[Objective]With the continuous expansion and increasing complexity of water diversion tunnels in hydropower projects,their long-term structural safety has become a critical engineering challenge.Conventional safety evaluation methods often rely on qualitative assessments or multi-index systems,which are highly subjective and fail to adequately account for progressive material deterioration and time-dependent deformation.This study proposes an integrated quantitative framework that combines real-time inversion of mechanical parameters and deformation evolution analysis to dynamically evaluate the structural safety of tunnels.[Methods]The proposed framework integrates four major components:data preprocessing,parameter inversion,deformation simulation,and safety evaluation.First,the raw deformation monitoring data are preprocessed by imputing missing values using the K-nearest neighbors(KNN)algorithm,identifying and correcting outliers with a sliding-window Z-score method,and reducing noise through logarithmic trend fitting.Second,a physics-informed inversion approach combining deep learning architectures-fully connected layers(FCL)and gated recurrent units(GRUs)—with Bayesian optimization is established to infer the current mechanical parameters of the tunnel from preprocessed deformation data.Third,an elastoviscoplastic damage creep constitutive model based on internal variable thermodynamics is employed to simulate deformation behavior under various material-degradation scenarios,represented by different strength reduction coefficients(Kr).Finally,based on the analysis of material creep behavior and deformation evolution patterns,a time-dependent"3S"safety evaluation index system is established.This system comprises the long-term deformation acceleration safety coefficient(S1),the nonlinear deformation initiation safety coefficient(S2),and the short-term deformation acceleration safety coefficient(S3).The safety state of the tunnel structure is quantified according to the relevant deformation evolution characteristics using the proposed 3S safety coefficients.The physical implications of these indices are as follows:S1 characterizes the critical point at which the structure transitions into an accelerated creep phase under continuous strength attenuation,indicating the long-term instability risk;S2 reflects the onset of deviation from the linear response during initial deformation,marking the beginning of dominance by nonlinear mechanical behavior;and S3 indicates the threshold for notable acceleration of deformation within a defined short-term period,serving as an indicator of potential sudden instability.[Results]The proposed method was implemented in the JLL Tunnel,a 20 km-long underground structure located in Hunan Province,China,which features complex geological conditions.The mechanical parameters were successfully inverted from field monitoring data,with simulated deformation curves showing high agreement with the measured values.Numerical simulations under different Kr conditions revealed distinct deformation patterns.For Kr ≥ 0.7,deformation stabilized after initial convergence.When 0.4≤Kr≤0.6,deformation exhibited slow growth,followed by an acceleration phase.For Kr≤0.3,deformation accelerated rapidly within a short time.The computed 3S safety coefficients were S,=2.4-3.0,S2=3.7-4.6,and S3=5.7-7.3,indicating that the tunnel is currently in a safe state with sufficient safety margins.These results validated the method's effectiveness in distinguishing between short-and long-term risks and in providing early safety warnings through deformation trajectory analysis.[Conclusions]This study proposes an integrated quantitative framework for tunnel structural safety evaluation that effectively combines real-time monitoring data,physics-based modeling,and deformation evolution analysis.The established 3S index system provides a refined insight into structural behavior under material degradation and enables safety assessment across multiple time scales.Compared with conventional methods,the proposed framework enhances objectivity,supports the dynamic prediction of time-dependent performance,and facilitates lifecycle safety management and preventive maintenance of tunnel structures.The methodology demonstrates strong generaliza bility and offers remarkable practical value for risk prevention and sustainable operation in tunnel engineering.

Developing sustainable cementitious materials with improved mechanical performance is important for reducing the environmental impact of concrete construction. This study investigates the effects of activated carbon (AC) incorporation on the mechanical performance and crack propagation behavior of concrete through an AI-informed multiscale framework. A deep learningassisted five-phase heterogeneous model is proposed to explicitly account for the distinct interfacial transition zones (ITZs) around both natural aggregates and AC particles, representing a departure from traditional three-phase approximations. Using the phase-field method, twodimensional (2D) high-fidelity finite element (FE) simulations were performed to analyze the influence of AC particles and ITZs on crack initiation, propagation, and damage localization under compressive loading. The simulation results suggest that dispersed AC particles can modify crack propagation paths by promoting crack deflection, branching, and increased crack-path tortuosity, thereby delaying the coalescence of dominant macrocracks. The crack evolution behavior is mainly affected by the spatial distribution and interaction of aggregate-related ITZs and ACrelated ITZs. Experimentally, regardless of the AC type used, the 5 wt% replacement level exhibited the most favorable mechanical performance within the investigated range, with an increase of more than 13% in elastic modulus and more than 14% in compressive strength. These improvements are likely associated with the formation of hydration products within the porous AC structure and the interaction between AC particles and crack propagation paths. Overall, this study provides an image-based multiscale modeling framework for understanding the mesocrack evolution mechanisms of AC-modified concrete and offers useful guidance for the mix design of sustainable cementitious composites with potential low-carbon benefits.

During the long-term service of water diversion tunnels, the mechanical properties of the surrounding rock and lining materials deteriorate over time due to various physicochemical actions, leading to cumulative internal mesoscopic damage. To evaluate the long-term safety of tunnels, initial mechanical parameters of the surrounding rock are obtained through displacement inversion. Utilizing a creep damage model based on internal variable thermodynamics, which considers material degradation and structural damage of tunnel, the long-term evolution process of tunnel deformation has been simulated. Furthermore, a long-term safety assessment of the tunnel is conducted based on deformation indicators. The results demonstrate that under different conditions of degradation and damage, the time-dependent deformation of tunnels exhibits distinct evolutionary characteristics. When the strength reduction coefficient is no less than 0.7, tunnel deformation follows a pattern of initial increase followed by stabilization. In contrast, when the strength reduction coefficient is below 0.6, tunnel deformation initially increases rapidly, then gradually slows down before accelerating again. As the strength reduction coefficient decreases, the magnitude of tunnel deformation increases, and the onset time for secondary acceleration occurs earlier. Based on these deformation evolution characteristics, a 3S evaluation method for assessing the structural safety factor of tunnels is proposed.

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.

ABSTRACT: With the continuous expansion of underground engineering activities into deeper areas, the occurrence of deep geological disasters such as rock bursts is becoming increasingly frequent. The incubation and development process of these disasters can essentially be revealed through the energy evolution law inside the surrounding rock. This study combines numerical simulation with deep learning to construct a surrogate model. It is designed to characterize the energy evolution of rocks in a deep‑buried water diversion tunnel .A constitutive model derived from internal variable thermodynamics was adopted. It enabled the numerical simulation of tunnel excavation and the generation of numerical samples. The ISSA-CNN-Attention- BiGRU deep learning algorithm was introduced. This enabled the construction of a surrogate model that links the mechanical parameters of dynamically excavated tunnel rock to its energy dissipation rate. Using this surrogate model, the energy evolution curve of the rocks is predicted. This facilitates the rapid forecasting of the energy dissipation rate time series over the entire construction period.

Objective High arch dams impose stringent requirements to ensure safety, requiring robust bearing capacity, deformation control, and resistance to seepage failure. The stability of the dam foundation serves as the cornerstone of the entire arch dam system. During operation, the enormous thrust generated by arch abutments acts on the dam-foundation interface, potentially inducing instability risks such as macroscopic fractures and shear sliding, particularly in weak foundation zones. These risks, if left unchecked, can compromise dam safety and may trigger catastrophic failure. Addressing weak zone reinforcement design in complex dam foundations poses a significant challenge, as no standardized system currently exists for prioritizing reinforcements or quantifying stability evaluation indicators. Methods To address this gap, this study proposes an energy-based method for stability evaluation and reinforcement design of weak dam foundation zones. A stability evolution analysis model was established using energy dissipation rate and domain integral variation, enabling the identification of critical weak zones and their evolutionary patterns. The study employed a three-dimensional numerical model of the arch dam-foundation system, accounting for complex geological factors such as faults, abutment slopes, and dam geometry. A thermodynamically driven creep constitutive model with internal variables was employed to conduct three-dimensional numerical simulations, revealing the stability evolution process of weak foundation zones. By analyzing energy dissipation rate curves and domain integrals, critical moments (marked by peak dissipation rates) and vulnerable areas (highlighted by energy concentration zones) were pinpointed. This method was then applied to parallel fault groups in a high arch dam foundation, with the reinforcement effectiveness analyzed in terms of energy dissipation rates, dam deformation, fault yield zones, and results from comparative testing using the super-water unit weight method. Results Results indicate that energy dissipation rates and domain integrals for abutment faults initially increased rapidly after reservoir impoundment, gradually decreased, and eventually stabilized. The stability evolution of dam foundation faults under impoundment exhibits distinct time-dependent behavior, progressing through three phases: instability, transition, and stabilization. A significant observation is the delayed occurrence of peak energy dissipation rates in downstream faults, reflecting a spatiotemporal hysteresis in arch thrust transmission. During normal operations, the thrust from the arch extends its influence on deep foundation stability to a distance approximately twice the width of the arch abutment. However, its impact on downstream stability ranges between 2-3 times the abutment width. Comparative analysis using the super-water unit weight method demonstrated reduced dam deformation, improved fault yield zone distribution, and significant decreases in energy dissipation rates and domain integrals for critical faults after reinforcement. Conclusions The proposed method reveals spatiotemporal hysteresis in arch thrust transmission and its disturbance on structural stability. For multifault dam foundations, upstream faults exhibit less susceptibility to hydraulic disturbances when compared to downstream faults. Weak zones in downstream faults are primarily concentrated near their intersections with the dam abutment as well as along the strike direction. The f123 and f120 faults on the left bank were identified as critical to global stability, with key reinforcement areas at elevations of 2440-2470 m (f123) and 2395-2425 m (f120). Targeted reinforcement measures effectively enhanced fault and foundation stability, significantly improving the overall stability of the arch dam-foundation system.

The position and shape of the slip surfaces in reservoir bank slopes, which is essential for stability analysis and risk evaluation, are dynamically influenced by external factors such as excavation and impoundment. This study proposes a potential multi-slip surfaces automatic identification and rapid extraction method based on energy dissipation rate index (EDR). Firstly, the spatial distribution of the EDR at each moment is calculated via an elasto-viscoplastic model within the internal variable thermodynamic framework. A ridge-finding technique is then applied to the EDR field to build a feature point set for slip surface extraction, which is subsequently denoised through convex hull check and statistical threshold. 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), enabling the progressive and automatic extraction of primary and secondary slip surfaces even under dense noise conditions. The proposed method is then validated and applied to a reservoir bank slope. The results indicate that the slope has five localized failure modes coexisting with the inherent surface of rupture. The proposed method effectively reduce the need for manual intervention and enhancing the robustness of the program.

ObjectiveCarbonaceous mudstone exhibits low mechanical strength, water-induced disintegration, and reduced structural compactness. It is widely distributed in the mountainous strata of Southwestern China, where tunneling activities frequently traverse formations dominated by carbonaceous mudstone. The surrounding rocks are subjected to prolonged coupled effects of in-situ stress and groundwater seepage pressure, which complicates the creep behavior of carbonaceous mudstone. Constitutive models are essential for characterizing creep-related mechanical properties and deformation mechanisms. However, current models derived from elastoplastic theory inadequately address hydro-mechanical coupling effects during creep. Therefore, establishing a coupled hydro-mechanical creep constitutive model is imperative for accurate deformation prediction in geotechnical engineering.MethodsThis study conducted laboratory creep tests under hydro-mechanical coupling conditions on carbonaceous mudstone samples collected from secondary lining fracture zones in the Yanglin Tunnel. The experiments revealed the evolution of deformation characteristics and macro-micro fracture mechanisms throughout the creep process. A three-stage nonlinear viscoelastic-plastic creep model incorporating hydro-mec-hanical coupling was developed based on rheological and elastoplastic theories, and parameter identification methods were established. The theoretical curves showed strong agreement with the experimental data, accurately captured the complete creep behavior of carbonaceous mudstone, and demonstrated the model's validity for engineering applications.Results and Discussions1) Creep curves exhibited stepwise progression, with deformation increasing significantly under higher osmotic pressures and reduced maximum bearing capacities. Failure deviatoric stress decreased exponentially with rising osmotic pressure. At constant confining pressure, elevated osmotic pressure shortened the total creep duration before failure. 2) Volumetric strain dilation occurred earlier under 23 MPa osmotic pressures compared to 1 MPa, which indicated accelerated crack initiation and unstable creep progression. 3) Accelerated creep rates manifested as nonlinear increases driven by coupled osmotic-deviatoric stress effects on crack damage evolution, which reflected macroscopic fracture propagation. 4) Under 7 MPa confining pressure, osmotic pressure reduced radial crack control and yielded stochastic failure modes such as composite fractures at 3 MPa. At 14 MPa confining pressure, macroscopic failure patterns became homogenized across osmotic pressures due to enhanced crack confinement. 5) SEM analysis revealed that tensile and shear fractures dominated microscale failure. Osmotic pressure raised intergranular reorganization, including fracture, refinement, and sliding, to form stress-adaptive microstructures. 6) Model-experiment consistency validated the applicability of the model across creep stages, including decay/steady-state creep at low stress and acceleration at high stress, which confirmed its capacity to characterize hydro-mechanical coupling effects.Conclusions1) Osmotic pressure intensifies creep deformation and reduces long-term strength in carbonaceous mudstone, inducing failure under lower deviatoric stresses. 2) Under low confining pressures, osmotic pressure reduces radial crack confinement, increasing stochastic failure mo-des characterized by composite fracture patterns. 3) At the microscopic level, osmotic pressure alters fracture morphologies: tensile fractures exhibit scaly brittle surfaces and root-like patterns, whereas shear fractures display dimples and transgranular cracks. The microstructural fracture patterns correlate well with macroscopic failure modes. 4) The proposed nonlinear viscoelastic-plastic creep model effectively characterizes full-stage creep behavior under varying confining/osmotic pressures. The model parameters derived from laboratory tests produce theoretical curves that closely match the experimental data by considering confining pressure and osmotic pressure as variables.

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.

Accurate prediction of time‑dependent multi‑responses of surrounding rock during tunnel excavation is critical for engineering safety and rational support design. Traditional numerical simulations are accurate but computationally prohibitive for real‑time use, whereas purely data‑driven models lack physical consistency. To address these issues, a physics‑informed surrogate model integrating Kolmogorov-Arnold Networks (KAN) and Long Short-Term Memory (LSTM) networks is proposed. The model takes seven static mechanical parameters of the rock mass as input and predicts the temporal evolution of deformation, stress, and damage characteristic zones including loosened, plastic, and disturbed zones induced by excavation. To overcome the lack of unified identification criteria, a quantitative determination method based on thermodynamic internal variable theory is introduced. The loosened zone is delineated by plastic shear strain, plastic volumetric strain, and their depth‑correlated Pearson coefficient curves; the disturbed zone boundary is determined using the cumulative energy dissipation density and its depth‑based Pearson coefficient curve. A Physical Knowledge Module (PKM) is embedded, encoding deformation growth, stress relaxation, and damage propagation into a composite loss function that combines data loss and physical loss. The framework is applied to the SJLS tunnel project. Bayesian optimization and Dirichlet sampling are employed for hyperparameter tuning. Results show that the model achieves high prediction accuracy (R2 > 0.99), outperforming standalone KAN, LSTM, and other baselines. The PKM enhances both predictive accuracy and physical consistency. The proposed framework serves as an efficient tool for real‑time prediction of multiple time‑dependent responses and for mechanical parameter inversion based on multi‑source monitoring data. A physics-informed KAN-LSTM framework for predicting tunnel responses is proposed. A numerical criterion for determining surrounding rock damage extents is established. Evolutionary pattern of time-dependent responses of surrounding rock are considered.

The foundation of high arch dams is subjected to substantial hydro-static and structural loads, making reinforcement of the dam toe essential for ensuring overall stability. This study investigates a high arch dam project in Southwest China using three-dimensional nonlinear finite element analysis to evaluate the strengthening effect of the toe structure on the dam-foundation system. The numerical simulations comprehensively examine deformation, stress distribution, yield zone development, unbalanced forces, and the evolution of the plastic complementary energy norm under overload conditions. The results demonstrate that the toe structure significantly improves stress transfer at the dam-foundation interface, alleviates stress concentration in the toe region, and enhances global structural stability. Specifically, the reinforcement mechanism is characterized by reduced dam displacement, mitigated principal tensile stresses on the upstream face, and delayed initiation and coalescence of yield zones. The analysis confirms that the toe structure plays a critical role in stabilizing the dam-foundation system by improving stress redistribution, delaying failure development, and increasing the over-load safety factor by approximately 10

ABSTRACT: Large deformation of surrounding rock under high in-situ stress conditions represents a critical challenge in deep tunnel engineering. This research investigates the deformation behavior and energy evolution of weak shale rock subjected to high geo-stress during excavation. Based on creep constitutive model with internal variables, a series of numerical simulations for TBM tunneling were conducted and surrounding rock parameters were obtained by inversion analysis. The characteristics and coevolutionary mechanisms of stress, deformation, loosening circle, and energy in surrounding rock were systematically compared and analyzed. Results demonstrate that excavation-induced large deformation constitutes a global response characterized by high initial deformation rate, substantial cumulative magnitude, and extensive influence scope. The rapid pre-excavation accumulation and slow post-excavation release of elastic strain energy directly drives the surrounding rock's intense initial and continued deformation. Plastic complementary energy effectively characterizes system non-equilibrium, with its abrupt increase revealing significant stress concentration as the fundamental deformation driver. The subsequent fluctuation pattern - initial decrease followed by increase - reflects the competing effects of tunnel self-balancing (energy release) and excavation advancement (energy disturbance).

The probabilistic stability evolution analysis of reservoir bank slopes is a crucial aspect of risk assessment, with core challenges including the consideration of deformation mechanisms and accurate determination of mechanical parameters. In this study, a novel time-varying reliability analysis framework based on sequential Bayesian updating of mechanical parameters is proposed. The inverse parameters account for damage time-dependent behavior, incorporating water effect and a strain-driven softening-hardening process that depends on sliding states. The likelihood function is enhanced to simultaneously consider observation error, surrogate model prediction error, and model structural error, with the introduction of physical penalty. Exploration of the high-dimensional parameter space is achieved via the Hamiltonian Monte Carlo (HMC) method and the physics knowledge-based time-dependent deformation surrogate model. The time-varying reliability analysis of the slope is performed using the multi-grid method. Taking a reservoir bank slope as a case study, the sequential updating of 12 mechanical parameters is conducted based on deformation time series from 16 monitoring points, thereby validating the proposed framework. The results indicate that the proposed framework effectively captures the posterior distribution of mechanical parameters, with the case slope remaining in a critically stable state after overall sliding, showing a high failure probability. Introducing model structural error can reduce parameter compensation, and a reasonable sequential updating step size can improve inversion accuracy.

The regression analysis method is a crucial approach for analyzing the slope deformation. To overcome the limitations of conventional methods in capturing the abrupt slope deformation features, a novel regression framework with activation function is established. A reservoir bank slope located in the upper reaches of the Lancang River, Yunnan Province, China, is selected as the case study. The slope deformation characteristic is investigated based on GNSS (global navigation satellite system) monitoring data, especially the dominant factor contributing to deformation resulting from water impoundment. The regression analysis shows an excellent correlation (R=0.99) between regression and measured displacements, confirming that the proposed multifactor regression model with the activation function effectively captures the deformation characteristics of the slope. It is demonstrated that the sliding component, primarily driven by the third water impoundment, has the greatest impact on deformation. The landslide mechanism may involve progressive creep deformation due to increased pore water pressure and reduced shear strength at the slope toe during the first two impoundment stages, bringing the slope to a critical state. The third impoundment intensified deformation, triggering gravity-driven sliding. Continuous sliding recompacted the sliding surface, gradually restoring frictional strength and stabilizing the slope.

In this paper, we report the detection of the very-high-energy (VHE, 100 GeV < E < 100 TeV) and ultra-high-energy (UHE, E > 100 TeV) γ-ray emissions from the direction of the young star-forming region W43, observed by the Large High Altitude Air Shower Observation (LHAASO). The extended γ-ray source was detected with a significance of ∼16 σ by KM2A and ∼17 σ by WCDA, respectively. The angular extension of this γ-ray source is about 0.5 degrees, corresponding to a physical size of about 50 pc. We discuss the origin of the γ-ray emission and possible cosmic ray acceleration in the W43 region using multi-wavelength data. Our findings suggest that W43 is likely another young star cluster capable of accelerating cosmic rays (CRs) to at least several hundred TeV.

In rock tunneling, the interaction between excavation tools and rocks will result in tool wear or deterioration, with the abrasiveness of rocks playing a significant role in this process. To assess the abrasiveness of rocks, Cerchar abrasivity tests and rock drilling tests were conducted on ten rock blocks. Initially, the Cerchar abrasivity index (CAI) of rock samples was obtained based on Cerchar abrasivity tests. The transformational relations between the CAI and other hardness and strength indexes (weighted hardness, equivalent quartz content, uniaxial compressive strength, etc.) were established using the multivariable regression. Afterward, measurement-while-drilling (MWD) data were collected during rock drilling tests. The MWD data and CAI of the ten rocks were integrated into a dataset, from which the MWD-CAI model was developed using the random forest algorithm. The results show that rock abrasiveness is more closely associated with hardness, while the correlation with compressive strength appears less significant. The transformational relations enable to predict CAI values by rock hardness and strength. Moreover, the MWD-CAI model is capable of predicting CAI values using drilling data. The findings can provide a reference for the quick assessment of rock abrasiveness.

The diffuse Galactic gamma-ray emission is a very important tool used to study the propagation and interaction of cosmic rays in the Milky Way. In this Letter, we report the measurements of the diffuse emission from the Galactic plane-covering Galactic longitudes from 15° to 235° and latitudes from -5° to +5°, in an energy range of 1 to 25 TeV-made with the Water Cherenkov Detector Array (WCDA) of the Large High Altitude Air Shower Observatory. After the sky regions of known sources are masked, the diffuse emission is detected with 24.6σ and 9.1σ significance in the inner Galactic plane (15°<l<125°, |b|<5°) and outer Galactic plane (125°<l<235°, |b|<5°), respectively. The WCDA spectra in both regions can be well described by a power-law function, with spectral indices of -2.67±0.05_{stat} in the inner region and -2.83±0.19_{stat} in the outer region, respectively. Combined with the Square Kilometer Array (KM2A) measurements at higher energies, a clear softening of the spectrum is found in the inner region, with change of spectral indices by ∼0.5 at a break energy around 30 TeV. The fluxes of the diffuse emission are higher by a factor of 1.5-2.7 than the model prediction assuming local cosmic ray spectra and the gas column density, which are consistent with those measured by the KM2A. Along the Galactic longitude, the spatial distribution of the diffuse emission shows deviation from that of the gas column density. The spectral shape of the diffuse emission may vary in different longitude regions. The WCDA measurements bridge the gap between the low-energy measurements by space detectors and the ultra-high-energy observations by KM2A and other experiments. These results suggest that improved modeling of the wideband diffuse emission is required.

Accurately obtaining the mechanical parameters of surrounding rock is fundamental for the scientific support design and stability calculations of tunnels. To address the limitations in precision associated with traditional inversion methods, a mechanical parameter inversion framework based on deep learning and Bayesian optimization is proposed. By incorporating the time-dependent deformation characteristics and mechanical properties of rock masses, the physics-informed framework achieves efficient inversion of the mechanical parameters of surrounding rock. The framework employs a deep learning algorithm based on fully connected layers and a multi-layer GRU architecture to construct a surrogate model for tunnel deformation. A physical knowledge module related to tunnel deformation behavior is developed, and a composite loss function is encoded to guide the data-driven process by physical information. The elastic-viscoplastic creep constitutive model, which grounded in internal variable thermodynamics, is used to simulate the time-dependent evolution of surrounding rock deformation. Based on the surrogate model optimized through hyperparameter tuning, a complex nonlinear mapping relationship between mechanical parameters and time-dependent deformation of the surrounding rock is established. A Bayesian optimization algorithm incorporating physical constraints is employed to account for the physical interrelations among parameters. The optimal mechanical parameters of the surrounding rock are inversely obtained based on burial depth and measured time-dependent deformation data. The proposed framework is applied to the JLL tunnel project. Results demonstrate that the determination coefficients (R2) of all inversion outcomes exceed 0.8, with a maximum value of 0.99. Comparative analyses with baseline models and ablation experiments further confirm the superiority of the proposed framework over other models.

In tunnel construction, tunnel boring machine (TBM) tunnelling typically relies on manual experience with sub-optimal control parameters, which can easily lead to inefficiency and high costs. This study proposed an intelligent decision-making method for TBM tunnelling control parameters based on multi-objective optimization (MOO). First, the effective TBM operation dataset is obtained through data preprocessing of the Songhua River (YS) tunnel project in China. Next, the proposed method begins with developing machine learning models for predicting TBM tunnelling performance parameters (i.e. total thrust and cutterhead torque), rock mass classification, and hazard risks (i.e. tunnel collapse and shield jamming). Then, considering three optimal objectives, (i.e., penetration rate, rock-breaking energy consumption, and cutterhead hob wear), the MOO framework and corresponding mathematical expression are established. The Pareto optimal front is solved using DE-NSGA-II algorithm. Finally, the optimal control parameters (i.e., advance rate and cutterhead rotation speed) are obtained by the satisfactory solution determination criterion, which can balance construction safety and efficiency with satisfaction. Furthermore, the proposed method is validated through 50 cases of TBM tunnelling, showing promising potential of application.

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.