Geotechnical parameter estimation is critical to the design, performance, safety, and cost and schedule management in Tunnel Boring Machine projects. Since these parameters vary within a certain range, relying on mean values for evaluation introduces significant risks to the project. Due to the non-homogeneous characteristics of geological formation, data may not exhibit a normal distribution and the presence of outliers might be deceptive. Therefore, the use of reliable analyses and simulation models is inevitable in the course of the data evaluation process. Advanced modeling techniques enable comprehensive analysis of the project data and allowing to model the uncertainty in geotechnical parameters. This study involves using Monte Carlo Simulation method to predict probabilistic distributions of field data, and therefore, establish a basis for designs and in turn to minimize project risks. In the study, 166 sets of geotechnical data Obtained from 35 boreholes including Standard Penetration Test, Limit Pressure, Liquid Limit, and Plastic Limit values, which are mostly utilized parameters in estimating project requirements, were used to estimate the geotechnical data distribution of the study field. In this context, firstly, the data was subjected to multi-parameter linear regression and variance analysis. Then, the obtained equations were implemented into a Monte Carlo Simulation, and probabilistic distributions of the geotechnical data of the field were simulated and corresponding to the 90% probability range, along with the minimum and maximum values at the 5% probability levels presented. Accordingly, while the average SPT N30 value is 42.86, but the highest occurrence rate is 50.81. For Net Limit Pressure, the average field data is 17.07 kg/cm², with the maximum occurrence between 9.6 kg/cm² and 13.7 kg/cm². Similarly, the average Plastic Limit value is 22.32, while the most probable value is 20.6. The average Liquid Limit value is 56.73, with the highest probability at 54.48, as indicated in the statistical data distribution. Understanding the percentage distribution of data likely to be encountered in the project allows for accurate forecasting of both high and low probability scenarios, offering a significant advantage, particularly in ordering TBM requirements.

Studies pertaining to the seismic bearing capacity analysis of skirted footings using the pseudo-static approach for estimation of the earthquake force in association with finite element method have been presented in this paper. An attempt has been made to explain the behaviors of the skirted footings by means of failure patterns obtained for rigid and flexible skirts. The skirts enhance the seismic bearing to some extent with an increase in seismic loading, after which it decreases nonlinearly. The effectiveness of skirts increases initially to some extent with an increase in seismic loading, after which it decreases nonlinearly. Other parameters that inversely affect the effectiveness of skirts are the depth of footing and the internal friction angle of the soil. The detailed finite element analysis regarding the various failure patterns of skirted footings under seismic forces shows the failure mechanism changes from a general shear failure to local shear failure with an increase in seismic force. An opposite trend has been observed with the increase in the angle of internal friction of the soil. The obtained analysis results suggest that a rigid skirted footing behaves similar to a conventional strip footing under seismic and static loadings. The excessive deflection of flexible skirts under combined gravity and seismic loading renders them relatively ineffective than rigid skirts.

Accurately estimating the performance of tunnel boring machines (TBMs) is crucial for mitigating the substantial financial risks and complexities associated with tunnel construction. Machine learning (ML) techniques have emerged as powerful tools for predicting non-linear time series data. In this research, six advanced meta-heuristic optimization algorithms based on long short-term memory (LSTM) networks were developed to predict TBM penetration rate (TBM-PR). The study utilized 1125 datasets, partitioned into 20% for testing, 70% for training, and 10% for validation, incorporating six key input parameters influencing TBM-PR. The performances of these LSTM-based models were rigorously compared using a suite of statistical evaluation metrics. The results underscored the profound impact of optimization algorithms on prediction accuracy. Among the models tested, the LSTM optimized by the particle swarm optimization (PSO) algorithm emerged as the most robust predictor of TBM-PR. Sensitivity analysis further revealed that the orientation of discontinuities, specifically the alpha angle (α), exerted the greatest influence on the model's predictions. This research is significant in that it addresses critical concerns of TBM manufacturers and operators, offering a reliable predictive tool adaptable to varying geological conditions.

Cross-flaws are frequently encountered in practical rock engineering projects near horseshoe-shaped cavities, and their presence can significantly impact the failure mode of these cavities. This study utilizes a combination of laboratory experiments and numerical simulations to investigate the influence of cross-flaws on the failure mode of a horseshoe-shaped cavity. During the experimental tests, we varied the length of secondary flaw and the angle of the cross-flaws in the specimens, followed by subjecting them to biaxial compression. Our experimental results show that when the angle α between the primary and the secondary flaws is small (0° and 45°), only one crack is initiated at the vault of the cavity, resulting in a shear failure mode. Conversely, when the angle α is large (90° and 135°), two cracks are more likely to initiate at the vault of the cavity, leading to the failure mode of falling blocks in the surrounding rock. Furthermore, the circumferential stress at the cavity vault from numerical simulations results is consistent with this observed phenomenon. When the angle α is small, only one circumferential tensile stress concentration is observed at the cavity vault, resulting in the initiation of a single crack. In contrast, when the angle α is large, two stress concentrations appear at the vault of the cavity, leading to the initiation of two cracks from these locations.

This study investigates free and damped vibration behaviours of porous functionally graded shells supported by Winkler-Pasternak foundation, considering different geometries. Utilizing a higher-order shear deformation theory, the displacement field is determined. The equations of motion are formulated using Hamilton's principle, and the solutions are obtained Navier's method employing double Fourier series. Parametric studies regarding the effects of porosity, material distribution, elastic foundation, shell geometry and damping are carried out. Results are given in tabular and graphical form for the free and forced vibration analyses, respectively.

There is growing interest in introducing artificial ground freezing (AGF) as a method to temporarily secure unstable ground during tunnel construction. In order to efficiently operate an artificial ground freezing system, basic modeling research is needed on the changes in freezing behavior according to various soil environmental conditions as well as design conditions. In this study, a thermal-hydraulic coupled analysis was performed to simulate the artificial ground freezing process of ground containing salt water. The effect of major variables, including pore water salinity, on artificial ground freezing test performance was investigated. Additionally, an artificial neural network-based prediction model was proposed to estimate the time required to achieve the desired arch thickness. The artificial neural network model demonstrated reliable accuracy (R² = 0.9942) in predicting the time it would take to reach the desired arch thickness. Among the major input variables considered, pore water salinity appeared to be the most influential input variable, and initial soil temperature showed the least importance.

In practical engineering, the design process for most retaining walls necessitates careful consideration of seismic resistance. The prevention of retaining wall overturning is of paramount importance, especially in cases where the foundation's bearing capacity is limited. To research the seismic active earth pressure (ES) of a relieving retaining wall rotating around base (RB), the shear dissipation graphs across various operating conditions are analyzed by using Optum software, and the earth pressure in each region was derived by the inclined strip method combined with the limit equilibrium method. By observing shear dissipation graphs across various operating conditions, the distribution law of each sliding surface is summarized, and three typical failure modes are obtained. The corresponding calculation model was established. Then the resultant force and its action point were obtained. By comparing the theoretical and numerical solutions with the previous studies, the correctness of the derived formula is proved. The variation of earth pressure distribution and resultant force under seismic acceleration are studied. The unloading plate's position, the wall heel's length, and seismic acceleration will weaken the unloading effect. On the contrary, the length of the unloading plate and the friction angle of the filling will strengthen the unloading effect. The derived formula proposed in this study demonstrates a remarkable level of accuracy under both static and seismic loading conditions. Additionally, it serves as a valuable design reference for the prevention of overturning in relieving retaining walls.

Recently, electrical resistivity surveys have been used to obtain information related to underground structures including burial structure type and depth. However, various field conditions hinder understanding measured electrical resistance, and thus there is a need to understand how various geometries affect electrical resistance. This study explores the effect of geometric parameters of a structure and electrodes on electrical resistance in the framework of the finite element method. First, an electrical resistance module is developed using the generalized mesh modeling technique, and the accuracy of the module is verified by comparing the results with the analytical solution for a cylindrical electrode with conical tip. Then, 387 cases of numerical analysis including geometric parameters of a buried structure and electrodes are conducted to quantitatively estimate the detection depth under a steady-state current condition. The results show that electrical resistance is increased as (1) shallower burial depth of structure, (2) closer distance between ground electrode and structure, (3) longer horizontal electrode distance. In addition, the maximum detection depth corresponding to converged electrical resistance is deeper as (4) closer distance between ground electrode and structure, (5) shorter horizontal electrode distance. The distribution of the electric potential around the electrodes and underground structure is analyzed to provide a better understanding of the measured electrical resistance. As engineering purpose, the empirical equation is proposed to calculate maximum detection depth as first approximation.