Search Results (299)

Predicting tunnel seepage field is a critical challenge in the construction of underground engineering projects. While traditional analytical solutions and numerical methods struggle with complex geometric boundaries, standard Physics-Informed Neural Networks (PINNs) encounter additional challenges in tunnel seepage problems, including training instability, boundary handling difficulties, and low sampling efficiency. This paper develops an enhanced PINN framework designed specifically for predicting tunnel seepage field by integrating Exponential Moving Average (EMA) weight stabilization, Residual Adaptive Refinement with Distribution (RAR-D) sampling, and Bayesian optimization for collaborative training. The framework introduces a trial function method based on an approximate distance function (ADF) to address the precise handling of circular tunnel boundaries. The results demonstrate that the enhanced PINN framework achieves an exceptional prediction accuracy with an overall average relative error of 5 × 10⁻⁵. More importantly, the method demonstrates excellent practical applicability in data-scarce scenarios, maintaining acceptable prediction accuracy even when monitoring points are severely limited. Bayesian optimization reveals the critical insight that loss weight configuration is more important than network architecture in physics-constrained problems. This study is a systematic application of PINNs to prediction of tunnel seepage field and holds significant value for tunnel construction monitoring and risk assessment. Full article

Zenker’s diverticulum (ZD) is an esophageal condition that results in an outpouching of the mucosal layer through a weakened area in the hypopharyngeal wall. This condition can cause symptoms like dysphagia, regurgitation, and aspiration, impacting patients’ quality of life. Historically, open surgery was the primary treatment. Although effective, this method is associated with longer recovery times and risks such as infections, nerve damage, and prolonged hospitalization. Rigid endoscopic stapling emerged as a less invasive alternative, offering high success rates for patients with favorable anatomy. Zenker’s peroral endoscopic myotomy (Z-POEM), adapted from treatments for achalasia, represents the latest advancement in ZD management. It involves creating a submucosal tunnel and precisely dividing the cricopharyngeus muscle. Z-POEM is minimally invasive and often provides quick relief with a high success rate of around 92%, while enabling outpatient treatment or brief hospital stays. However, it requires specialized expertise, and long-term data on recurrence rates are still emerging. This review discusses the evolution of these treatment modalities through comprehensive searches of PubMed, MEDLINE, and ScienceDirect databases. Studies reporting on treatment outcomes, complication rates, operative times, and clinical success associated with open surgery, rigid endoscopic stapling, and Z-POEM were included, with emphasis on meta-analyses, multicenter studies, and large case series highlighting Z-POEM’s comparable success to open surgery and increased patient tolerance. Open surgery achieves long-term symptom resolution rates of 90–95% but is associated with higher complication rates (up to 30%) and prolonged recovery times. Rigid endoscopic stapling offers symptom relief in approximately 90% of cases, with lower morbidity and shorter hospital stays (1–2 days), though anatomical limitations restrict its use. Z-POEM has demonstrated clinical success rates of 85.5–93%, with major complications reported in 4.8–5% of cases and recurrence rates as low as 1.4% at one-year follow-up in larger diverticula. Z-POEM’s minimally invasive nature and suitability for high-risk patients make it increasingly preferred in specialized centers. Management of Zenker’s diverticulum has evolved significantly, with endoscopic techniques, particularly Z-POEM, offering comparable success to open surgery but with fewer complications and faster recovery. Ongoing advances in endoscopic equipment and technique, along with emerging data on long-term outcomes, are likely to further refine treatment algorithms for ZD, especially for elderly and high-risk populations. Future directions in ZD management include ongoing research to enhance the safety and efficacy of endoscopic techniques, with new technologies on the horizon that could further improve outcomes and accessibility. Full article

Accurate boundary detection is critical for autonomous trackless rubber-wheeled vehicles in underground coal mines, as it prevents lateral collisions with tunnel walls. Unlike open-road environments, underground tunnels suffer from poor illumination, water mist, and dust, which degrade visual imaging. To address these challenges, this paper proposes a navigable area computation for underground autonomous vehicles via ground feature and boundary processing, consisting of three core steps. First, a real-time point cloud correction process via pre-correction and dynamic update aligns ground point clouds with the LiDAR coordinate system to ensure parallelism. Second, corrected point clouds are projected onto a 2D grid map using a grid-based method, effectively mitigating the impact of ground unevenness on boundary extraction; third, an adaptive boundary completion method is designed to resolve boundary discontinuities in junctions and shunting chambers. Additionally, the method emphasizes continuous extraction of boundaries over extended periods by integrating temporal context, ensuring the continuity of boundary detection during vehicle operation. Experiments on real underground vehicle data validate that the method achieves accurate detection and consistent tracking of dual-sided boundaries across straight tunnels, curves, intersections, and shunting chambers, meeting the requirements of underground autonomous driving. This work provides a rule-based, real-time solution feasible under limited computing power, offering critical safety redundancy when deep learning methods fail in harsh underground environments. Full article

Polyurethane grouting materials are polymer materials formed through the reaction of polyisocyanates and polyols. They play important roles in underground engineering, tunnel construction, and mining due to their fast reaction rate, high bonding strength, and excellent impermeability. However, traditional polyurethane grouting materials have shortcomings such as high reaction heat release, high brittleness, and poor flame retardancy, which limit their applications in high-demand engineering projects. This paper systematically reviews the research progress on modified polyurethane grouting materials. Four major modification technologies are summarized: temperature reduction modification, flame retardant modification, mechanical enhancement, and environmental adaptability improvement. A multi-dimensional performance characterization system is established, covering slurry properties, solidified body performance, microstructure characteristics, thermal properties and flame retardancy, diffusion grouting performance, and environmental adaptability. The application effects of modified polyurethane grouting materials in grouting reinforcement, grouting water plugging, and grouting lifting are analyzed. Future development directions are projected. This review is particularly valuable for researchers and engineers working in tunneling, mining, geotechnical engineering, and infrastructure rehabilitation. Full article

In underground coal mines, narrow and irregular tunnels, dust, and gas hazards pose significant challenges to robotic path planning, particularly for shotcrete operations. The traditional A* algorithm has the limitations of limited safety, excessive node expansion, and insufficient dynamic obstacle avoidance capabilities. To address these, a hybrid algorithm integrating adaptive A*, Q-learning, and the Dynamic Window Approach (DWA) is proposed. The A* component is enhanced through improvements to its evaluation function and node selection strategy, incorporating dynamically adjustable neighborhood sampling to improve search efficiency. Q-learning re-plans unsafe trajectories in complex environments using a redesigned reward mechanism and an adaptive exploration strategy. The DWA module facilitates real-time obstacle avoidance in dynamic scenarios by optimizing both the velocity space and the trajectory evaluation process. The simulation results indicate that the proposed algorithm reduces the number of path nodes by approximately 30%, reduces the computational time by approximately 20% on 200 × 200 grids, and increases the path length by only 10%. These results demonstrate that the proposed approach effectively balances global path optimality with local adaptability, significantly improving the safety and real-time performance in complex underground environments. Full article

To ensure the safe and reliable operation of hydrogen-blended natural gas (HBNG) pipelines in urban utility tunnels, this study conducted a comprehensive CFD simulation of the leakage and diffusion characteristics of HBNG in confined underground environments. Utilizing ANSYS CFD software (2024R1), a three-dimensional physical model of a utility tunnel was developed to investigate the influence of key parameters, such as leak sizes (4 mm, 6 mm, and 8 mm)—selected based on common small-orifice defects in utility tunnel pipelines (e.g., corrosion-induced pinholes and minor mechanical damage) and hydrogen blending ratios (HBR) ranging from 0% to 20%—a range aligned with current global HBNG demonstration projects (e.g., China’s “Medium-Term and Long-Term Plan for Hydrogen Energy Industry Development”) and ISO standards prioritizing 20% as a technically feasible upper limit for existing infrastructure, on HBNG diffusion behavior. The study also evaluated the adequacy of current accident ventilation standards. The findings show that as leak orifice size increases, the diffusion range of HBNG expands significantly, with a 31.5% increase in diffusion distance and an 18.5% reduction in alarm time as the orifice diameter grows from 4 mm to 8 mm. Furthermore, hydrogen blending accelerates gas diffusion, with each 5% increase in HBR shortening the alarm time by approximately 1.6 s and increasing equilibrium concentrations by 0.4% vol. The current ventilation standard (12 h⁻¹) was found to be insufficient to suppress concentrations below the 1% safety threshold when the HBR exceeds 5% or the orifice diameter exceeds 4 mm—thresholds derived from simulations showing that, under 12 h⁻¹ ventilation, equilibrium concentrations exceed the 1% safety threshold under these conditions. To address these gaps, this study proposes an adaptive ventilation strategy that uses variable-frequency drives to adjust ventilation rates in real time based on sensor feedback of gas concentrations, ensuring alignment with leakage conditions, thereby ensuring enhanced safety. These results provide crucial theoretical insights for the safe design of HBNG pipelines and ventilation optimization in utility tunnels. Full article

Urban estuarine environments face increasing water safety risks due to microbial contamination from combined sewer overflows (CSOs), particularly during heavy rainfall events. In megacities like Tokyo, where waterfronts are widely used for recreation, such contamination poses significant public health risks. The challenge is compounded by the variability in both intensity and spatial distribution of rainfall across the catchment, combined with complex tidal dynamics making effective water quality management difficult. To address this challenge, we conducted a series of hydrodynamic–microbial fate simulations to examine the spatial and vertical behavior of Escherichia coli (E. coli) under different rainfall–tide conditions. Focusing on the Sumida River estuary, rainfall data from eight drainage areas were classified into six event types using cluster analysis. Two contrasting events were selected for detailed analysis: a light rainfall (G2, 15 mm over 13 h) and an intense event (G6, 272 mm over 34 h). Vertical water quality profiling was performed along an 8.5 km transect from the Kanda–Sumida River confluence to the Tokyo Bay Tunnel, illustrating E. coli and salinity. The results showed that the rainfall intensity and tidal phase at the event onset are critical in shaping both the magnitude and vertical distribution of microbial contamination. The intense event (G6) led to deep microbial intrusion (up to 6–7 m) and major salinity disruption, while the lighter event (G2) showed surface-layer confinement. Salinity gradients were more strongly affected during G6, indicating freshwater intrusion. Tidal phase also influenced transport: the flood-high condition retained E. coli, whereas ebb-low tides facilitated downstream flushing. These findings highlight the influence of rainfall intensity and tidal timing on microbial distribution and support the use of vertical profiling in estuarine water quality management. They also support the development of dynamic, event-based water quality risk assessment tools. With appropriate local calibration, the modeling framework is transferable to other urban estuarine systems to support proactive and adaptive water quality management. Full article

The tunnel construction environment is complex, and workers’ cognitive load directly affects safety and efficiency, making a dynamic assessment urgently needed. This study aims to explore the cognitive load of tunnel construction workers under different working environments using EEG technology. In the experimental design, subjects adapted to the virtual reality (VR) environment and received instructions before wearing a wireless EEG system and VR equipment to begin the formal experiment. Each subject underwent four rounds of experiments, corresponding to four different scenarios: control, night shift, noise, and confined space. Each round included three tasks of low, medium, and high difficulty. Analysis of EEG data showed that tunnel construction tasks in different environments significantly affected cognitive load, especially during night shifts and in confined spaces, with cognitive load increasing significantly with task difficulty. The results provide a theoretical basis for optimizing the management of tunnel construction environments and task design. Full article

Precise forecasting of rockburst intensity categories is vital to safeguarding operational safety and refining design protocols in deep underground engineering. This study proposes an intelligent forecasting framework through the integration of k-medoids-SMOTE and the BSLO-optimized Random Forest (BSLO-RF) algorithm. A curated dataset encompassing 351 rockburst instances, stratified into four intensity grades, was compiled via systematic literature synthesis. To mitigate data imbalance and outlier interference, z-score normalization and k-medoids-SMOTE oversampling were implemented, with t-SNE visualization confirming improved inter-class distinguishability. Notably, the BSLO algorithm was utilized for hyperparameter tuning of the Random Forest model, thereby strengthening its global search and local refinement capabilities. Comparative analyses revealed that the optimized BSLO-RF framework outperformed conventional machine learning methods (e.g., BSLO-SVM, BSLO-BP), achieving an average prediction accuracy of 89.16% on the balanced dataset—accompanied by a recall of 87.5% and F1-score of 0.88. It exhibited superior performance in predicting extreme grades: 93.3% accuracy for Level I (no rockburst) and 87.9% for Level IV (severe rockburst), exceeding BSLO-SVM (75.8% for Level IV) and BSLO-BP (72.7% for Level IV). Field validation via the Zhongnanshan Tunnel project further corroborated its reliability, yielding an 80% prediction accuracy (four out of five cases correctly classified) and verifying its adaptability to complex geological settings. This research introduces a robust intelligent classification approach for rockburst intensity, offering actionable insights for risk assessment and mitigation in deep mining and tunneling initiatives. Full article

Simultaneous Localization and Mapping (SLAM) is a critical technology for robot intelligence. Compared to cameras, Light Detection and Ranging (LiDAR) sensors achieve higher accuracy and stability in indoor environments. However, LiDAR can only capture the geometric structure of the environment, and LiDAR-based SLAM often fails in scenarios with insufficient geometric features or highly similar structures. Furthermore, low-cost mechanical LiDARs, constrained by sparse point cloud density, are particularly prone to odometry drift along the Z-axis, especially in environments such as tunnels or long corridors. To address the localization issues in such scenarios, we propose a forward-enhanced SLAM algorithm. Utilizing a 16-line LiDAR and a monocular camera, we construct a dense colored point cloud input and apply an efficient multi-modal feature extraction algorithm based on centroid distance to extract a set of feature points with significant geometric and color features. These points are then optimized in the back end based on constraints from points, lines, and planes. We compare our method with several classic SLAM algorithms in terms of feature extraction, localization, and elevation constraint. Experimental results demonstrate that our method achieves high-precision real-time operation and exhibits excellent adaptability to indoor environments with similar structures. Full article

While foraging tunnels of the red imported fire ant, Solenopsis invicta, have been well studied, much less is known about the tunnels constructed between neighboring nests, despite their perceived importance in intra-colony exchange and collaboration. In this study, we investigated such tunnels by excavating 80 pairs of nests (with distances of <1 m between nests) located in different types of habitats. For each pair of nests, we recorded the number of inter-nest tunnels and observed their shape, diameter, subsurface depth, and ant presence within them. Moreover, we analyzed the relationships between the probability of constructing inter-nest tunnels and several nest/habitat characteristics, including distance between nests, colony social form, nest size, soil type, and vegetation cover, as well as the relationships between tunnel numbers and these factors. The results show that the number of inter-nest tunnels ranges from one to 11. These tunnels open to the two nests at terminals, are elliptical in cross-section, <1.5 cm in diameter, and mostly at 1–3 cm (range 1–12 cm) subsurface depth. Among the 36 pairs of nests possessing tunnels, 31 pairs (86.1%) had worker or alate ants within their tunnels. Polygynous colonies are more likely (52.4%) to construct inter-nest tunnels than monogynous colonies (17.6%). Nest pairs that have a small nest, located in habitats with higher vegetation cover and loamy or sandy loam soil, tend to have inter-nest tunnels. We also showed that the capacity of constructing inter-nest tunnels falls in the regime similar to foraging tunnels. As nests were treated with chemicals, 33 nests were relocated and 47 new nests resulted within 2 weeks, but no definite tunnels were constructed between original nests and corresponding new nests. Our results highlight the significance of including such tunnels when analyzing intra-colony exchange, collaboration, and adaptive strategies in S. invicta. Uses of tunnels by fire ants during nest relocation, and the requirement of destroying them during control program implementation, were discussed. Full article

An asymptotic stability velocity tracking controller is designed to enable the autonomous maneuvering flight of unmanned helicopters. Firstly, taking the UH-60A without pilots as the research object, a high-efficient rotor aerodynamic modeling is developed, which incorporates a free-wake vortex method with the flap response of blades. The consummate flight dynamic model is complemented by wind tunnel-validated fuselage/tail rotor load regressions. Secondly, a linear state–space equation is derived via the small perturbation linearization method based on the flight dynamic model within the body coordinate system. A decoupled model is formulated based on the linear state–space equation by employing the implicit model approach. Subsequently, a system of ordinary differential equations is constructed, which is related to the deviation between actual velocity and its expected value, along with higher-order derivatives of this discrepancy. The I&I (immersion and invariance) theory is then employed to facilitate the design of a non-cascade control loop. Finally, the response of desired velocity in longitudinal channel is simulated with step signal to compare the control effect with a PID (proportional–integral–derivative) controller. By adjusting the coefficients, the response progress of the PID controller is similar to the effect of adaptive controller with I&I theory. However, there is no obvious overshoot in the process with I&I adaptive controller, and the average response amplitude accounts for 16.69% of the random white noise, which is 7.38% of the oscillation level under the PID controller. The parameter tuning complexity when employing I&I theory is significantly lower than that of the PID controller, which is evaluated by mathematical derivations and simulations. Meanwhile, the sidestep and pirouette maneuvers are simulated and analyzed to examine the controller in accordance with the performance criteria outlined in the ADS-33E-REF standards. The simulation results demonstrate that the speed expectation-oriented asymptotic stability control can achieve a fast response. Both sidestep and pirouette maneuvers can satisfy the desired performance requirements stipulated by ADS-33E-REF. Full article

Hidradenitis suppurativa (HS) is a chronic, relapsing inflammatory dermatosis of the pilosebaceous unit characterized by nodules, abscesses, and dermal tunnels. Recent transcriptomic studies have implicated dysregulation of innate and adaptive immune responses, epidermal barrier dysfunction, and systemic metabolic alterations. This review synthesizes findings from 16 studies investigating the HS transcriptome using bulk and single-cell RNA sequencing. Differential gene expression analyses revealed extensive upregulation of inflammatory cytokines and chemokines, particularly in lesional and perilesional skin. These changes were also mirrored in non-lesional skin, suggesting diffuse immune dysregulation beyond visibly affected areas. Downregulated pathways include those involved in lipid metabolism, muscle contraction, and neuronal signaling, potentially linking HS to obesity, metabolic syndrome, and neuropsychiatric comorbidities. Single-cell transcriptomics confirmed the enrichment of keratinocytes and immune cells (B cells, plasma cells, M1 macrophages, and T cells) with proinflammatory profiles in HS lesions. Keratinocyte dysfunction further implicated a compromised epidermal barrier in disease pathogenesis. While transcriptomic studies have advanced mechanistic understanding and highlighted therapeutic targets—such as the IL-1β–TH17 axis and B cell signaling pathways—methodological heterogeneity limits cross-study comparisons. Integration of multi-omics data and standardized phenotyping will be essential to identify robust biomarkers, stratify HS subtypes, and guide personalized therapeutic approaches. Full article

In wind tunnel tests for aerospace and bridge engineering, the accurate prediction of Mach number remains a core challenge to ensure the reliability of airflow dynamics characterization. Pure data-driven models often fail to meet high-precision prediction requirements due to the lack of physical mechanism constraints and insufficient generalization capability. This paper proposes a physical information-based long short-term memory network (P-LSTM), which constructs a physical loss function by embedding isentropic flow equations from gas dynamics, thereby constraining the Mach number prediction solution space within the physically feasible domain. This approach effectively balances the neural network’s ability to capture temporal features with the interpretability of physical mechanisms. Aiming at the scarcity of data in new wind tunnel scenarios, an adaptive weight transfer learning method (AWTL) is further proposed, realizing efficient knowledge transfer across different-scale wind tunnels via cross-domain data calibration, adaptive source-domain weight reweighting, and target-domain fine-tuning. Experimental results show that the P-LSTM method achieves a 50.65–62.54% reduction in RMSE, 48.00–54.05% in MAE, and 47.88–73.68% in MD compared with traditional LSTM for Mach number prediction in the 0.6 m continuous wind tunnel flow field. The AWTL model also outperforms the direct training model significantly in the 2.4 m continuous wind tunnel, with RMSE, MAE, and MD reduced by 85.26%, 95.12%, and 71.14%, respectively. These results validate that the proposed models achieve high-precision Mach number prediction with strong generalization capability. Full article

To enable intelligent inspection of underground cable systems, this study presents a gecko-inspired quadruped robot that integrates multi-degree-of-freedom motion with a deep learning-based visual detection system. Inspired by the gecko’s flexible spine and leg structure, the robot exhibits strong adaptability to confined and uneven tunnel environments. The motion system is modeled using the standard Denavit–Hartenberg (D–H) method, with both forward and inverse kinematics derived analytically. A zero-impact foot trajectory is employed to achieve stable gait planning. For defect detection, the robot incorporates a binocular vision module and an enhanced YOLOv8 framework. The key improvements include a lightweight feature fusion structure (SlimNeck), a multidimensional coordinate attention (MCA) mechanism, and a refined MPDIoU loss function, which collectively improve the detection accuracy of subtle defects such as insulation aging, micro-cracks, and surface contamination. A variety of data augmentation techniques—such as brightness adjustment, Gaussian noise, and occlusion simulation—are applied to enhance robustness under complex lighting and environmental conditions. The experimental results validate the effectiveness of the proposed system in both kinematic control and vision-based defect recognition. This work demonstrates the potential of integrating bio-inspired mechanical design with intelligent visual perception to support practical, efficient cable inspection in confined underground environments. Full article