Rock fragmentation is an important indicator for assessing the quality of blasting operations. However, accurate prediction of rock fragmentation after blasting is challenging due to the complicated blasting parameters and rock properties. For this reason, optimized by the Bayesian optimization algorithm (BOA), four hybrid machine learning models, including random forest, adaptive boosting, gradient boosting, and extremely randomized trees, were developed in this study. A total of 102 data sets with seven input parameters (spacing-toburden ratio, hole depth-to-burden ratio, burden-to-hole diameter ratio, stemming length-to-burden ratio, powder factor, in situ block size, and elastic modulus) and one output parameter (rock fragment mean size, X₅₀) were adopted to train and validate the predictive models. The root mean square error (RMSE), the mean absolute error (MAE), and the coefficient of determination (R²) were used as the evaluation metrics. The evaluation results demonstrated that the hybrid models showed superior performance than the standalone models. The hybrid model consisting of gradient boosting and BOA (GBoost-BOA) achieved the best prediction results compared with the other hybrid models, with the highest R² value of 0.96 and the smallest values of RMSE and MAE of 0.03 and 0.02, respectively. Furthermore, sensitivity analysis was carried out to study the effects of input variables on rock fragmentation. In situ block size (X_B), elastic modulus (E), and stemming length-to-burden ratio (T/B) were set as the main influencing factors. The proposed hybrid model provided a reliable prediction result and thus could be considered an alternative approach for rock fragment prediction in mining engineering.

Assessing the stability of pillars in underground mines (especially in deep underground mines) is a critical concern during both the design and the operational phases of a project. This study mainly focuses on developing two practical models to predict pillar stability status. For this purpose, two robust models were developed using a database including 236 case histories from seven underground hard rock mines, based on gene expression programming (GEP) and decision tree-support vector machine (DT-SVM) hybrid algorithms. The performance of the developed models was evaluated based on four common statistical criteria (sensitivity, specificity, Matthews correlation coefficient, and accuracy), receiver operating characteristic (ROC) curve, and testing data sets. The results showed that the GEP and DT-SVM models performed exceptionally well in assessing pillar stability, showing a high level of accuracy. The DT-SVM model, in particular, outperformed the GEP model (accuracy of 0.914, sensitivity of 0.842, specificity of 0.929, Matthews correlation coefficient of 0.767, and area under the ROC of 0.897 for the test data set). Furthermore, upon comparing the developed models with the previous ones, it was revealed that both models can effectively determine the condition of pillar stability with low uncertainty and acceptable accuracy. This suggests that these models could serve as dependable tools for project managers, aiding in the evaluation of pillar stability during the design and operational phases of mining projects, despite the inherent challenges in this domain.

Monitoring of the mechanical behavior of underwater shield tunnels is vital for ensuring their long-term structural stability. Typically determined by empirical or semi-empirical methods, the limited number of monitoring points and coarse monitoring schemes pose huge challenges in terms of capturing the complete mechanical state of the entire structure. Therefore, with the aim of optimizing the monitoring scheme, this study introduces a spatial deduction model for the stress distribution of the overall structure using a machine learning algorithm. Initially, clustering experiments were performed on a numerical data set to determine the typical positions of structural mechanical responses. Subsequently, supervised learning methods were applied to derive the data information across the entire surface by using the data from these typical positions, which allows flexibility in the number and combinations of these points. According to the evaluation results of the model under various conditions, the optimized number of monitoring points and their locations are determined. Experimental findings suggest that an excessive number of monitoring points results in information redundancy, thus diminishing the deduction capability. The primary positions for monitoring points are determined as the spandrel and hance of the tunnel structure, with the arch crown and inch arch serving as additional positions to enhance the monitoring network. Compared with common methods, the proposed model shows significantly improved characterization abilities, establishing its reliability for optimizing the monitoring scheme.

In this study, the design and development of a sensor made of low-cost parts to monitor inclination and acceleration are presented. Α micro electro-mechanical systems, micro electro mechanical systems, sensor was housed in a robust enclosure and interfaced with a Raspberry Pi microcomputer with Internet connectivity into a proposed tilt and acceleration monitoring node. Online capabilities accessible by mobile phone such as real-time graph, early warning notification, and database logging were implemented using Python programming. The sensor response was calibrated for inherent bias and errors, and then tested thoroughly in the laboratory under static and dynamic loading conditions beside high-quality transducers. Satisfactory accuracy was achieved in real time using the Complementary Filter method, and it was further improved in LabVIEW using Kalman Filters with parameter tuning. A sensor interface with LabVIEW and a 600 MHz CPU microcontroller allowed real-time implementation of high-speed embedded filters, further optimizing sensor results. Kalman and embedded filtering results show agreement for the sensor, followed closely by the low-complexity complementary filter applied in real time. The sensor's dynamic response was also verified by shaking table tests, simulating past recorded seismic excitations or artificial vibrations, indicating negligible effect of external acceleration on measured tilt; sensor measurements were benchmarked using high-quality tilt and acceleration measuring transducers. A preliminary field evaluation shows robustness of the sensor to harsh weather conditions.

To guarantee safe and efficient tunneling of a tunnel boring machine (TBM), rapid and accurate judgment of the rock mass condition is essential. Based on fuzzy C-means clustering, this paper proposes a grouped machine learning method for predicting rock mass parameters. An elaborate data set on field rock mass is collected, which also matches field TBM tunneling. Meanwhile, target stratum samples are divided into several clusters by fuzzy C-means clustering, and multiple submodels are trained by samples in different clusters with the input of pretreated TBM tunneling data and the output of rock mass parameter data. Each testing sample or newly encountered tunneling condition can be predicted by multiple submodels with the weight of the membership degree of the sample to each cluster. The proposed method has been realized by 100 training samples and verified by 30 testing samples collected from the C1 part of the Pearl Delta water resources allocation project. The average percentage error of uniaxial compressive strength and joint frequency (J_f) of the 30 testing samples predicted by the pure back propagation (BP) neural network is 13.62% and 12.38%, while that predicted by the BP neural network combined with fuzzy C-means is 7.66% and 6.40%, respectively. In addition, by combining fuzzy C-means clustering, the prediction accuracies of support vector regression and random forest are also improved to different degrees, which demonstrates that fuzzy C-means clustering is helpful for improving the prediction accuracy of machine learning and thus has good applicability. Accordingly, the proposed method is valuable for predicting rock mass parameters during TBM tunneling.

It is crucial to predict future mechanical behaviors for the prevention of structural disasters. Especially for underground construction, the structural mechanical behaviors are affected by multiple internal and external factors due to the complex conditions. Given that the existing models fail to take into account all the factors and accurate prediction of the multiple time series simultaneously is difficult using these models, this study proposed an improved prediction model through the autoencoder fused long- and short-term time-series network driven by the mass number of monitoring data. Then, the proposed model was formalized on multiple time series of strain monitoring data. Also, the discussion analysis with a classical baseline and an ablation experiment was conducted to verify the effectiveness of the prediction model. As the results indicate, the proposed model shows obvious superiority in predicting the future mechanical behaviors of structures. As a case study, the presented model was applied to the Nanjing Dinghuaimen tunnel to predict the stain variation on a different time scale in the future.

The extraction of coal measure gas has been shifted toward thin gas reservoirs due to the depletion of medium-thick gas reservoirs. The coproduction of coalbed gas, shale gas, and tight sandstone gas (called a multisuperposed gas system) is a key low-cost technology for the enhancement of natural gas production from thin gas reservoirs in coal measure. As an emerging engineering exploitation technology at its early stage of development, gas coproduction confronts various engineering challenges in hydraulic fracturing, bottom-hole pressure regulation, well network arrangement, and extraction sequence. The current understanding of the opportunities and challenges in the gas coproduction from the multisuperposed gas system is not comprehensive enough. In this case, the previous achievements in the field of gas coproduction should be urgently reviewed to provide valuable guidance and recommendations for further development. This review first discusses the regional and spatial distribution characteristics and possible reservoir combinations of gas reservoirs in coal measure. Then, the basic properties of different reservoirs, engineering challenges, and interlayer interference are comparatively analyzed and discussed. The current simulation models for gas coproduction and potential future research directions are further explored. The results indicate that the coupling effects of reservoir heterogeneity, interwell interference, and geological structure for increasing coproduction prediction accuracy should be included in future simulation models for gas coproduction. Careful investigation is required to explore the mechanisms and their further quantifications on the effects of interlayer interference in gas coproduction. The fractal dimension as a scale can play an important role in the characterization of the gas and water transport in different reservoirs. The machine learning methods have tremendous potential to provide accurate and fast predictions for gas coproduction and interlayer interference.

Oil and gas exploration studies have been increasingly moving deeper into the earth. The rocks in deep and ultra-deep reservoirs are exposed to a complex environment of high temperatures and large geo-stresses. The Tarim oilfield in the Xinjiang Uygur Autonomous Region (Xinjiang for short), China, has achieved a breakthrough in the exploration of deep hydrocarbon reservoirs at a depth of over 9000 m. The mechanical properties of deep rocks are significantly different from those of shallow rocks. In this study, triaxial compression tests were conducted on heat-treated carbonatite rocks to explore the evolution of the mechanical properties of carbonatite rocks under high confining pressure after thermal treatment. The rocks for the tests were collected from reservoirs in the Tarim oilfield, Xinjiang, China. The experiments were performed at confining pressures ranging from atmospheric to 120 MPa and temperatures ranging from 25 to 500℃. The results show that the critical confining pressure of the brittle-ductile transition increases with increasing temperature. Young's modulus is negatively correlated with the temperature and positively correlated with the confining pressure. As the confining pressure increases, the failure mode of the specimens gradually transforms from shear fracture failure into “V”-type failure and finally into bulging failure (multiple shear fractures). With increasing temperature, the failure angle tends to decrease. In addition, an improved version of the Mohr‒Coulomb strength criterion with a temperature-dependent power function was proposed to describe the failure strength of carbonatite rocks after exposure to high temperature and high confining pressure. The surface of the strength envelope of this criterion is temperature dependent, which could reflect the strength evolution of rock under high confining pressures after thermal treatment. Compared with other strength criteria, this criterion is more capable of replicating physical processes.

Chinese coal reservoirs are characterized by low pressure and low permeability, which need to be enhanced so as to increase production. However, conventional methods for permeability enhancement can only increase the permeability in fractures, but not the ultra-low permeability in coal matrices. Attempts to enhance such impermeable structures lead to rapid attenuation of gas production, especially in the late stage of gas extraction. Thermal stimulation by injecting high-temperature steam is a promising method to increase gas production. The critical scientific challenges that still hinder its widespread application are related to the evolution law of permeability of high-temperature steam in coal and the thermal deformation of coal. In this study, an experimental approach is developed to explore the high-temperature steam seepage coupled with the thermal deformation in coal under triaxial stress. The tests were conducted using cylindrical coal specimens of ϕ50 mm × 100 mm. The permeability and thermal strain in coal were investigated when high-temperature steam was injected at 151.11, 183.20, 213.65, and 239.76℃. The experimental results reveal for the first time that as the amount of injected fluid increases, the steam permeability shows periodic pulsation changes. This paper introduces and explains the main traits of this discovery that may shed more light on the seepage phenomenon. When the injected steam temperature increases, the amplitude of pulsating permeability decreases, whereas the frequency increases; meanwhile, the period becomes shorter, the pulsation peak appears earlier, and the stabilization time becomes longer. The average peak permeability shows a “U-shaped” trend, decreasing first and then increasing as the steam temperature increases. Meanwhile, with the extension of steam injection time, the axial, radial, and volumetric strains of coal show a stage-wise expansion characteristic at different temperatures of steam injection, except for the radial strains at 151.11℃. A two-phase flow theory of gas-liquid is adopted to elucidate the mechanism of pulsating seepage of steam. Moreover, the influencing mechanism of inward and outward thermal expansion on the permeability of coal is interpreted. The results presented in this paper provide new insight into the feasibility of thermal gas recovery by steam injection.

With recent technological advancements, tunnel boring machines (TBM) have developed and exhibited high performance in large diameters and weak ground conditions. Tunnels are crucial structures that significantly influence the timelines of highway and railway projects. Therefore, the construction of tunnels with TBMs becomes a preferred option. In this study, a comparative analysis between TBM and the New Austrian Tunneling Method (NATM) for tunnel construction is performed in the construction of the T1 tunnel with a diameter of 13 m, which is the longest tunnel in the Eşme-Salihli section of Ankara-İzmir High-Speed Railway Project (Türkiye). The selection of TBM type, measures taken in problematic sections, and application issues of TBM are discussed. The impact of correct description of geological and geotechnical conditions on both selection and performance of TBM is presented. An earth pressure balanced type TBM is chosen for the construction of the T1 tunnel. Because of the additional engineering measures taken before excavation in problematic areas, the tunnel was completed with great success within the initially planned timeframe. From this point of view, this study is an important case and may contribute to worldwide tunneling literature.

Shield tunnel, composed of several segments, is widely used in urban underground engineering. When the tunnel is under load, relative displacement occurs between adjacent segments. In the past, distributed optical fiber sensing technology was used to perform strain monitoring, but there is an urgent need to determine how to transform strain into displacement. In this study, optical frequency domain reflectometry was applied in laboratory tests. Aiming at the shear process and center settlement process of shield tunnel segments, two kinds of quantitative calculation methods were put forward to carry out a quantitative analysis. Meanwhile, the laboratory test process was simulated numerically utilizing the discrete element numerical analysis method. Optical fiber, an atypical geotechnical material, was innovatively applied for discrete element modeling and numerical simulation. The results show that the measured displacement of the dial gauge, the calculated results of the numerical model, and the displacement quantitatively calculated from the optical fiber data agree with each other in general. The latter two methods can potentially be utilized in engineering application of deformation monitoring at shield tunnel joints, but need to be further calibrated and adjusted in detail.

To study the energy evolution and failure characteristics of saturated sandstone under unloading conditions, rock unloading tests under different stress paths were conducted. The energy evolution mechanism of the unloading failure of saturated sandstone was systematically explored from the perspectives of the stress path, the initial confining pressure, and the energy conversion rate. The results show that (1) before the peak stress, the elastic energy increases with an increase in deviatoric stress, while the dissipated energy slowly increases first. After the peak stress, the elastic energy decreases with the decrease of deviatoric stress, and the dissipated energy suddenly increases. The energy release intensity during rock failure is positively correlated with the axial stress. (2) When the initial confining pressure is below a certain threshold, the stress path is the main factor influencing the total energy difference. When the axial stress remains constant and the confining pressure is unloading, the total energy is more sensitive to changes in the confining pressure. When the axial stress remains constant, the compressive deformation ability of the rock cannot be significantly improved by the increase in the initial confining pressure. The initial confining pressure is positively correlated with the rock's energy storage limit. (3) The initial confining pressure increases the energy conversion rate of the rock; the initial confining pressure is positively correlated with the energy conversion rate; and the energy conversion rate has a high confining pressure effect. The increase in the axial stress has a much greater impact on the elastic energy than the confining pressure. (4) When the deviatoric stress is small, the confining pressure mainly plays a protective role. Compared with the case of triaxial compression paths, the rock damage is more severe under unloading paths, and compared with the case of constant axial stress, the rock damage is more severe under increasing axial stress.