As an effective measure to solve the cracking problem of primary support in large deformation tunnels, yielding support has an important influence on the mechanical properties of primary support and the energy characteristics of the surrounding rock. Based on the stress-strain curves of common yielding members, the mechanical characteristics of yielding members are summarized, and a numerical method is developed to realize the mechanical properties of yielding members based on the Beam element and Interface element. The concrete elastoplastic damage constitutive model is used to carry out numerical calculation, and the influence of yield member position and its constant resistance on the mechanical properties of the initial support and surrounding rock energy is discussed. The results indicate that: It is beneficial to reduce the internal force and damage of spray layer and the stress and energy of surrounding rock by setting compression members in different positions, and when the maximum initial geostress is oriented horizontally, the pressure-relief effect of yield members subjected to horizontal deformation is more pronounced. When the yield member position is fixed and the constant resistance is different, the maximum damage of spray layer, bending moment and axial force all increase first and then decrease with the increase of constant resistance, and the rock stress and elastic strain energy density around the hole also increase first and then decrease with the increase of constant resistance.
To elucidate the microscopic mechanisms of pore structure evolution and external fluid interactions in siliceous shale reservoirs, systematic sensitivity experiments were conducted on siliceous shale reservoirs in the eastern margin of the Ordos Basin, focusing on external fluids such as variable-viscosity slick water and flowback fluid. Combined with the nuclear magnetic resonance (NMR) T2 spectrum, the mechanism of action of variable viscosity slick water and flowback fluid on the pore structure of siliceous shale was studied. The evolution mechanism of pore throat size distribution was analyzed, and the difference in action mechanism between variable viscosity slick water and flowback fluid was clarified. The results show that: The variable viscosity of slickwater has a harmful effect on the total porosity of siliceous shale. The total porosity of siliceous shale can be reduced by about 6.8%. Among them, the decrease of macropores is the largest, followed by mesopores. The proportion of macropores and mesopores can be reduced by 46.2% and 7.4%, respectively, while the proportion of micropores shows an increasing trend. The total porosity of siliceous shale changes little after the action of flowback fluid. The change rate of total porosity after the action of flowback fluid is 0.04% and -0.02%. Among them, the proportion of medium pores increases, and the degree of change in micro-nano and small pores is small. The slickwater has weak damage to the total porosity of siliceous shale, micro-nano pores are not sensitive, small pores are not a sensitive-weak improvement, and large pores are moderately weak damage; the flowback fluid is not sensitive to the total porosity of siliceous shale and is not sensitive to micro-nano pores, small pores, mesopores, and macropores. The research results can provide a reference for fracturing process optimization and efficient development of shale oil.
A quantitative risk assessment method for tunnel water inrush disasters is developed to enhance safety risk prevention capabilities under the influence of water-bearing structures. Addressing the water inrush risks induced by the coupling effects of water-bearing structure spatial positioning and surrounding rock disturbance, an integrated evaluation methodology combining geological detection and mechanical analysis is proposed. The methodology comprises three core components: First, the transient electromagnetic method is employed for advanced water detection to accurately identify the three-dimensional spatial distribution of water-bearing structures. Second, graphical analysis is applied to partition the surrounding rock disturbance into failure zone Df, stress-disturbed elastic zone De, and natural stress zone, establishing a spatial superposition risk assessment model between water-bearing structures and disturbance zones. Finally, for medium or high-risk scenarios, a danger coefficient Kc calculation formula constrained by force equilibrium is derived to dynamically adjust risk levels. Engineering validation demonstrates: When a water-bearing structure in a tunnel case was 9.00 m from the tunnel wall, combined with disturbance boundary values of Df = 2.40 m and De = 12.65 m, the initial risk assessment identified medium risk. After recalculating Kc = 0.87, which is less than 1.00, the risk level was upgraded to high, consistent with subsequent actual water inrush incidents. Two key achievements are summarized as follows: a dual-level risk assessment framework of u201Cpreliminary determination-quantitative rechecku201D is established, achieving the transition from qualitative to quantitative risk grading, and the defined boundary values of the disturbance zone can provide a scientific basis for the minimum safe distance in advanced geological detection. The research outcomes form a complete technical system covering the entire process of u201Cdetection-analysis-evaluation-preventionu201D for tunnel water inrush risks, substantially improving assessment accuracy and offering innovative solutions for safe tunnel construction in complex geological conditions.
Although some aging-friendly facilities have been built in Chinese subway stations in accordance with relevant specifications, existing stations have problems such as facility aging, unclear information signs, imperfect service functions, and loopholes in emergency management. Through field investigations and literature analysis, the aging-friendly features of Japanese subway stations are deeply analyzed. Japanese subway stations fully consider the physical functions and travel needs of the elderly in aspects such as in-station passage facilities, information exchange facilities, service facilities, and emergency support measures. Their designs are elaborate and human-centered. By learning from the experience of Japan, China should improve the planning and design standards for aging-friendly facilities, introduce intelligent technologies, optimize information display and services, strengthen personnel training and deployment, and build a comprehensive emergency management system. This can enhance the aging-friendly level of subways, ensure the travel safety and comfort of the elderly, and promote the high-quality development of social public transportation services.
To address the challenges of low efficiency and high computational costs faced by deep learning models in tunnel surrounding rock integrity analysis, a dual-model strategy based on YOLO and DeepLab is proposed, enabling rapid assessment of tunnel surrounding rock integrity. This method constructs a rock fracture recognition model based on the YOLO algorithm and introduces the DeepLab image semantic segmentation algorithm to intelligently extract fracture parameters, with results visualized. By integrating the processed image information, a new indicator for evaluating the development degree of rock fracturesu2014the Fracture Factor (Ff)u2014is introduced to quantitatively analyze the development degree of surrounding rock fractures and determine the surrounding rock integrity coefficient (Kv). Using 5 000 images from typical domestic engineering projects for testing, the results show that the dual-model strategy based on YOLO and DeepLab achieves a fracture recognition accuracy of 96.41% and a fracture segmentation accuracy of 94.48%. Compared to traditional deep learning algorithms, the Fracture Factor (Ff) provides an assessment of surrounding rock integrity that is closer to the integrity coefficient in the BQ method. The dual-model strategy significantly reduces computational costs while ensuring accuracy, enhancing the feasibility of deep learning models in practical engineering applications.
