Precise simulation of weak structural surfaces in geomechanical model test is critical to their success. This study addresses a key challenge in geotechnical testing: the difficulty of selecting and configuring similar materials that accurately replicate shear strength parameters. An indirect inversion method and process is introduced for determining similar material proportions based on improved adaptive genetic algorithm-backpropagation (IAGA-BP) neural network, which significantly enhances inversion accuracy and concurrently reduces the number of experimental samples required. Talcum powder, sand, Vaseline, and oil are used to configure similar materials, with a focus on simulating the shear strength parameters f and c for weak structural surfaces in the model test. A total of 168 direct shear tests were conducted across various material proportions and normal stresses, producing 168 shear strength datasets and 42 friction coefficient and cohesion datasets. The selected material and configuration method allow for a wide range of shear strength parameters, with a friction coefficient of 0.264-0.687 and cohesion of 0.03-20.66 kPa. In the proposed indirect inversion method, the BP neural network is employed to indirectly predict the shear strength instead of friction coefficient and cohesion. Subsequently, the f and c values are fitted by Mohr-Coulomb criterion based on the predicted shear strength data. Finally, the IAGA algorithm is used to search for the optimal mix ratio close to the target value. To determine the applicability and robustness of the proposed mix ratio determination method, a systematic study is conducted to evaluate the performance of the indirect and traditional inversion methods, investigate the impact of the sample size on prediction accuracy, and analyze the influence of various factors on f and c. The test results demonstrate that, even with a limited number of experiments, the indirect inversion method achieves higher accuracy than the traditional inversion method in mix ratio determination.

During TBM tunneling, timely and effective prediction of energy evolution of surrounding rock is critical for forecasting potential hazards like rockburst, serving as a fundamental safeguard for deep underground construction. So far, most researchers often underestimate the importance of rapid prediction of the energy evolution of tunnel surrounding rock, resulting in the inability to predict specific information such as the location and time of rock bursts. In this study, a surrogate model for predicting the evolution of energy dissipation rate of tunnel surrounding rock based on the static TFT model is proposed to achieve fast time series prediction. Building upon the Temporal Fusion Transformer (TFT) framework, the static TFT model which considers the time invariant nature of tunnel surrounding rock data is proposed. 4373 numerical samples containing 9 surrounding rock energy influencing factors and 12 output features are established and trained on the model guided by the proposed mixed data and physical loss function. The model's performance is evaluated through sample size impact, and ablation feature experiments, as well as comparing the predictive accuracy and fitting effectiveness with baseline models. It is found that the proposed model achieves superior performance across all metrics in predicting surrounding rock energy evolution without redundant features. Specifically, it attains an MAE of 0.0447J center dot m- 3 center dot s- 1, an R2 of 0.9201, and an MSE of 0.0148J2 center dot m- 6 center dot s- 2 for energy dissipation rate prediction. These outcomes signify a substantive advancement in rapid energy evolution forecasting for tunnel surrounding rock and provide an early-warning basis for related geohazards.

针对大型地下洞室围岩变形垂直方向开挖距离效应研究的不足，结合工程监测数据与FLAC3D数值模拟，系统分析了分层下挖条件下围岩变形的时空演化特征及距离效应。并通过定义变形响应强度，构建垂直方向变形曲线（VDP）解析模型。研究结果表明：大型地下洞室分层下挖引起的围岩变形呈现显著“台阶状”增长特征，其变形响应强度随开挖面距离增大呈非线性衰减；通过非线性回归方法反演得出卸荷强度衰减函数λ（x），并建立VDP垂直向变形曲线解析模型，可定量表征分层下挖厚度与围岩变形增量的关联性。工程验证显示，应用于施工期围岩变形监控，VDP拟合曲线与实际监测数据相关系数R达0.92~0.96；同时，估算补埋式监测仪器的丢失变形增量偏差仅7.4 %，在高达11 m开挖区间预测误差仅为1.54 mm。基于“机理揭示-模型构建-工程验证”的完整技术链，研究成果可为地下洞室智能建造安全监控与稳定性评估提供新工具，具有重要应用价值与实践意义。

Objective The operational safety of large dams during earthquakes is a critical concern for national security and public welfare. In China's southwestern region, which is both a seismically active zone and a hub for numerous high dams and large reservoirs, the convergence of abundant hydropower resources and high seismic risk presents a significant challenge. Traditional dam safety monitoring systems often face severe technical bottlenecks under extreme seismic conditions, characterized by susceptibility to disconnection (power and communication outages), prolonged data acquisition times, and poor inter-system coordination. This results in a critical gap: awareness of an earthquake event without timely knowledge of the dam's structural integrity. This study aims to address these industry-wide challenges by developing and implementing a robust, multi-level emergency monitoring system tailored for seismic conditionsMethods Focusing on the Yalong River Basin, a major clean energy base, this research establishes a comprehensive, closed-loop technical system encompassing "information perception - rapid transmission - analysis and decision-making." Key methodological innovations include: 1) Enhanced System Resilience: A distributed Uninterruptible Power Supply (UPS) system and a redundant Ka-band satellite communication link were designed and deployed. This ensures continuous operation and data transmission capability during power grid failures and terrestrial network disruptions caused by earthquakes. 2) Advanced Information Perception: Field hardware was upgraded using DAU3000 intelligent data acquisition units, which feature a fully parallel architecture. This significantly improves data collection efficiency and transmission speed. 3) Dual-Source Triggering Mechanism: An innovative technical scheme employing "Reservoir Seismic Network + Dam Strong Motion Monitoring System" for dual-source joint excitation was proposed. Earthquake risk scoring functions (R₁ based on seismic source parameters, and R₂ based on weighted strong motion instrument responses) were designed. This mechanism automatically triggers the dam safety monitoring automation system to initiate encrypted patrol surveys when thresholds are exceeded, ensuring rapid post-earthquake response. 4) Efficient System Linkage: A real-time synchronization mechanism between on-site systems and the central data center (Chengdu) was developed, replacing traditional periodic database polling with active data pushing. Furthermore, a cross-system virtual triggering function was created, providing a manual override and fault-tolerant backup to initiate data collection remotely via personal computers or mobile terminals, enhancing system coordination and robustness.Results and Discussions The constructed multi-level linkage response architecture for dam safety emergency monitoring has been successfully validated through practical application during actual seismic events within the basin. The system demonstrates high reliability and efficiency: For earthquake events with an instrumental intensity of II degrees or above, the system successfully triggers automated responses. The strong motion monitoring system completes data resolution and pushes earthquake intensity information typically within 2-3 minutes. Subsequently, the safety monitoring automation system completes the collection of all key monitoring point data across an entire dam within approximately 3 minutes. All monitoring data can be synchronized to the central Yashuiyun cloud platform in Chengdu within 5 minutes via the conventional dedicated network, or within 15 minutes via the Ka satellite backup link. Compared to the pre-upgrade state involving manual processes and single-source triggers, the system achieves an approximate 88.2% reduction in single-station patrol survey time, a 94.4% reduction in data synchronization time to the information center, and a 95.8% reduction in strong motion information resolution time. The system transforms the emergency monitoring workflow from a manual, experience-driven process to a fully automated, data-driven operation.Conclusions This research systematically addresses the key bottlenecks of power supply, communication, information perception, and system coordination in dam safety monitoring under seismic conditions. The implemented multi-level emergency monitoring system fundamentally enhances the real-time perception capability of the structural behavior of dams in the Yalong River Basin. The closed-loop technical system, integrating power and communication guarantees, efficient information sensing, and automated system linkage, provides a replicable paradigm for industry-wide dam safety management. It effectively facilitates a shift in emergency management strategy from "post-event disposal" to "pre-event warning and in-event control." The system's open and compatible architecture allows for future integration with advanced intelligent perception technologies such as drone inspections and machine vision, promising further enhancements in coping with complex disaster scenarios.

Adequate calibration of material parameters is the prerequisite for credible long-term deformation prediction of reservoir bank slopes. In this study, a creep parameter inversion method accounting for water effect and mechanical characteristics of rock masses is proposed. The elasto-viscoplastic model based on internal variables is introduced in inversion, which incorporates transient pore pressure effect, progressive strength degradation in hydro-fluctuation belt and saturated zone, as well as dam-foundation interaction induced by periodic water level fluctuations. The inversion process integrates a metaheuristic algorithm (improved adaptive genetic algorithm, IAGA) with a BP neural network-based (BPNN) surrogate model. A segmented and incremental strategy is implemented in objective function to capture the spatio-temporal heterogeneity of the deformations observed at each point. Besides, random perturbation coefficients, derived from statistical experimental results of multiple hydropower projects, are introduced to constrain friction coefficient (f) and cohesion (c), addressing the heteroscedastic nature of strength parameters. Leveraging deformation measurements spanning approximately 17 years from 27 observation points during the construction and impoundment periods, an inversion is performed on 48 creep parameters across 8 materials of a near-dam slope. Based on the calibrated parameters, predictions are made for the convergence time, stabilization time, and ultimate deformation. The results indicate that the calculated deformation aligns well with field observations, revealing that certain portions of the slope remain in a stress adjustment phase. The predicted deformation convergence is expected between 2025 and 2036, with stabilization occurring between 2034 and 2039, and an ultimate deformation ranging from 165 to 215 mm.

The surrogate model serves as an efficient simulation tool during the slope parameter inversion process. However, the creep constitutive model integrated with dynamic damage evolution poses challenges in development of the required surrogate model. In this study, a novel physics knowledge-based surrogate model framework is proposed. In this framework, a Transformer module is employed to capture strain-driven softening-hardening physical mechanisms. Positional encoding and self-attention are utilized to transform the constitutive parameters associated with shear strain, which are not directly time-related, into intermediate latent features for physical loss calculation. Next, a multi-layer stacked GRU (gated recurrent unit) network is built to provide input interfaces for time-dependent intermediate latent features, hydraulic boundary conditions, and water-rock interaction degradation equations, with static parameters introduced via external fully-connected layers. Finally, a combined loss function is constructed to facilitate the collaborative training of physical and data loss, introducing time-dependent weight adjustments to focus the surrogate model on accurate deformation predictions during critical phases. Based on the deformation of a reservoir bank landslide triggered by impoundment and subsequent restabilization, an elasto-viscoplastic constitutive model that considers water effect and sliding state dependencies is developed to validate the proposed surrogate model framework. The results indicate that the framework exhibits good performance in capturing physical mechanisms and predicting creep behavior, reducing errors by about 30 times compared to baseline models such as GRU and LSTM (long short-term memory), meeting the precision requirements for parameter inversion. Ablation experiments also confirmed the effectiveness of the framework. This framework can also serve as a reference for constructing other creep surrogate models that involve non-time-related across dimensions.

Accurate identification of multi-slip surfaces in complex slope failure modes is essential for risk assessment and reinforcement design. In this study, a multi-slip surfaces extraction method based on energy dissipation rate (EDR) index is proposed. Firstly, the elasto-viscoplastic damage model is employed to simulate slope failure procedure. Then, a ridge-finding technique is applied to the EDR field to generate a feature point set, which is subsequently denoised through convex hull check and statistical threshold. Additionally, density-based spatial clustering of applications with noise (DBSCAN) is adopted for the initial clustering of feature points, followed by multi-model fitting via random sample consensus (RANSAC). The proposed method is validated based on a homogeneous slope and an undrained clay slope with a thin weak layer. Moreover, it is applied to a practical reservoir bank slope to analyze the spatial distribution of multi-slip surfaces and stability evolution during impoundment. The results indicate that, under complex stress conditions, EDR offers distinct advantages over traditional indices such as displacement and shear strain increment. The analyzed reservoir bank slope exhibits a potential risk of overall instability along the inherent surface of rupture, alongside several local instability modes. The developed approach provides a novel strategy for automated identification of multi-slip surfaces, significantly reducing manual intervention.

Direct shear is the stress path that rockmass undergoes often and is thus important to the stability evaluation of underground openings in deep tunnels. This study introduces a self-developed true triaxial direct shear device capable of achieving large displacement direct shear, providing a basis for the multi-stage shear experiments. Direct shear experiments were conducted on intact granite samples from the Beishan underground research laboratory under two stress paths respectively with lateral stress greater and smaller than the normal stress. Additionally, the influence of loading and unloading paths on triaxial direct shear behaviors at the residual stage was investigated through multi-stage shear experiments. It shows the peak shear strength experiences an increase of 11.3 similar to 38.1 % when the lateral stress increases from 5 MPa to 35 MPa. Furthermore, the roughness of rock fracture surfaces was also affected importantly by the lateral stress. At multiple scales, increasing lateral stress expands the tensile fracture range, enhancing rock brittleness. When normal stress is 20 MPa, the increase of lateral stress from 5 MPa to 35 MPa reduces the brittle coefficient from approximately 5.40 to 3.80. Additionally, the effect of lateral stress on residual strength is closely tied to the contact state of the fractured rock surface, and in multistage tests, when lateral stress exceeds normal stress, the shear strength during lateral stress unloading consistently surpasses that during loading. The findings quantified the effect of lateral stress on the direct shear properties of rocks and can provide a reference for modifying the rock strength and deformability in the stability evaluation of deep hard rock openings.

Objective The construction of hydraulic tunnels in high-stress surrounding rock environments often leads to the occurrence of rock bursts, thereby posing a substantial threat to engineering safety. Among the various active prevention and control measures for rock bursts, drilling pressure relief in surrounding rocks is considered a relatively economical and effective method. By creating drilled holes in the rock mass, stress concentration can be redistributed, thereby reducing the likelihood of sudden failures and improving the overall stability of the tunnel structure. Methods In order to investigate the mechanical properties and damage characteristics of sandstone in hydraulic tunnels under different combinations of drilling numbers and drilling depths, a series of uniaxial compression tests were conducted. These tests utilized an advanced uniaxial compression testing machine and the VIC-3D noncontact full-field strain measurement system. The experiment involved eight different combinations of drilling holes in the sandstone specimens. This study comprehensively analyzed key parameters such as compressive strength, the accumulation and release characteristics of elastic strain energy, and the residual volume rate of sandstone. A regression analysis was conducted to establish a quantitative relationship between the residual volume rate of sandstone and its compressive strength. In addition, the crack evolution and damage characteristics of sandstone under different drilling hole configurations were studied using digital image correlation (DIC) technology and fracture phase field simulation. Furthermore, numerical simulations based on finite element methods were performed to compare the effects of straight holes and 10u00B0 inclined holes on stress redistribution within the rock mass. Results The experimental and numerical results led to the following key findings: (1) when the radius of the drilled holes remains constant, an increase in the drilling depth leads to a decrease in the compressive strength of sandstone. This finding indicates that deeper drilling can effectively weaken the rock mass and facilitate stress relief. (2) Under the condition of identical hole radius and depth, an increase in the number of drilled holes results in a discontinuous reduction in the compressive strength of sandstone. Moreover, the arrangement of the drilled holes plays a crucial role in determining the overall strength of sandstone. For instance, the specimens with asymmetrical three-borehole configurations exhibited lower compressive strength than those with symmetrical four-borehole configurations. This finding suggests that asymmetrical arrangements can enhance energy dissipation efficiency and reduce the overall stress level within the rock. (3) The elastic strain energy of sandstone exhibits a strong positive correlation with compressive strength. Moreover, as the ratio of loss energy to elastic strain energy approaches zero, the intensity of sandstone destruction considerably increases. This outcome highlights the role of energy release in the failure process of rock materials. (4) DIC strain field analysis and numerical simulations confirm that sandstone under uniaxial compression follows a characteristic butterfly-shaped damage pattern. The three-borehole asymmetric configuration showed lower compressive strength, greater far-field stress reduction, earlier failure onset, and higher economic feasibility for pressure relief applications than the four-borehole symmetric configurations. (5) Under identical rock formation and borehole depth conditions, the impact of straight and 10u00B0 inclined boreholes on stress redistribution is found to be similar. However, practical construction decisions should be made, considering site-specific conditions and operational requirements. Conclusions This study provides valuable insights for optimizing the design of borehole pressure relief schemes for hydraulic tunnels. The findings provide a reference for engineers seeking to improve tunnel stability through effective stress redistribution strategies. By systematically evaluating different drilling configurations, this study contributes to the development of more efficient and cost-effective methods for mitigating rock bursts in high-stress environments.

The construction/operation of ultrahigh arch dams may impose significant perturbations to surrounding mountains, resulting in landslide motions of rock slopes and endangering the safety of hydropower systems and human habitats. For example, the Laxiwa Hydropower Station in China witnessed its nearby Guobu slope displacing significantly after the reservoir impoundment and having so far displaced up to 40 m. It is of great importance to understand the mechanisms driving this large deformation. Here, we present some preliminary results from a combined remote sensing and numerical modelling investigation of this slope before, during, and after the reservoir impoundment. Analysis based on the differential interferometric synthetic aperture radar (DInSAR) data indicates that the slope had already been actively creeping at a rate of 10 cm/year (e.g. in years of 2003–2005). We develop a geological model including different rock mass compartments and various discontinuity structures as well as a realistic representation of the suspended ancient landslide. We model the coupled hydro-mechanical and creep behaviour of the slope in response to reservoir impoundment. A good agreement is reached between the simulation results and field measurements of slope displacement time series recorded at different elevations of the slope surface. Our results show that the reservoir impoundment causes notable pressure changes at the toe region of the slope, leading to strong deformations (under coupled poroelastic and primary creep effects) that propagate upslope with the ancient landslide partially reactivated. These deformations tend to decelerate significantly after the impoundment due to the transition to secondary creeps.

The stability of rock slopes along the reservoir bank is very important for achieving safe operation of a hydropower station during its impoundment period. The Guobu slope at the right bank of the Laxiwa Hydropower Station is a typical ultrahigh rock slope (700 m) and has continuously shown large deformations during and after the reservoir impoundment. In this paper, some preliminary results are presented from a combined field observation, in-situ monitoring, and experimental investigation of this slope. Based on a detailed analysis of the monitoring data of the slope and the impoundment procedure of the reservoir, it is found that the deformation of the Guobu slope shows a clear correlation with the temporal variation of the reservoir water level. According to the filed observation and inspection of the slope surface and internal deformation, the Guobu slope may be divided into three zones with their boundaries approximately defined: (i) the upper highly fractured rock mass with large deformation, (ii) the middle moderately fractured rock mass at a near-critical state, and (iii) the lower less fractured rock mass at a stable condition. Based on the observed correlation between the slope deformation and reservoir impoundment, a conceptual model interpreting the deformation mechanism of the Guobu slope is proposed. Reduction of effective stress and mechanical properties of the rock mass at the toe region of the slope, as a result of the rise of the reservoir water level, triggers the toe of the slope to deform and further induces the rock mass at the upslope region to show large deformation dominated by toppling. Furthermore, we present some experimental results to elucidate the effect of water on rock strength, which support the above conceptual model based on the mechanism of water-induced strength degradation of granitic rocks.

ObjectiveThere have been numerous cases of dam failures caused by slope instability in water conservancy projects domestically and internationally, leading to significant casualties and property losses. Therefore, the stability of the slopes near the river banks is crucial for the operational safety of dams and hydropower stations. The Jinping Dam, features a massive engineering slope on its left bank. For approximately 10 years following the commencement of operations at the Jinping I Hydropower Station, certain areas near the dam on the left bank have exhibited persistent deformation, and the underlying mechanisms are still not fully understood. The presence of deep cracks on the left bank significantly influences the selection of the arch dam axis during the design phase. MethodsThis paper utilizes numerical simulation to analyze the variation characteristics of in-situ stress during the evolution of the river valley, aiming to clarify the influencing factors and mechanical mechanisms behind the formation of deep fractures. Additionally, by leveraging monitoring data on slope deformation and conducting a creep analysis of river valley evolution, this paper examines the relationship between deep fractures and the continuous deformation of the slope during operation. Results This study found that the formation of deep fractures was closely related to valley incision, the complex geological conditions, and the tectonic stress of the valley slope. Owing to these factors, the unloading depth gradually decreased with a decrease in elevation. Consequently, when the rock mass at higher elevations experienced unloading at greater depths, the rapid release of strain energy occurred during the later stages of valley incision, which led to the unloading failure of the deep rock mass and the formation of deep fractures. Moreover, although the rock mass at lower elevations only experienced surface unloading failure, as the depth increased, the ratios of principal compressive stress and principal tensile stress decreased and the former remained at a high value. This resulted in compressive shear failure and the formation of deep fractures in the deep rock mass. Monitoring data showed that the arch thrust of the dam body hindered deformation in the empty direction, and displacement at high elevations was mainly due to gravity-driven tipping deformation. A comparison of apparent and deep deformation revealed that lamprophyre dikes, faults, and other weak zones were the main factors affecting slope deformation during operation, while the influence of deep fractures was not significant. Additionally, the tectonic stress of the mountain on the left bank had minimal impact on slope deformation. The results of the creep calculations were consistent with the observed deformation patterns of the slope during operation, further confirming that the influence of tectonic stress on slope deformation had dissipated.ConclusionsIn summary, this paper identifies the causes and mechanical mechanisms of deep fractures through finite element calculations and analysis of measured monitoring data, enhancing the engineering understanding of such slope issues. Compared with weak zones such as faults, deep fractures have less influence on the long-term deformation of the slope. The influence of tectonic stress on slope deformation has dissipated, which provides a foundation for further studies on the left bank slope of Jinping I Hydropower Station.

Abstract Onsager fluxes proposed by D.G.B. Edelen assume that the same symmetry, nonlinear Onsager reciprocal relations, holds near and far from equilibrium. This assumption leads to exact differential 1-form J ⋅ dX everywhere, where J and X are thermodynamic fluxes and forces, respectively. However, thermodynamic fluxes far from equilibrium are characterized by symmetry breaking, which lead to the inexact differential 1-form. It is shown in this paper that the inexact differential 1-form J ⋅ dX should be represented by multiple independent scalar-valued functions. Generalized Onsager fluxes are obtained based on such representation. Generalized Onsager fluxes do not satisfy the nonlinear Onsager reciprocal relations and contain multiple independent scalar-valued functions, so they are suitable to thermodynamic fluxes far from equilibrium. Generalized Onsager fluxes embody Onsager fluxes as a special case. Therefore, generalized Onsager fluxes provide a unified framework for thermodynamic fluxes near and far from equilibrium.

Understanding and predicting the shear mechanical characteristics of rock fractures subjected to different pore water pressures is vital for the safety and the stability of the reservoir bank slope. A series of direct shear tests were conducted on granite fractures covering various reservoir impoundment-relevant pore water pressures. The results reveal that as pore water pressure increases, the peak shear strength, residual shear strength, shear stiffness, and peak shear displacement show a decreased trend, while the peak dilation increases with the pore water pressure. The shear-induced reductions in fracture surface parameter ( Z 2 ) and the sheared-off volumes of asperities decrease with the increase in pore water pressure. An empirical function is presented to successfully describe the evolutions of the shear parameters and the volume of sheared-off asperities with pore water pressure. Under drained conditions, it is found that shear failure will cause a transient drop in pore water pressure, and the pore water pressure fluctuation coefficient can be used as an early warning indicator for potential damage of fractures. In addition, a singular-value decomposition method is employed to fit the peak and residual shear strength with pore water pressure, which indicates that the effective stress principle can satisfactorily explain the hydro-mechanical mechanism of rock fractures. Before the shear failure, the effective stress coefficient is close to unit, while it is reduced after the shear failure due to the increase in the actual contact area caused by asperity degradation and gouge formation.

ABSTRACT Slope failure near a reservoir area is mainly influenced by reservoir impoundment and rainfall. In this paper, a bank slope on the Lancang River near a hydropower station in Southwest China is studied. The FEM numerical model is established to investigate the deformation and stability of the bank slope, especially the influence of the reservoir impoundment and rainfall. Unbalanced force (UST) and the factor of safety of the potential failure surface are presented to quantitatively evaluate the monolithic stability of the slope. The numerical analysis results show that the slope reaches a critical state after the first impoundment. The maximum deformation is mainly located at the toe of the slope towards the riverbed. The failure mechanism can be explained that the bank slope especially the toe of the slope is softened by rainfall and water impoundment and the effective stress effect makes it easier for the slope to yield, which results in significant deformation or landslide failure. It is also demonstrated that the toe reinforcement has a significant effect on restraining the shear displacements of the slope and preventing the progressive failure of the landslide. This study presented a feasible approach for more accurate analysis of reservoir slope stability during impounding and rainfall, and also provided practical measures for slope reinforcement. INTRODUCTION Landslides will cause major threat to both lives and property worldwide, especially in regions of steep terrain and heavy rainfall (Xu et al., 2014). Due to the impoundment of many large reservoir dams, more and more landslides have been triggered and cause serious damage. Water is an important factor that influences the mechanical behaviour of geotechnical materials and is responsible for many engineering hazards (Vásárhelyi and Ván, 2006; Wang et al., 2014). The weakening effects caused by water will significantly influence the stability of the reservoir slope.

Investigating deformation behavior of fractured reservoir bank slopes during impoundment plays an essential role in safety control of high arch dams. Because of changes in water levels during impoundment, the fractures are subjected to varying water pressure and cyclic wetting–drying conditions. In this numerical study, mechanical, physical, and chemical water–rock interactions in rock fractures are incorporated into discrete element modeling of fractured reservoir bank slopes, and effects of mechanical, physical, and chemical water–rock interactions on deformation of fractured reservoir bank slopes are investigated based on a case study of valley deformation at the Xiluodu Hydropower Station. The mechanical water–rock interaction is considered by the effective stress law, while the physical and chemical water–rock interactions are modeled by some empirical deterioration laws. The results showed that the modeled and monitored deformation behaviors of fractured reservoir bank slopes are in good agreement. The slip of shear zones due to water pressure elevation and mechanical parameter weakening is the main reason of impoundment-induced valley contraction. Discrete element method considering water–rock interactions can serve as a robust and reasonable tool to predict the deformation magnitude of fractured reservoir bank slope.

The stability of reservoir bank slope during impoundment is significant for the safe operation of hydropower stations. The deformation evolution of a slope adjacent to dam is analyzed based on the field monitoring data. The main influencing factors and spatial distribution of the slope deformation during impoundment are identified based on the multiple regression model and K -means cluster analysis, respectively. Subsequently, the deformation mechanism of the slope is analyzed by three-dimensional nonlinear finite element method. Then, an inversion analysis method based on improved adaptive genetic algorithm and back propagation neural network is proposed. The measured deformations of 44 monitoring points are used to inverse a total of 27 mechanical parameters, including deformation parameters, strength parameters and Biot coefficients of 9 materials. Finally, the influence of sample numbers on the prediction accuracy and the robustness of the neural network are discussed. The results indicate that the deformation rate of the slope is substantially associated with the impounding process. The water pressure component and aging component of the deformation account for a relatively high proportion while the temperature component is negligible. The deformation is mainly affected by material softening and effective stress of shallow-buried fractured rock mass, in which the latter dominates. The water load on the dam surface is the deformation inducement of the deep rock mass. The proposed inversion method can reasonably obtain the weakening rate of materials and make the numerical model accurately fit in with the actual state of the slope.

Understanding and predicting the frictional behavior of rock joints subject to dynamic cyclic loading is essential for the seismic safety of many rock engineering problems. In this paper, cyclic shear experiments were conducted based on a shaking table apparatus. A series of cyclic friction tests covering various earthquake-relevant frequencies was conducted on dry planar joints of granitic rock. The friction behavior especially the dynamic evolution of the frictional strength under cyclic loading was investigated, with the influence of loading frequency on the frictional strength weakening and energy dissipation quantified. Based on the experimental results, we propose a phenomenological model characterizing the dynamic frictional behavior of planar granite joints under cyclic loading. This empirical formulation relates the evolution of joint frictional strength to the number of cycles of dynamic loading. As the frequency increases, the frictional strength and dissipated energy density both show a pronounced reduction, while the strength weakening ratio and the critical number of cycles (at which the residual strength is reached) increase in a logarithmic fashion. The frictional strength evolution is described by the product of the peak frictional strength and a damage function, which is further related to the number of cycles and the frequency. In addition, a close inspection of fracture surface morphology after the cyclic loading was conducted to understand the role of surface wear and gouge formation in the frequency-enhanced dynamic friction weakening of rock joints.

Closely inline with the urgent needs of long-term safety regulation of major water conservancy and hydropower projects in China, the dynamic evolution and coupling mechanism of seepage field, stress field and parameter field (geotechnical mechanical performance parameters and functional indexes of reinforcement system) in the reservoir area and project area are deeply studied under the conditions of excavation and unloading, reservoir water storage, water level alternation, flood discharge rain and fog and climate change in the reservoir area, and the theory and method for predicting the change trend of geological environment are established and formed. The mutual feeding mechanism between different structures (reservoir bank slope, high dam rock foundation and large underground cavern group) and the geological environment is systematically studied, the dynamic analysis method and regulation technology of working properties of structures are improved and developed based on the evolution of geological environment, and the complete set of long-term life extension equipment and construction method of reinforcement system (especially anchorage system) adapted to the changing environment is developed, and it is demonstrated and applied in practical engineering.

Currently, various machine learning (ML) techniques have been developed to solve geotechnical engineering problems. However, the lack of representative field samples limits the application of ML models. A shield jamming risk prediction method based on numerical samples and random forest (RF) classifier is proposed. The database with samples of different shield jamming risk levels is established by numerical simulation of the TBM construction process. By setting different values and combinations, seven influencing parameters, i.e., advance rate, overcut, elastic modulus, tensile strength, in situ stress, maximum thrust and friction coefficient are considered. The shield jamming risk level is determined according to the ratio of the total friction force to TBM residual thrust. Feature importance analysis indicates that elastic modulus, overcut and in-situ stress are the major influencing factors of shield jamming. Based on the labeled database, the RF model integrating multiple decision trees is established to capture the complex relationship between shield jamming risk and different influencing factors. The trained model has shown a good prediction performance on test set, and the prediction results of six field instances in DXL tunnel are in good agreement with the actual shield jamming situation. Compared with other conventional classifiers, i.e., support vector machine (SVM), k-nearest neighbors (KNN), decision tree (DT) and logistic regression (LR), the proposed FR classifier has stronger prediction accuracy and generalization ability. Additionally, the influence of sample numbers and sample imbalance is discussed.