducing the mismatch. This study provides an additive and practically applicable method for pressure-design correction and risk assessment in mud-contaminated refracturing operations.

Summary Novel knotted temporary plugging balls have been widely applied in various fields, owing to their unique geometric structure. The optimization of temporary plugging in multicluster horizontal wells is a critical challenge for enhancing hydraulic fracturing efficiency. For this study, we developed and experimentally validated a computational fluid dynamics-discrete element method (CFD-DEM) coupled model to simulate the migration and plugging behavior of novel knotted temporary plugging balls. An innovative discrete ball-chain method was introduced to characterize the flexibility of fibrillated tails, while the immersed boundary method (IBM) was applied to directly resolve fluid/solid interactions. The simulation results revealed a three-stage plugging mechanism comprising axial migration, adaptive alignment, and compliant deformation. In the second stage, the critical capture distance between the ball and perforation is defined as the key parameter governing successful plugging. The tail structure significantly extends this distance, which scales with the tail length and number. Plugging placement can be steered by adjusting the density contrast between the tails and the core. Furthermore, multicore knotted balls exhibited self-adaptive selective and secondary plugging under nonuniform erosion. Based on these findings, a variable-density multicore adaptive plugging strategy is proposed that integrates low-, medium-, and high-density designs to improve the perforation plugging probability in multicluster completions. The results provide guidance for material design and operational optimization.

Our country has abundant coal rock gas resources. Hydraulic fracturing is a key technology for effectively developing coal rock gas reservoirs. Due to the differences in the mechanical properties, microstructure, gas occurrence states, and productivity-controlling factors of deep coal rock compared with shallow and medium-depth coal reservoirs, as well as the significant property variation in reservoirs between blocks, the adaptability of current fracturing technologies still faces challenges. Innovation in reservoir stimulation technologies is essential for the efficient development of coal rock gas. The difficulties in reservoir stimulation brought by the geological characteristics of coal rock gas reservoirs are discussed first. To deal with the geological features and the challenges of fracturing technology, an efficient development concept of matrix pore-cleat/fracture simultaneous stimulation, summarized as “point desorption, line dredging, fracture geometry improvement, and propped bulk fracture network”, is proposed, and the following key issues are figured out based on its connotation: ① fracturing-induced matrix pore structure and pore surface property modification, enhanced gas desorption, imbibition displacement, and adsorbed-phase and free-phase methane collaborative and efficient gas supply; ② activate cleats/fractures, shorten gas diffusion distances, and facilitate matrix reserves releasing; ③ promote uniform fracture propagation, enhance fracture complexity, and precisely control fracture morphology; ④ full-scale proppant support for coal rock reservoirs that “precisely matches proppant particle size with multi-level fracture widths, supports bedding-plane fractures, and provides three-dimensional support for cleat and main fractures”. Results indicate that it is necessary to further study the relationship between fracture parameters and production dynamics based on the gas storage and production characteristics of deep coal rocks, in order to identify fracture parameters that can realize the adsorbed and free gas “continuous-cooperative” supply, and to provide support for fracture property control and treating design optimization. Conduct in-depth research on hydraulic fracture network propagation rules in coal rock reservoirs and fracture propagation numerical simulation technologies, and combine the net pressure log-log diagram during fracturing to effectively control the fracture network propagation behavior in coal rock gas reservoirs. Use high-viscosity and leakage-weakening fluid first to create the main fractures, then use low-viscosity fluid to create complex fractures, achieving “controlled near-wellbore fracture complexity and sufficiently extended fractures” to form a “long fracture network”. To deal with the requirement of high conductivity and larger volume for fracture networks in deep coal rock formations, maximizing fracture volume and optimizing flow capacity with limited proppant and fluid is an effective way to reduce costs and increase efficiency. Propose the full-scale proppant-support fracturing technology for deep coal rock reservoirs to achieve long-term connectivity of “main fractures + bedding planes + cleats” and increase the effective support volume of fractures. Optimize different proppants and fiber combinations through long-term fracture conductivity tests for multi-size fractures with different proppant placement patterns. Improve existing fracturing fluid systems, explore water-reducing, high-sand-ratio, low-cost fracturing fluids, and clean desorption-promoting agents.

Multi-cluster fracturing in vertically multi-layered formations faces critical challenges due to pronounced interlayer heterogeneity and uneven flow distribution among lateral fractures, resulting in unclear regulatory mechanisms of operational parameters on hydraulic fracture propagation. In this study, based on the equilibrium height growth model and the PKN-C model, the coupled solution of fracture length and height propagation is achieved through pressure iteration. By integrating the dynamic flow distribution mechanism of the resistance method, a simultaneous propagation model for multiple fractures in multi-layered formations is established. Furthermore, regulatory mechanisms of stimulation parameters on vertical-lateral competitive fracture propagation are systematically analyzed. The results demonstrate that interlayer mechanical contrasts induce non-smooth abrupt transitions in hydraulic fracture height profiles. The uniformity of multi-cluster fractures exhibits positive correlations with cluster number and spacing, while showing a negative correlation with injection rate. Additionally, cluster number and small cluster spacing (6-8 m) predominantly regulate vertical propagation, whereas injection rate and larger cluster spacing (>= 10 m) exert stronger control over lateral propagation. The proposed workflow provides theoretical guidance for parameter optimization in multi-layered formations stimulation.

Engineering parameters of enhanced geothermal systems (EGS) directly govern heat recovery performance and economic viability. To address the practical engineering challenge of selecting the most robust well configuration amidst conflicting technical and economic objectives, this study develops a comprehensive evaluation and decision-making framework tailored for sedimentary geothermal systems. First, a 3D discrete fracture network model is established to simulate the thermal extraction behavior of six distinct well patterns. These configurations are evaluated through a multi-dimensional index system comprising production temperature, power generation, total heat extraction, flow impedance, and cost of electricity (COE). To systematically identify the optimal well pattern, a balance must be struck among these conflicting criteria. Consequently, we introduce a novel multi-criteria decision-making (MCDM) method, GRA-COBRA, which employs a cooperative game model for weight determination and integrates Grey Relational Analysis (GRA) with Comprehensive Distance-Based Ranking (COBRA) for evaluation. Application of the GRA-COBRA methodology identifies the single-injection-triple-production (1I3P) pattern as the optimal configuration, achieving the best equilibrium between energy output and economic feasibility. Finally, comparative analysis and sensitivity tests confirm that the proposed framework provides project developers with a more rational and robust decision-making tool compared to traditional methods.

The creation of fracture networks via hydraulic fracturing is essential for effective reservoir stimulation, making the interaction between hydraulic and natural fractures a key research focus. As constant-rate injection is still widely adopted in field applications, the majority of existing research is based on this premise. In contrast, the propagation characteristics of hydraulic fractures under variable-rate injection have yet to be fully elucidated. Therefore, this study established a numerical model for hydraulic fracture propagation in naturally fractured reservoirs. The effects of constant-rate, cyclic-rate, and step-rate injection schemes on the interaction between hydraulic and natural fractures was systematically investigated. Furthermore, the influences of the period and amplitude associated with cyclic-rate injection, together with different step-rate injection schemes, on the propagation morphology of hydraulic fractures were analyzed. The results indicate that, compared with constant-rate injection, adopting cyclic-rate or step-rate injection can enhance the interaction patterns between hydraulic fractures and natural fractures. These transforms the originally simple crossing mode into a combined mode involving both crossing and opening. Furthermore, both the period and amplitude of the cyclic injection scheme significantly influence these interaction patterns. However, when the in-situ stress difference exceeds a critical value, the impact of cyclic-rate injection on fracture propagation diminishes, and the in-situ stress becomes the dominant controlling factor. The above findings are of significant importance for guiding field operations and improving fracturing effectiveness.

Effective propping of multi-stage fractures is challenging in shale oil and gas reservoirs stimulated by full-domain propped fracturing. This paper discusses the three-dimensional multi-level fracture propping technique, a component of the full-domain propped fracturing technology. The mechanism of proppant transport within multi-level fractures is thoroughly analyzed, and the control strategies for three-dimensional multi-level fracture propping and proper implementation paths are elaborated. Due to retardation by narrow fracture walls and severe fracturing fluid leakoff, proppant exhibits poor transport capacity and high settling velocity within shale fractures, resulting in limited longitudinal and lateral placement coverage. Additionally, proppant struggles to divert into branch fractures at fracture junctions, which ultimately reduces the effective propped volume of multi-level fractures. Adjusting pumping rate, fracturing fluid viscosity and proppant particle size can modify the intra-fracture proppant placement pattern to a certain extent, yet such measures show limited performance in improving far-fracture placement. To achieve three-dimensional multi-level fracture propping in shale, two targeted technologies are proposed. One is high-efficiency proppant placement in main fractures based on structure-driven proppant transport, which adopts proppant-fiber clusters as fundamental transport units instead of pure proppant grains to alter particle settling and packing behaviors, thereby greatly expanding the propped volume of main fractures. The other is graded propping by optimizing transport unit dimensions and improving particle entry capacity to mitigate insufficient propping in branch fractures. The full-domain propped fracturing technology has been fully or partially applied in pilot and comparative tests at nearly 300 wells in 11 oil and gas fields within China, with satisfactory results obtained. This technology is expected to be widely deployed for developing various unconventional oil and gas reservoirs in the future.

The deformation prediction of fractured rock slopes is a key challenge in hydropower engineering due to the heterogeneity and stochastic distribution of fracture systems. Traditional approaches often struggle to identify dominant fracture parameters and maintain predictive accuracy while preserving geological interpretability. This study proposes a hybrid feature selection and prediction framework that integrates Discrete Fracture Network (DFN) modeling, Finite Difference Method (FDM), Global Sensitivity Analysis (GSA), and Extreme Gradient Boosting (XGBoost). Eighteen fracture-related geometric and mechanical parameters were initially considered, with maximum slope displacement from DFN-FDM simulations as the output. Sensitivity ranking was performed using Pearson correlation, Sobol variance-based index, and PAWN density-based method (Probabilistic Analysis of Whisker Numbers). The results indicate that 11 parameters dominate slope deformation, reflecting both geometric configuration and mechanical heterogeneity. Based on ranked features, XGBoost surrogate models were established and evaluated through 10-fold cross-validation. The GSA-XGBoost achieves the lowest error, outperforming both the original model and PCA-based (Principal Component Analysis) surrogate model, while reducing dimensionality and retaining geological interpretability. Application to a fractured slope at BDa Hydropower Station demonstrates the framework’s capacity for parameter prioritization and reliable deformation prediction. This approach provides practical guidance for slope stability evaluation and support design in fractured rock masses.

Landslide-generated waves involve large deformations of the landslide and a mixture of solid material and water, posing significant challenges for numerical simulation. This study proposes a smoothed particle hydrodynamics and discrete element method (SPH-DEM) coupling framework based on mixture theory to study water-soil coupling in landslide-generated waves. The water is modeled by the SPH method as a weakly compressible Newtonian fluid, while the soil landslide is directly simulated at the particle scale with DEM. The governing equations for the fluid phase are derived from mixture theory using the intrinsic fluid density. The model explicitly captures the spatiotemporal variation of fluid volume fraction resulting from landslide deformation and mixing with water. The delta-SPH formulation is incorporated to enhance the accuracy and stability of fluid simulations, and a U-tube seepage test is conducted to validate the model's reliability for seepage computations. The proposed method is further applied to three experiments on granular landslide-generated waves. The results accurately capture the evolution of the free water surface, landslide deformation, and the mixing between granular and water, while also revealing the evolution of the inter-particle interaction chains within the landslide during motion. Compared with previous methods, the proposed approach demonstrates improved accuracy in wave prediction and better performance in modeling landslide dynamics.

Accurate and efficient simulation of fracture propagation patterns is crucial for optimizing hydraulic fracturing design. However, the impact of rock mechanics heterogeneity and interfaces on fracture propagation is highly complex. Traditional numerical simulation methods often suffer from slow convergence, high computational cost, and the need for manual adjustments of physical parameters under heterogeneous conditions, making fracture propagation simulation a challenging task. To address these issues, this study proposes a fracture propagation prediction method based on a generative neural network named Fracture Propagation GAN (FPGAN). By employing the FPGAN model as a surrogate for fracture propagation simulation, the efficiency of simulating fracture propagation under heterogeneous conditions is significantly enhanced while maintaining the accuracy of the original images. Fracture propagation time-series images under various mechanical parameters were generated using the Finite Discrete Element Method (FDEM) to construct both basic and complex datasets of fracture propagation images. The FPGAN model was trained on these datasets to enable rapid prediction of fracture morphologies under heterogeneous conditions. Experimental results demonstrate that the FPGAN model can predict hydraulic fracture propagation images for any given combination of mechanical parameters within 1 min, achieving computational efficiency improvements of several orders of magnitude compared to traditional numerical methods. The proposed FPGAN model provides a robust foundation for analyzing the influence of different mineral compositions on fracture generation and exhibits significant potential in hydraulic fracture propagation simulation.

To address the lack of effective models for evaluating the conductivity of hydraulically fractured networks with discontinuous proppant placement in shale reservoirs, this study uses tree-shaped fracture networks to characterize the hierarchical and branched morphology of complex hydraulic fractures, introduces the distributed dislocation technique to solve the nonlinear closure problem of fractures under discontinuous proppant placement, and establishes an equivalent fracture conductivity model for tree-shaped fracture networks with discontinuous proppant placement based on the hydraulic-electric analogy and a series-parallel circuit approach. The results show that an optimal placement spacing exists under discontinuous proppant placement, which can maximize fracture conductivity while reducing proppant consumption. For example, under representative conditions, when the proppant pillar width is 0.5 m, the optimal spacing between adjacent proppant pillars is about 0.20 m, and the corresponding maximum effective fracture conductivity is about 1.9 × 10-11 m3. Compared with the fully filled condition, the fracture conductivity can be improved by 1–2 orders of magnitude. The optimal placement spacing is mainly controlled by closure stress, rock Young’s modulus, proppant-pillar width, and initial fracture width. At the fracture-network scale, the effects of geometric parameters, such as the number of bifurcation levels, bifurcation angle, number of branches, and aspect ratio, on the overall fracture-network conductivity are further analyzed. The established model can provide theoretical guidance for fracture-network conductivity evaluation and proppant-injection strategy optimization.

During the shale gas extraction, a larger fracture closure stress causes proppants to embed in the shale surface, greatly reducing fracture conductivity. The heterogeneity of shale makes its mechanical properties exhibit obvious scale effects, and the embedmentbehavior of proppants of different sizes is thus affected by multi-scale mechanical responses. To investigate the effect of particle size on the embedment mechanism, this work conducted micro-indentation tests using indenters of varying radii to model the embedment process of a single proppant. The test results show that when the embedment depth is 1 μm, with the radius increasing from 25 μm to 400 μm, the corresponding load does not show a linear growth trend, while the Young's modulus increases by about 76%. When the embedment depth is 3 μm, 5 μm, and 10 μm, respectively, the load gradually increases with the increase in radius. It is worth noting that when the embedment depth reaches 10 μm, the Young's modulus corresponding to each particle size tends to be stable, about 25 GPa. Based on the Drucker-Prager yield criterion, an analysis model of proppant embedment depth considering the particle size effect was established. Compared with the traditional Hertz and Thornton models, the proposed model can more accurately reflect the plastic behavior and strengthening mechanism of shale, and is applicable for predicting and evaluating multi-scale embedment responses.

Fibers are key materials for regulating the transport and placement of proppants, but their mechanism of action remains unclear. In this study, a solid–solid–liquid three-phase coupled computational fluid dynamics and discrete element method approach was developed to incorporate adhesive flexible fibers, enabling the simulation of their regulatory mechanisms in proppant transport and placement within rough fractures. The results show that fibers intertwine into a network structure that drags and wraps adjacent particles, allowing proppants to escape isolated migration and forming a novel and efficient “fiber–proppant cluster cooperative transport” mode. Adhesion promotes cluster formation and prevents fiber overflow, while velocity differences inside and outside clusters cause uneven drag and shear failure. When forces reach equilibrium, stable clusters migrate toward the fracture tip. Greater fracture wall roughness leads to more complex placement patterns but has little effect on filling ratio. With increasing fiber concentration and length, the settling slope angle first decreases and then increases, indicating optimal values that scale with proppant particle size. Higher fluid velocity and viscosity further reduce the settling slope angle. These findings provide theoretical guidance for developing novel fiber fracturing materials and optimizing fracturing parameters.

Landslide-induced surge waves involve complex fluid-solid interactions, making accurate and efficient numerical simulation methods crucial. This study proposes a volume of fluid-discrete element method (VOF-DEM) coupling method, in which the VOF method is embedded within a two-way-coupled computational fluid dynamics-DEM framework. Turbulence is modeled using a Reynolds-stress model, extending the approach to particle-laden, turbulent free-surface flows. In parallel, a virtual dual-mesh porosity algorithm is introduced to compute local fluid porosity with high accuracy. The coupled method is validated by comparing simulated particle column collapses with experiments results. The findings demonstrate that the method effectively captures landslide downslope velocity, surge waveform evolution, propagation characteristics, and deposition morphology. Applied to the RongSong deposit landslide at the RuMei Hydropower Station, the simulation successfully reproduces the landslide motion and surge waves generation and propagation. The results provide reliable physical insights for predicting and mitigating landslide-induced surge waves hazards in mountainous reservoirs.

Due to cost considerations, quartz sand has commonly replaced ceramic proppant in the field. The significant crushing of quartz sand raises concerns about its ability to meet the demands of the fracture network conductivity. Therefore, it is crucial to study the crushing characters of quartz Sand. For faster analysis crushing, a new image processing-based quantitative method is developed to determine the compression proppant crushing rate. Compared with conventional screening methods to verify accuracy. The influence of factors such as sand concentration, particle size combination, and placement method on the proppant crushing rate was analyzed. The results of this analysis were consistent with previous studies, which further confirming the applicability of the proposed method. This research provides a theoretical foundation for hydraulic fracturing and optimization of sand placement in order to maintain long-term high conductivity in fractures.

The nucleation and growth of 10–12 tensile twins in TA2 pure titanium during in-situ tensile test are explored by electron backscattered diffraction (EBSD). The dynamic process of twin growth can be clearly revealed by observe the evolution of microstructure at the same position with cumulative strains increases. According to the statistics and analysis of experimental data, it is found that the Schmid factor (SF) plays an important role during the nucleation and growth of twins. The twin variants with high SF are easy to be activated. However, approximately 18

Rock-breaking is a fundamental process in subsea tunnel construction, particularly during deep trench excavation. Understanding the failure process and permeability evolution of rock masses under hydro-mechanical coupling is essential for optimizing excavation plans and ensuring construction safety. Due to the challenges of capturing crack propagation and permeability evolution solely through experiments, numerical simulations provide a valuable supplement. This study investigates the weathered granite from the deep section of the Shenzhen-Zhongshan Bridge subsea tunnel, employing triaxial compression tests and CFD-DEM numerical simulations under hydro-mechanical coupling conditions. The experimental results reveal that the stress-strain curves can be divided into five distinct stages, with plastic strain increments becoming more concentrated as confining pressure and fluid pressure increase, particularly after the peak strength stage. Numerical simulations further demonstrate that higher confining pressure promotes shear failure, while elevated confining and fluid pressures accelerate the formation of stable microcracks. Contact forces and seepage forces are predominantly concentrated around the primary cracks, with contact forces distributed uniformly along the crack path and seepage forces most pronounced at the inlets and outlets of the seepage channels. These findings offer significant insights into the internal hydro-mechanical behavior of rocks, contributing to the refinement of rock-breaking strategies and the mitigation of geological risks in subsea tunnel construction projects.

Summary Hydraulic fracturing is an effective method for enhancing both the initial reservoir production and ultimate recovery. Nevertheless, the conductivity of proppant fractures is a pivotal factor in the optimization of fracture designs within the context of fracture modification. Experimental testing methods for proppant fracture conductivity are costly and time-consuming, and the physical model is excessively complex and incomplete to account for all the influencing factors, resulting in low computational efficiency. A backpropagation neural network (BPNN) model was constructed using the RAdam optimization algorithm to identify a more efficacious method for predicting the conductivity of proppant fractures. The model was used to predict the fracture conductivity of two data types pertaining to the experimental data on the conductivity of geothermal and volcanic reservoirs. The prediction model is enhanced for three key areas. First, an isolated forest algorithm is used to assess and discard anomalous data points. Second, the objective function is optimized by employing the RAdam optimization algorithm, which has the advantages of both Adam and stochastic gradient descent (SGD). This guarantees rapid convergence and prevents the initial training phase from converging to a locally optimal solution. Moreover, this approach enhances the stability of the model training process. Finally, the rectified linear unit (ReLU) activation function may result in issues related to neuronal activity, including the potential for its disappearance. This study addresses this problem by employing the Kaiming initialization method. The experiments used a series of evaluation metrics, including the mean square error and coefficient of determination, to assess the predictive performance of the two data sets in the two distinct models. The experimental results indicate that the BPNN with RAdam optimization is a more effective approach for data pertaining to volcanic and geothermal reservoirs. Moreover, the prediction of the geothermal reservoir data is more precise than that of the volcanic reservoir data. This model can be used for rapid predictions based on existing fracture conductivity data, which can better guide the design of fracture modifications and is of great importance.

High-elevation landslides impacting water bodies may trigger secondary disasters of landslide-generated impulse waves (LGIWs), which can cause substantial losses. This study employs a coupled discrete element method (DEM) and smoothed particle hydrodynamics (SPH) method to estimate the potential LGIWs hazard induced by the Meilishi (MLS) high-elevation landslide in the Gushui Reservoir. The DEM is used to simulate the landslide failure and sliding processes, while the SPH method is used to capture the water motion. To quantify the interaction between the DEM and SPH simulations, the dynamic boundary condition approach is utilized herein. The results show that the landslide mass exhibits significant deformation and disintegration characteristics during the sliding process. The volume release rate and the volume release ratio at different elevations show similar variation characteristics; however, the sliding velocity and kinetic energy release rate at lower elevations are significantly higher than those at higher elevations. The landslide forms an accumulation body approximately 1013.5 m long in the river channel without causing river blockage. The amplitude of the leading wave is much higher than that of the subsequent waves. There is no obvious wave reflection in the collapse area due to the convex shape of the opposite bank. The maximum water level (MWL) on the right side of the dam is higher than on the left side. The numerical model is then validated using large-scale prototype-specific physical model experiments, and the results indicate a high degree of consistency between the two models. Subsequently, the validated model is used to analyze the influence of entry velocity on the LGIWs hazard. Under the most dangerous condition, the landslide impacts the water with an entry velocity of 81.8 m/s, and the MWL is 2285.92 m on the dam face, showing a low possibility of overtopping. This study provides beneficial insight for accurately understanding the LGIWs hazard induced by MLS landslide and serves as a valuable reference for disaster prevention and mitigation.