Journal of Rock Mechanics and Geotechnical Engineering最新文献

The spatiotemporal distributions of microbes in soil by different methods could affect the efficacy of the microbes to reduce the soil hydraulic conductivity. In this study, the specimens of bio-mediated sands were prepared using three different methods, i.e. injecting, mixing, and pouring a given microbial solution onto compacted sand specimens. The hydraulic conductivity was measured by constant-head tests, while any soil microstructural changes due to addition of the microbes were observed by scanning electron microscope (SEM) and mercury intrusion porosimetry (MIP) tests. The amount of dextran concentration produced by microbes in each type of specimen was quantified by a refractometer. Results show that dextran production increased exponentially after 5–7 d of microbial settling with the supply of culture medium. The injection and mixing methods resulted in a similar amount and uniform distribution of dextran in the specimens. The pouring method, however, produced a nonuniform distribution, with a higher concentration near the specimen surface. As the supply of culture medium discontinued, the dextran content near the surface produced by the pouring method decreased dramatically due to high competition for nutrients with foreign colonies. Average dextran concentration was negatively and correlated with hydraulic conductivity of bio-mediated soils exponentially, due to the clogging of large soil pores by dextran. The hydraulic conductivity of the injection and mixing cases did not change significantly when the supply of culture medium was absent.

The technology of tunnel boring machine (TBM) has been widely applied for underground construction worldwide; however, how to ensure the TBM tunneling process safe and efficient remains a major concern. Advance rate is a key parameter of TBM operation and reflects the TBM-ground interaction, for which a reliable prediction helps optimize the TBM performance. Here, we develop a hybrid neural network model, called Attention-ResNet-LSTM, for accurate prediction of the TBM advance rate. A database including geological properties and TBM operational parameters from the Yangtze River Natural Gas Pipeline Project is used to train and test this deep learning model. The evolutionary polynomial regression method is adopted to aid the selection of input parameters. The results of numerical experiments show that our Attention-ResNet-LSTM model outperforms other commonly-used intelligent models with a lower root mean square error and a lower mean absolute percentage error. Further, parametric analyses are conducted to explore the effects of the sequence length of historical data and the model architecture on the prediction accuracy. A correlation analysis between the input and output parameters is also implemented to provide guidance for adjusting relevant TBM operational parameters. The performance of our hybrid intelligent model is demonstrated in a case study of TBM tunneling through a complex ground with variable strata. Finally, data collected from the Baimang River Tunnel Project in Shenzhen of China are used to further test the generalization of our model. The results indicate that, compared to the conventional ResNet-LSTM model, our model has a better predictive capability for scenarios with unknown datasets due to its self-adaptive characteristic.

This study focuses on the analytical prediction of subsurface settlement induced by shield tunnelling in sandy cobble stratum considering the volumetric deformation modes of the soil above the tunnel crown. A series of numerical analyses is performed to examine the effects of cover depth ratio (C/D), tunnel volume loss rate (η_t) and volumetric block proportion (VBP) on the characteristics of subsurface settlement trough and soil volume loss. Considering the ground loss variation with depth, three modes are deduced from the volumetric deformation responses of the soil above the tunnel crown. Then, analytical solutions to predict subsurface settlement for each mode are presented using stochastic medium theory. The influences of C/D, η_t and VBP on the key parameters (i.e. B and N) in the analytical expressions are discussed to determine the fitting formulae of B and N. Finally, the proposed analytical solutions are validated by the comparisons with the results of model test and numerical simulation. Results show that the fitting formulae provide a convenient and reliable way to evaluate the key parameters. Besides, the analytical solutions are reasonable and available in predicting the subsurface settlement induced by shield tunnelling in sandy cobble stratum.

In order to evaluate the performance of deep geological disposal of radioactive waste, an underground research laboratory (URL) was constructed by Andra in the Callovo-Oxfordian (COx) claystone formation at the Meuse/Haute-Marne (MHM). The construction of URL induced the excavation damage of host formations, and the ventilation in the galleries desaturated the host formation close to the gallery wall. Moreover, it is expected that the mechanical behaviour of COx claystone is time-dependent. This study presents a constitutive model developed to describe the viscoplastic behaviour of unsaturated and damaged COx claystone. In this model, the unsaturation effect is considered by adopting the Bishop effective stress and the van Genuchten (VG) water retention model. In terms of the viscoplastic behaviour, the nonstationary flow surface (NSFS) theory for unsaturated soils is used with consideration of the coupled effects of strain rate and suction on the yield stress. A progressive hardening law is adopted. Meanwhile, a non-associated flow rule is used, which is similar to that in Barcelona basic model (BBM). In addition, to describe the damage effect induced by suction change and viscoplastic loading, a damage function is defined based on the crack volume proportion. This damage function contains two variables: unsaturated effective stress and viscoplastic volumetric strain, with the related parameters determined based on the mercury intrusion porosimetry (MIP) tests. For the model validation, different tests on COx claystone under different loading paths are simulated. Comparisons between experimental and simulated results indicated that the present model is able to well describe the viscoplastic behaviour of damaged COx claystone, including swelling/shrinkage, triaxial extension and compression, and triaxial creep.

Jet grouting is one of the most popular soil improvement techniques, but its design usually involves great uncertainties that can lead to economic cost overruns in construction projects. The high dispersion in the properties of the improved material leads to designers assuming a conservative, arbitrary and unjustified strength, which is even sometimes subjected to the results of the test fields. The present paper presents an approach for prediction of the uniaxial compressive strength (UCS) of jet grouting columns based on the analysis of several machine learning algorithms on a database of 854 results mainly collected from different research papers. The selected machine learning model (extremely randomized trees) relates the soil type and various parameters of the technique to the value of the compressive strength. Despite the complex mechanism that surrounds the jet grouting process, evidenced by the high dispersion and low correlation of the variables studied, the trained model allows to optimally predict the values of compressive strength with a significant improvement with respect to the existing works. Consequently, this work proposes for the first time a reliable and easily applicable approach for estimation of the compressive strength of jet grouting columns.

Lateral dispersion significantly directs the assessment of rockfall hazard and design of countermeasures. In the present study, the dependence of lateral dispersion on different controlling factors has been systematically evaluated by performing laboratory tests using three different rock block types, namely circular block, and two types of elliptical block. The three types of rock block are released onto an inclined surface with the identical initial status. Parallel, anti-parallel, and oblique impact tests set at slope angles of 22.5° and 45°are conducted to study the block-slope interaction of rockfall. Lateral dispersion of rockfall is less influenced by the block shape for the oblique impact, while the post-impact behaviors are greatly affected by the block shape. The key factors influencing the deviation of the post-impact trajectory direction are the slope angle (θ) and direction difference (Δφ). An empirical model is then developed to characterize the deviation distribution of lateral dispersion by 5th and 95th percentile values with the inclusion of the two key factors. Linear function can be used to describe the 5th percentile boundary, while hyperbolic function is good for the 95th percentile boundary, which need to be validated by field tests in the subsequent research.

Enzyme-induced carbonate precipitation (EICP) is an emerging technique to improve the soil and most studies are carried out at room temperature. However, considering some foundations are in high-temperature environments (>40 °C), the higher urease activity at high temperature results in the solidification inhomogeneity, limiting the application of EICP. The higher urease activity at high temperature hampers the application of EICP because of solidification inhomogeneity. The garlic extract has been used as a type of urease inhibitor in medical science and food engineering. Here, we propose to use it to control urease activity for sand solidification at high temperature. The effects of garlic extract addition on urease activity and precipitation rates for calcium carbonate (CaCO₃) were studied. Extra tests were conducted to study the effect of garlic extract addition on the solidification homogeneity. The results showed that the garlic extract addition significantly decreased urease activity. To reduce the rate of CaCO₃ precipitation at different temperatures, a suitable concentration of garlic extract was necessary to obtain a suitable urease activity. In the sand solidification test, garlic extract addition resulted in a smaller difference in sonic time values or CaCO₃ contents at different parts of samples. The improved solidification homogeneity can achieve higher strength. The correlation between sonic time values and CaCO₃ content was higher than that between CaCO₃ content and strength. Appropriate concentrations of garlic extract were obtained at 35 °C, 40 °C, 45 °C, 50 °C, and 55 °C. The proposed garlic extract addition method was significant to improve the homogeneity of solidified soil in practical engineering applications.

Lime-treatment of clayey soil significantly increases its shear and tensile strengths. Consequently, the tensile strength of lime-treated soils deserves careful investigation because it may provide an appreciable benefit for the stability of earth structures. This study investigates the tensile and shear strengths of an untreated and lime-treated (3% of lime) plastic clay at different curing times (7 d, 56 d and 300 d), through triaxial tension and compression tests. Triaxial tension tests are performed using “diabolo-shaped” soil samples with reduced central section, such that the central part of the specimen can be under axial tension while both end-sections remain in axial compression. Consolidated undrained (CU) conditions with measurement of pore water pressure allow analyzing the failure conditions through effective stress and total stress approaches. The results of triaxial tension tests reveal that the failure occurs under tensile mode at low confining pressure while extensional shear failure mode is observed under higher confining pressure. Consequently, a classical Mohr-Coulomb shear failure criterion must be combined with a cut-off tensile strength criterion that is not affected by the confining pressure. When comparing shear failure under compression and tension, a slight anisotropy is observed.

Unconventional resources like shale gas has been the focus of intense research and development for two decades. Apart from intrinsic geologic factors that control the gas shale productivity (e.g. organic matter content, bedding planes, natural fractures, porosity and stress regime among others), external factors like wellbore orientation and stimulation design play a role. In this study, we present a series of true triaxial hydraulic fracturing experiments conducted on Lushan shale to investigate the interplay of internal factors (bedding, natural fractures and in situ stress) and external factors (wellbore orientation) on the growth process of fracture networks in cubic specimens of 200 mm in length. We observe relatively low breakdown pressure and fracture propagation pressure as the wellbore orientation and/or the maximum in situ stress is subparallel to the shale bedding plane. The wellbore orientation has a more prominent effect on the breakdown pressure, but its effect is tapered with increasing angle of bedding inclination. The shale breakdown is followed by an abrupt response in sample displacement, which reflects the stimulated fracture volume. Based on fluid tracer analysis, the morphology of hydraulic fractures (HF) is divided into four categories. Among the categories, activation of bedding planes (bedding failure, BF) and natural fractures (NF) significantly increase bifurcation and fractured areas. Under the same stress regime, a horizontal wellbore is more favorable to enhance the complexity of hydraulic fracture networks. This is attributed to the relatively large surface area in contact with the bedding plane for the horizontal borehole compared to the case with a vertical wellbore. These findings provide important references for hydraulic fracturing design in shale reservoirs.