Underground salt caverns are critical for large-scale energy storage, yet their stability assessment often relies on time-consuming numerical simulations. This study proposes a hybrid ensemble approach to estimate the stability of energy storage salt caverns. A dataset of 419 samples was constructed, with15 key features were selected from cavern geometry, operational parameters, in-situ stress distribution, and core mechanical properties. The model outputs two indicators reflecting the stability of the salt cavern: the maximum displacement of the cavern wall and the reduction in cavern volume. A stacking model combining Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGB), and Random Forest (RF) was developed, with hyperparameter optimization via the Hiking Optimization Algorithm (HOA). The results show that the model achieves a coefficient of determination (R2) value greater than 0.99 for predicting maximum wall displacement and over 0.97 for volume reduction. The proposed method was validated using the Vlsekriterijumska Optimizacija i Kompromisno Resenje (VIKOR) analysis, which demonstrates significant advantages compared with five traditional models. Furthermore, a comparative engineering scenario analysis based on the Jintan salt mine validates the model's capability to rapidly assess stability and optimize operational schemes. This study confirms that the proposed data-driven framework can serve as an efficient surrogate for numerical modeling, offering valuable decision-making support for the future potential of salt cavern site selection, design, and operation.

Soundless chemical demolition agents (SCDA) are recognized as a safe and sustainable alternative for rock breakage, with the potential to reduce reliance on explosives in mining. However, their mechanical performance is considerably hampered in the extreme cold climate, with slow or no rock fracturing observed. To overcome this deficiency, two methods were developed in recent studies by 1) electric heating the SCDA borehole with high-temperature wire, and 2) increasing the SCDA mixing water temperature. Good results were obtained with both methods. This paper explores the merits of a hybrid approach combining the two methods to further enhance the SCDA efficiency in the extreme cold environment. The new method, termed accelerated rock breakage (ARB), is tested in the temperature-controlled chamber at ambient temperatures of-20 degrees C to-60 degrees C using 203.2 mm (8-inch) cubic granite rock samples. The results demonstrate that the ARB method is superior to previous methods. For instance, at-40 degrees C ambient temperature, increasing the mixing water temperature to 50 degrees C while applying 25 V to the wire reduces the time to first crack (TFC) and minimum demolition time (MDT) to only 0.7 and 1.1 hours, respectively. The paper reports on the results of a detailed parametric study of 8 rock samples.

Traditional methods for slope stability assessment often encounter limitations in prediction accuracy and adaptability when applied to complex geological conditions and nonlinear characteristics typical of mining areas. To address these challenges and enhance analytical efficiency, this study develops a comprehensive slope stability database comprising 897 cases. Based on this dataset, we proposed two intelligent prediction frameworks grounded in hybrid ensemble learning utilizing the hiking optimization algorithm to perform hyperparameter optimization for both light gradients boosting machine and extreme gradient boosting models. Considering the randomness of the optimization algorithm, a new base model pairing strategy is proposed. For each base learner, we constructed three optimized configurations resulting in six base models. The models are then integrated using stacking and voting strategies to form hybrid stacking and hybrid voting models. All models are evaluated using five-fold cross-validation. The hybrid ensemble models achieve an accuracy of 0.9222, both significantly outperforming base models. Moreover, this study also explores the optimal base model selection strategy and compares the applicability of the models across different slope types. The interpretability analysis using SHapley Additive exPlanations (SHAP) reveals that the internal friction angle (phi) and cohesion (C) are the most influential factors governing stability prediction across both hybrid frameworks. To support practical engineering applications, we developed a user-friendly graphical user interface, enhancing the operability and applicability of the models. Comparative analysis demonstrates that the proposed method significantly improves the accuracy of slope stability prediction, outperforming traditional machine learning approaches. Overall, this research achieves key advancements in both database construction and intelligent prediction methodology, offering theoretical foundations and technical pathways for slope stability assessment under complex geological conditions.

Rockburst is a complex problem that underground mines face around the world as they reach deposits at greater depths. Based on literature, it is well established that rockbursts can be caused by numerous factors, most notably geological features, mechanical rock properties, induced seismicity, and mining parameters. In this research, the Young Davidson (YD) Mine in Northern Ontario, Canada, is used as the case study. The mine experiences seismic events at relatively shallow depths of less than 1 km. In the current study, rock burstability is examined through a comprehensive rock testing program of the different YD Mine lithologies. It is found that rock burstability potential is low to moderate at best. A 3D mine-wide numerical model was then built, and the updated mechanical rock properties, as well as the measured in-situ stress regime, were used. The stress analysis revealed that high stress concentration occurs in the ore pillars. Seismic events were primarily attributed to the following factors: the stress state as a result of the mining geometry, the nature of the sublevel stoping system, which leads to highly stressed secondary stopes, and the oblique orientation to the orebody strike of the in-situ stress tensor with an exceptionally high horizontal to vertical stress in-situ stress ratio. These findings support the hypothesis that mining geometry can be an independent cause for mining-induced seismicity regardless of rock properties.

Accurate prediction of retaining-pile displacement is important for deformation control during staged deep-excavation construction, where monitoring series often exhibit pronounced nonstationarity and multiscale temporal variation. To account for the distinct temporal characteristics of the trend and fluctuation components, this study proposes a CEEMDAN–SVR–PSO-LSTM hybrid framework, termed CSPL. Complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is first used to decompose the monitored displacement series into a slowly varying residual and oscillatory intrinsic mode functions (IMFs). Support vector regression (SVR) is employed to predict the trend component, whereas particle swarm optimization (PSO)-optimized long short-term memory (LSTM) is used to predict the fluctuation components. The component-wise predictions are then reconstructed to obtain the final displacement prediction. The proposed model is validated using monitoring data from two Zhengzhou Metro projects. For Case 1, the model achieves average R2, RMSE, and MAPE values of approximately 0.94, 0.33 mm, and 2.8%, respectively, across different monitoring depths, showing better overall predictive performance than BP, EMD-LSTM, and VMD-GRU. For Case 2, external validation using 90 depth-wise monitoring points over six construction stages yields millimeter-level errors, supporting stable predictive performance under different geological and support-system conditions. The results further indicate that the late construction stages and the upper pile segments deserve particular attention in deformation control. The proposed model provides a data-driven tool for construction-stage displacement prediction and deformation control in underground geotechnical engineering.

Accurate determination of in situ stress is fundamental for the safe and efficient design of underground construction projects such as tunnels, caverns, and deep mining excavations. Conventional techniques—particularly overcoring and hydraulic fracturing—have been widely adopted for decades, but their practical use is often constrained by high operational cost, rigorous field requirements, and logistical limitations at depth. As engineering projects advance into deeper and more complex geological environments, these constraints have prompted growing interest in laboratory-based, core-derived stress measurement approaches. Such methods utilize the stress-relief deformation that occurs when drill cores are extracted, enabling stress estimation without extensive downhole instrumentation. This paper presents a critical review of core-based stress measurement techniques based on a structured survey of peer-reviewed literature retrieved from major scientific databases (Web of Science, Scopus, and Google Scholar), covering studies published from the 1960s to 2025. The review examines Anelastic Strain Recovery (ASR), Differential Strain Curve Analysis (DSCA), Deformation Rate Analysis (DRA), acoustic-emission-based Kaiser effect approaches, and the emerging Diametrical Core Deformation Technique (DCDT). Recent studies show that DCDT, which measures instantaneous elastic diametrical deformation of cores, provides a more direct and physically transparent link to differential in situ stress, with reduced sensitivity to time-dependent effects. The DCDT, based on precise measurement of instantaneous elastic deformation upon coring, offers high-resolution stress estimation with minimal disruption to field operations. Its compatibility with optical scanning, laser micrometers, and CT imaging highlights its potential as a practical alternative to conventional techniques. A comparative synthesis of assumptions, accuracy, and applicability is provided, and key limitations and future research needs of core-based stress measurement methods are identified. The findings of this review provide practical guidance for selecting stress measurement techniques and support the application of core-based methods, particularly DCDT, in deep underground engineering, where cost-effective and reliable stress characterization is required.

Rock fragmentation in hard rock mining has traditionally relied on explosives which raises significant environmental and safety concerns for both workers and local communities. In response, soundless chemical demolition agents (SCDAs) primarily composed of lime (CaO), offer safer and more sustainable alternative to traditional blasting due to their soundless, vibrationless, and fumeless properties. This study is focused on examining the effect of mixing water temperature on the rock breakage performance of two commercially available SCDA brands, namely Betonamit type R (BT-R) and Dexpan type 3 (DXP-3). The experiments were conducted using 15 cm cubic granite rock specimens as the host material under various ambient temperatures. The study revealed an increase in mixing water temperature significantly accelerates the mechanical performance of SCDAs, particularly in cold ambient conditions. For instance, increasing the mixing water temperature from 20°C to 40°C reduced the time to first crack (TFC) by 36 % for BT-R and 74 % for DXP-3 under an ambient temperature of 0°C. A corresponding reduction in minimum demolition time (MDT) was also observed. At higher ambient temperatures, the impact of mixing water temperature was found to be less pronounced for both SCDA types. It is concluded that high mixing water temperature would be highly recommended for cold climate applications in both open pit and underground mining.

This review paper explores the mechanisms and factors influencing the corrosion of rockbolts, particularly in aggressive underground mining environments characterized by high humidity, fluctuating temperatures, and corrosive chemical agents. The study also highlights the role of chloride ions and sulfide inclusions in initiating and propagating corrosion. Additionally, the review explores the emergence of fiber-reinforced polymer (FRP) rockbolts as a viable alternative, offering insights into their non-corrosive properties, lightweight nature, and high tensile strength. The comparative analysis between traditional steel rockbolts and FRP alternatives reinforces the potential for FRP rockbolts to enhance the longevity and safety of mining support structures, while also considering the practical implications of their implementation in the mining industry.

The stability of underground open stopes is vital for safe mining, and accurate prediction is essential. This study developed a stability prediction framework integrating hybrid stacking ensemble modeling with the Optuna optimization algorithm to enhance traditional stability graphs. The workflow begins with revising the dataset compiled by Mawdesley. The revised dataset was then divided into training and testing subsets, followed by standardized preprocessing applied to both subsets. To reduce overfitting, a five-fold cross-validation strategy was systematically incorporated with the Optuna algorithm for hyperparameter optimization across ten machine learning algorithms. Three global indicators (accuracy, Kappa coefficient, and area under the curve (AUC)) and three local indicators (precision, recall, and F1 score) were employed for performance evaluation. Following performance ranking, the top three algorithms were selected as base learners, while a ridge regression classifier was designated as the meta learner for complementary characteristics and efficiency. The constructed stacking model was subsequently optimized. The optimized stacking ensemble model achieved an accuracy of 0.8689, precision of 0.8542, recall of 0.7519, F1 score of 0.7857, Kappa coefficient of 0.5760, and an AUC value of 0.8944 on the test set. Subsequent application of this model facilitated the generation of an updated stability graph, demonstrating marked improvements over traditional empirical approaches in both prediction accuracy and resistance to subjective bias. Furthermore, a comparative analysis revealed Optuna’s superior efficiency and effectiveness relative to Bayesian optimization, simulated annealing, and random search methodologies. This study provides a more reliable and accurate tool for open stope stability prediction.

Mechanized rock excavation is a commonly used method in underground projects in mining and tunneling. The performance of mechanized rock breaking depends on the rock's cuttability, which is traditionally analyzed through extensive field experiments, significantly affecting work efficiency. Therefore, establishing a method to quickly and accurately assess the rock mass characteristics and cuttability based on drilling, tunneling, and mining parameters is of great importance in practice. In this study, three databases were developed for drilling parameters, tunneling parameters, and mining parameters from field and experimental measurements. The hiking optimization algorithm (HOA) and Ivy algorithm (IVYA) were introduced to optimize the gradient boosting decision tree (GBDT) model. For each database, two hybrid ensemble models were developed to predict uniaxial compressive strength (UCS) and identifying cuttability levels. The training and testing dataset division ratio for all models is 4:1, with cross-validation applied to prevent overfitting. The results indicate that the developed models can be directly applied to real-time analysis of rock strength characteristics and cuttability based on drilling, tunneling, and mining parameters, facilitating the real-time adjustment of cutting equipment operation parameters and improving rock-breaking efficiency. Finally, a user-friendly graphical user interface (GUI) was developed for easy use by non-algorithm operators on site.

Concrete carbonation is a key factor leading to the instability of concrete structures. As the carbonation process progresses, the strength and durability of concrete gradually decrease, resulting in cracking and corrosion issues. Accurately predicting the carbonation depth is crucial for ensuring structural safety and extending service life. This study investigates the carbonation-induced deterioration of fly ash concrete (FAC), which is critical for structural durability. A comprehensive dataset of 883 carbonation instances was established, and a hybrid model integrating gradient boosting decision tree (GBDT) with an arithmetic optimization algorithm (AOA) was developed. The model's performance was evaluated against five classical models using five-fold cross-validation and four performance metrics: mean absolute error (MAE), coefficient of determination (R-2), root mean square error (RMSE), and variance accounted for (VAF). The AOA-GBDT model demonstrated superior predictive accuracy, achieving an R-2 of 0.9570, MAE of 2.1893, RMSE of 10.3105, and VAF of 95.7496. The technique for order of preference by similarity to ideal solution (TOPSIS) ranking further confirmed its outperformance over GBDT, KNN, BPNN, SVR, and LR. Shapley additive explanations (SHAP) analysis revealed that exposure time and fly ash content (FA) significantly influence carbonation depth. Feature analysis suggests that a 40 % FA replacement ratio optimally balances durability and economic efficiency. Additionally, a user-friendly prediction tool was developed to facilitate structural maintenance. Based on these findings, an economic circular model for fly ash utilization is proposed, integrating FAC into a closed-loop system that enhances sustainability by linking power generation, construction, and resource recycling. This study advances carbonation prediction methodologies and contributes to the sustainable use of fly ash in the built environment.

Rock formations naturally contain intricate internal fractures due to various environmental factors. Such fractures result in significant weakening of the mechanical properties of the rock mass. As factures have different geometric features and fill material characteristics, it is difficult to replicate their complex behavior in the laboratory. This poses a serious limitation on the experimental investigation of the mechanical properties of fractured rocks. Sand powder 3D printing (3DP) can overcome the limitations of casting methods in preparing samples with complex fractures and thus is widely applied in soft rock mechanics experiments. This paper utilizes Computer Tomography (CT) scanning to obtain the probability distribution patterns of fractures in fractured rock samples. Additionally, it combines sand powder 3DP technology to generate soft rock-like samples with internal networks of filled fractures. Uniaxial compression experiments employing digital image correlation (DIC) and acoustic emission (AE) techniques are used to investigate the mechanical properties, deformation characteristics, and fracture evolution patterns of samples with different fracture densities. By increasing the fracture density, the peak strength of the soft rock-like samples exponentially decreases, and the deformation characteristics linearly decrease. Crack propagation paths mostly follow the prefabricated fracture trajectories and loading direction. An RA-AF analysis suggests that the failure mode of the soft rock-like samples transitions from diagonal shear failure to block-shaped shear failure with increasing fracture density. These research findings represent a novel sand powder 3DP approach for studying the complex mechanics of complex fractured rocks.

Expansive cement (EC) is generally a slurry that it is poured into vertical holes for surface rock breakage applications. This paper describes the development of a novel cartridge for extending EC applications from gravity-filled vertical holes to horizontal, uptilted, and wet boreholes. Four cartridge prototypes were made from low-cost and readily available plastics using three-dimensional printers. The performance of each cartridge was investigated in unconfined rock slab tests. The polylactic acid (PLA) cartridge was found to be superior to the thermoplastic polyurethane, polyethylene terephthalate glycol, and acrylonitrile butadiene styrene cartridges. Through partial heat containment, the PLA cartridge accelerated the EC hydraulic reaction and shortened the onset of rock destruction by 30% relative to vertical, gravity-filled EC. Finally, rock breakage with EC was demonstrated in an underground mine using PLA cartridges. This novel type of cartridge could not only suit various applications beyond the scope of the current EC surface applications but also significantly improve the rock fracturing efficiency of EC. Le ciment expansif (EC) est g & eacute;n & eacute;ralement une boue que l'on verse dans des trous verticaux pour briser la roche en surface. Cet article d & eacute;crit le d & eacute;veloppement d'une nouvelle cartouche permettant d'& eacute;tendre les applications du ciment expansif des trous verticaux remplis par gravit & eacute; aux trous de forage horizontaux, inclin & eacute;s et humides. Quatre prototypes de cartouches ont & eacute;t & eacute; fabriqu & eacute;s & agrave; l'aide d'imprimantes tridimensionnelles & agrave; partir de plastiques peu co & ucirc;teux et facilement disponibles. Les performances de chaque cartouche ont & eacute;t & eacute; & eacute;tudi & eacute;es lors d'essais en dalles rocheuses non confin & eacute;es.La cartouche en acide polylactique (PLA, de l'anglais polylactic acid) s'est av & eacute;r & eacute;e sup & eacute;rieure aux cartouches en polyur & eacute;thane thermoplastique, en poly & eacute;thyl & eacute;ne t & eacute;r & eacute;phtalate glycol et en acrylonitrile butadi & eacute;ne styr & eacute;ne. Gr & acirc;ce au confinement partiel de la chaleur, la cartouche PLA a acc & eacute;l & eacute;r & eacute; la r & eacute;action hydraulique de l'EC et a raccourci le d & eacute;but de la destruction de la roche de 30 % par rapport & agrave; l'EC verticale remplie par gravit & eacute;. Enfin, la rupture de la roche avec l'EC a & eacute;t & eacute; d & eacute;montr & eacute;e dans une mine souterraine & agrave; l'aide de cartouches en PLA. Ce nouveau type de cartouche pourrait non seulement convenir & agrave; diverses applications d & eacute;passant le cadre des applications actuelles de l'EC en surface, mais aussi am & eacute;liorer de mani & eacute;re significative l'efficacit & eacute; de la fracturation de la roche par l'EC.

The state of stress in a mining front constantly changes with mining activities. In a recent study, the authors developed and verified with laboratory measurements an analytical model for the calculation of mining-induced stresses based on measuring the deformations of a diamond drill rock core extracted perpendicular to the mining front or face. The method is called diametrical core deformation technique (DCDT). In this study, the DCDT is used in combination with 3D numerical modelling to develop a practical methodology for the assessment of mining face stability. To demonstrate the methodology, a diamond drill rock core was retrieved from an access drift face 530 m below the surface of an underground mine in northern Quebec. The state of stress in the mining front is estimated from the DCDT and used to adjust the orientation and principal stress magnitudes in the local area around the access drift in a 3D linear-elastic numerical model using an iterative approach. As the rock core is partially fractured due to previous face advance blasting, the numerical model is further adjusted to model the observed damage zone. The 3D model after adjustments is used to examine the mining front stability with the Hoek–Brown failure criterion. It is postulated that the proposed methodology is suitable for the stability assessment of any mining front with or without an observed damage zone.

Interest in explosive-free rock fracturing has grown exponentially in the past two decades due to concerns about the environmental impacts of traditional rock fracturing methods with explosive energy. The present paper essentially presents the findings of an experimental study that aimed to develop a novel method for rock fracturing with soundless chemical degradation agents (SCDAs) at cold ambient temperatures. This is of great significance because commercially available SCDAs do not perform well, or at all, in these conditions. This study is part of a multi-phase project under the umbrella of Canada's Clean Growth Program. The newly established approach is verified through the rock fracture test results with SCDAs of two series of concrete and granite blocks exposed to cold temperatures up to −20°C. This method, the so-called high-temperature wire method (HTWM), utilizes a high-temperature spiral wire inserted into the SCDAs borehole and subjected to a DC voltage. The heat flux generated by the wire helps the SCDAs cure and expand to fracture the block. Based on the experimentally obtained results, it has been shown that the proposed new HTWM is quite promising because it enables rock fracturing with SCDAs at cold ambient temperatures reaching −20°C. It should be emphasized that such a result was previously unattainable with current industry practice.

Explosive-free rock breakage methods have been the subject of increasing research in the past three decades as they are considered more environmentally friendly than rock fragmentation with explosive energy. This paper summarizes some of the research findings of a 4-year project on rock breakage with expansive cement. Experimental and numerical studies for the estimation of peak expansive pressure in thick-walled steel cylinders were conducted and validated with direct pressure measurement. A series of rock slabs from Stanstead granite was then tested with a central hole injected with expansive cement. Another series of slabs was subjected to uniaxial pressure of 5 MPa with and without relief holes around the expansive cement hole. The results are compared with those obtained from unloaded rock slabs. Finally, the results of a field experiment involving the slashing of an intersection at a Canadian underground are presented and recommendations for practical applications are discussed.

The electromagnetic radiation (EMR) monitoring and early warning technology has experienced decades of successful applications for worldwide coal and rock dynamic disasters, yet a fundamental model unifying physical mechanism and generation process for EMR is still lacking. The effective revealing of EMR’s mechanism is crucial for dynamic disaster control and management. With this motive, a multi-scale experimental study was conducted in the earlier stage. At the micro-scale, the charge’s existence and non-uniform distribution on rock’s micro-surface were confirmed by atomic force microscope (AFM), and deduced the relationship with load changes. At the meso-scale, the time sequence synchronization and frequency domain consistency of EMR and micro-vibration (MV) in the rock fracture under load have been confirmed. Therefore, it is inferred that the vibration of the crack surface acts as the power source of rock fracture-induced EMR, and the original charge on the crack surface and the charge generated by the new crack surface are the electrical basis of EMR. Based on the above two experimental findings, this paper proposes a new mechanism of rock fracture-induced EMR defined as the electricity-vibration coupling mechanism, stating that, the vibrating charged crack generates the EMR. Subsequently, a generation model was constructed based on vibrating charged crack clusters to elucidate this mechanism. The experimental results demonstrated that the EMR waveform calculated by the model and measured by antenna exhibited good correspondence, thereby verifying the effectiveness of the constructed EMR model. The proposal of this new mechanism and the model further clarified the EMR’s mechanism induced by rock fracture. Moreover, the inter-relationship among crack propagation, vibration, and EMR was developed by this model, which could be immensely beneficial in EMR-based identification and prediction of dynamic disasters in complex mining environments worldwide.