Xuebin Xie

Rock cohesion (c ) and angle of internal fricti on (φ ) are core parameters for rock mass stability analysis and engineering design; however, traditional triaxial tests are costly and time-consuming, limiting engineering efficiency.Based on a total of 199 sets of measured data from four rock types (shale, limestone, quartzite, and quartz-mica schist) in the Himalayan region, this study uses P-wave velocity (V_p ), density (ρ ), uniaxial compressive strength (UCS), and tensile strength (TS) as input variables. It employs four models—Support Vector Regression (SVR), Random Forest (RF), multilayer perceptron (MLP), and extreme gradient boosting (XGBoost) to predictc andφ . Hyperparameters were tuned using grid search and Bayesian optimization. We compared unified modeling with rock-type-specific modeling, performed interpretability analysis using SHAP, and tested robustness by introducing Gaussian noise. The results show that XGBoost produced the best predictions atc (test setR^2 = 0.9901, RMSE = 0.512 MPa), while the Bayesian-optimized SVR model yielded the best results atφ (R^2 = 0.9776, RMSE = 0.744°). Rock-type-specific modeling improved theR^2 for limestone atφ by 0.3541; the SHAP contribution for UCS and TS exceeded 70%; Random Forest demonstrated the best noise resistance, with a decrease inR^2 of less than 0.04 under 10% noise. In summary, the strategy proposed in this paper allows for the selection of prediction schemes based on data quality and lithological differences, providing a feasible approach for rapidly obtaining rock strength parameters.

This study systematically optimized the geometric parameters of three typical cross-sections for curved-wall highway tunnel inner contours. Aiming to minimize the net excavation area, a unified genetic algorithm-based optimization framework was established for systematic comparison of four typical curved-wall section types and implemented on the Matlab(R2023b) platform, incorporating encoding, selection, crossover, and mutation operations for global optimization of geometric parameters across different section types. The optimized sections were further validated for structural performance using Midas GTS NX. Results show that the proposed multi-type optimization framework effectively reduced tunnel excavation areas across all section types, with mean optimization rates of 2.60% ± 0.21%, 2.11% ± 0.03%, 4.70% ± 0.02%, and 2.54% ± 0.02% (95% CI) achieved for single-circle, triple-circle Model-1, triple-circle Model-2, and five-circle sections, respectively, providing quantitative evidence for section-type selection in highway tunnel design. In terms of structural performance, the optimized sections demonstrated favorable axial force and bending moment characteristics. The findings provide a quantitative basis combining economic efficiency and structural rationality for tunnel section design, offering significant engineering application value and technical support for standardized and refined highway tunnel design in China.

Two types of rock specimens simulating grout curtains were prepared: one featuring solely an axial through-hole and the other featuring combined axial and transverse holes (termed combined-hole specimens). Coupled permeability-damage tests under cyclic loading-unloading were conducted to analyze the effects of peak stress level, confining pressure, seepage pressure difference, and defect geometry (curtain filling configuration) on permeability evolution and damage accumulation. The results indicate that: (1) Peak stress level is the primary factor governing damage development rate; (2) Higher confining pressure mitigates damage in the surrounding rock and decelerates permeability increase; (3) Larger seepage pressure difference (≥ 0.5 MPa) significantly enhances the stability and accuracy of permeability measurements; (4) For combined-hole configurations, increasing the transverse hole area (and thus the grout volume) provides more lateral seepage paths, effectively reducing axial permeability by up to two orders of magnitude. This study provides an experimental framework for assessing the long-term sealing performance of grouted curtains under repeated dynamic loading akin to blasting disturbances.

Uniaxial compression testing provides essential mechanical property characterization for intact rock specimens. The accuracy of specimen preparation critically affects compression test results through end-surface geometry deviations: parallelism, perpendicularity, and diameter tolerance. Specimen end-surface parallelism is affected by surface irregularities (e.g., protrusions, warping), whereas perpendicularity deviations indicate angular misalignment of the specimen with the loading axis. This study develops a 3D uniaxial compression model using RFPA3D, with rigid loading plates to simulate realistic boundary conditions. Three typical end-surface defects are modeled: protrusions (central/eccentric), grooves, and unilateral warping. Specimens with varying tilt angles are generated to evaluate perpendicularity deviations. Simulation results reveal that central end-surface protrusions induce: (1) localized stress concentration, which forms a dense core, and (2) pronounced wedging failure when protrusion height exceeds critical thresholds. Eccentric protrusions trigger characteristic shear failure modes, while unilateral warping causes localized failure through stress concentration at the deformed region. Importantly, end-surface grooves substantially alter stress distributions, generating bilateral stress concentration zones when groove width exceeds critical dimensions.

Rockburst is a major disaster in deep underground engineering, and its prediction is crucial for engineering safety. This study proposes an optimization method based on multidimensional feature selection and integrated learning that systematically evaluates the impact of different indicator dimensions by constructing an indicator–indicator system and an indicator–rockburst hierarchy using a combination of seven-, six-, five-, four-, and three-dimensional indicators in conjunction with six machine-learning models, such as XGBoost, LightGBM, and CatBoost. The results show that tree models (e.g., CatBoost, LightGBM, etc.) are naturally resistant to multicollinearity, and PCA preprocessing destroys their nonlinear feature relationships, leading to performance degradation. CatBoost has the best performance and strong overfitting resistance; LightGBM is the second most efficient and suitable for real-time applications. The indicator–indicator system has better overall performance but less stability, and the indicator–rockburst system has slightly lower performance but a more stable downward trend. The six-dimensional system in both types of systems can balance the performance and complexity and is the optimal choice for engineering applications. This study provides theoretical support and practical reference for the selection of rockburst prediction and an evaluation index system.

Currently, the central task of predicting rock fragmentation is becoming increasingly important in the field of rock mechanics and engineering blasting. This direction has been shown to be crucial to ensure the safety and durability of construction projects. In this study, a BP neural network is constructed to optimize the network weights and bias with the help of CPO algorithm, and its practicality and reliability are tested through the case of an iron ore mine in Hunan Province, China. The model is trained and tested using typical blasting data, and the results show that it performs efficiently, with a short prediction time and a high level of confidence. The predicted values were consistent with actual engineering measurements, achieving an RMSE of only 0.015813, which indicates strong potential for guiding practical blasting block size predictions.

To explore the influence of the arching effect on stress distribution in jointed backfill structures, this study employs three-dimensional numerical modeling to systematically analyze the mechanical behavior of backfill materials. A finite-difference approach was adopted to establish a representative stope model incorporating interface elements to simulate rock–backfill interactions. The methodology involved parametric studies examining key material properties (internal friction angle, cohesion, elastic modulus, Poisson’s ratio) and geometric configurations, with boundary conditions derived from typical mining scenarios. The results demonstrate that stress distribution follows nonlinear relationships with all investigated parameters. Increasing the internal friction angle and the cohesion reduce internal stresses, though the arch effect exhibits a distinct upper limit. Mechanical properties significantly influence stress transfer characteristics, with the elastic modulus governing stiffness response and the Poisson’s ratio affecting lateral stress development. Geometric parameters control the spatial extent of arching, with larger dimensions modifying the stress redistribution pattern. This research quantitatively establishes the operational limits of arching in backfill structures, providing crucial thresholds to prevent stability risks from overestimating its benefits. The findings offer practical guidelines for optimizing backfill design in deep mining and paste filling applications, contributing both technical solutions for mine safety and fundamental insights for geomechanical theory. The developed methodology serves as a robust framework for future studies on complex backfill behavior under various loading conditions.

Early warning for rock blasting is crucial for ensuring the safety of deep underground engineering. Existing methods primarily focus on classifying rock blasting levels, which makes it difficult to provide timely warnings. This paper proposes a novel early warning framework for rock blasting based on time series prediction of acoustic emission (AE) parameters. Based on uniaxial rock tests, ten AE parameters (rise time, ring count, energy, duration, amplitude, average frequency, RMS voltage, average signal level, peak frequency, and initial frequency) are identified as potential indicators for rock blasting early warning. These ten parameters collectively affect the accuracy of AE monitoring. Factor analysis is employed to process the normalized AE data, simplifying the data structure and identifying common variables. Additionally, it is found that the BP neural network optimized by Particle Swarm Optimization (PSO) is more suitable for predicting the future evolution of these AE parameters. This makes it possible to establish a comprehensive multi-indicator early warning system. The proposed framework provides a new perspective for rock blasting early warning systems.

The increase in mining depth necessitates higher strength requirements for hard rock pillars, making mine pillar stability analysis crucial for pillar design and underground safety operations. To enhance the accuracy of predicting the stability state of mine pillars, a prediction model based on the subtraction-average-based optimizer (SABO) for hyperparameter optimization of the least-squares support vector machine (LSSVM) is proposed. First, by analyzing the redundancy of features in the mine pillar dataset and conducting feature selection, five parameter combinations were constructed to examine their effects on the performance of different models. Second, the SABO-LSSVM prediction model was compared vertically with classic models and horizontally with other optimized models to ensure comprehensive and objective evaluation. Finally, two data sampling methods and a combined sampling method were used to correct the bias of the optimized model for different categories of mine pillars. The results demonstrated that the SABO-LSSVM model exhibited good accuracy and comprehensive performance, thereby providing valuable insights for mine pillar stability prediction.