Tunnelling and Underground Space Technology最新文献

{"title":"Investigating user perception of underground public space from a temporal perspective: A framework using social media data and combining LLMs and BERT models","authors":"Chen-Xiao Ma,&nbsp;Fang-Le Peng,&nbsp;Zi-Jian Li,&nbsp;Tian-Ze Zhang,&nbsp;Yong-Kang Qiao","doi":"10.1016/j.tust.2025.107390","DOIUrl":"10.1016/j.tust.2025.107390","url":null,"abstract":"<div><div>User perception of underground public space (UPS) includes both space perception and sentiment perception derived from its users. Examining how user perception evolves over time is essential for user-oriented UPS planning and renovation. Social media data provides a novel data source for capturing dynamic interactions between UPS and users. This study developed a text mining-based analytical framework to investigate 15 UPS cases in Shanghai. Large language models (LLMs) and Bidirectional Encoder Representations from Transformers (BERT) models were employed to analyze user perception from a temporal perspective. A total of 27 perception elements, categorized into six groups, were identified and annotated using LLMs. Elements related to space functions were most frequently perceived by users, while space public facilities received the least attention. The diversity of high-frequency perception elements increased from 2006 to 2024. On the other hand, sentimental polarity analysis of complete user reviews using the BERT model effectively assessed UPS satisfaction. User sentiments regarding space function elements have remained stable since 2006. However, user satisfaction with space environment, design and layout, and management elements exhibited an increasing trend. Furthermore, the user-perceived functions of UPS differed from its original planning objectives, highlighting the demands for multi-functional and adaptable UPS during its planning and design. Moreover, regression analysis of UPS sentiments also revealed that optimizing UPS atmosphere and management during renovation enhanced space performance.</div><div>Abbreviations: API, application programming interface; NLP, natural language processing; UUS, urban underground space; BERT, Bidirectional Encoder Representations from Transformers; LLM, large language model; UPS, underground public space.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107390"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145796219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development, challenges and potentials of underground energy storage systems","authors":"Huachuan Wang ,&nbsp;Shangtong Yang ,&nbsp;Stephanie Flude ,&nbsp;Ketong Zhang ,&nbsp;Qianbing Zhang ,&nbsp;Jian Zhao ,&nbsp;Zoe K. Shipton","doi":"10.1016/j.tust.2025.107360","DOIUrl":"10.1016/j.tust.2025.107360","url":null,"abstract":"<div><div>The rapid development of intermittent renewable energy (such as wind and solar power) results in a high demand for large-scale energy storage. Underground energy storage (UES) presents a Gigawatt-scale opportunity to address the growing needs for efficient, reliable, and sustainable energy systems. Although the previous study has analysed or reviewed technical characteristics of UES systems, most was focused on one specific type of technology, while a systematic and quantitative comparison among different categories of UES systems is still limited. This paper comprehensively reviews the existing and emerging UES technologies, providing a science introduction of concepts, principles and technical developments of different UES categories to the community. A systematic comparison of system and engineering parameters among different UES systems is conducted based on the massive data collected from established and proposed projects, which can provide a guidance and reference for designers and engineers to make informed decisions with regards to the optimal UES scheme. To help with the UES site selection and design optimisation, geological and structural properties are further summarised and discussed, which are key components during the play fairway analysis. Besides, new engineering challenges are proposed regarding the deterioration of structural and rock materials in UES multiphysical environments, which will not only affect the accuracy of numerical and theoretical assessment and prediction, but also threaten the serviceability and durability of UES infrastructures. Finally, a global potential map for UES is established to help policy-makers formulate region-specific UES policies and foster international collaboration, further accelerating the deployment of UES infrastructures.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107360"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of tunnel rock mass integrity using multi-modal data and Generative Large Language Model: TunnelRIP-GPT","authors":"Chen Wu ,&nbsp;Hongwei Huang ,&nbsp;Yi-Qing Ni ,&nbsp;Haoxuan Zhang","doi":"10.1016/j.tust.2025.107291","DOIUrl":"10.1016/j.tust.2025.107291","url":null,"abstract":"<div><div>The prediction of tunnel rock mass integrity is critical for ensuring construction safety. Tunnel rock mass offers a vast amount of multimodal data, yet accurately integrating these multimodal to predict rock mass integrity remains challenging. This study proposes a novel framework combining an Enhanced Computer Vision (CV) model and interactive Generative Large Language Models (LLMs) to address this gap. First, the Stable Diffusion method is employed to augment fracture image datasets, mitigating data imbalance issues. An improved CV method achieves 83% test accuracy in rock fracture segmentation. Subsequently, seven parameter sets-encompassing apparent, internal, physico-mechanical, and design characteristics are constructed from geological sketches, ground-penetrating radar, and drilling reports. <em>A Tunnel rock integrity prediction GPT</em> method is developed, which achieves a higher prediction accuracy of 92% and overcomes the limitation of conventional Transformer and CNN methods, which cannot directly integrate these multi-modal data (including images, numerical values, and texts). Finally, generalizability is confirmed through validation on untrained datasets. The study further evaluates the predictive efficacy of individual parameter sets, enabling tailored solutions for diverse tunnel conditions, and demonstrates the potential of generative LLMs to replace traditional CV and digital image processing in geotechnical applications.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107291"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A knowledge-data dual-driven framework for predicting surrounding rock classification in TBM tunneling and generalization analysis","authors":"Wei Luo ,&nbsp;Yadong Xue ,&nbsp;Yulong Sun ,&nbsp;Wei Zhang ,&nbsp;Jie Wang","doi":"10.1016/j.tust.2025.107378","DOIUrl":"10.1016/j.tust.2025.107378","url":null,"abstract":"<div><div>Accurate and real-time prediction of surrounding rock classification (SRC) in tunnel boring machine (TBM) operations is critical for construction risk management and dynamic support design. Although machine learning has shown potential in improving SRC prediction accuracy, its generalization performance in practical engineering applications remains insufficiently explored. This study proposes a knowledge-data dual driven SRC prediction framework (KD-SRC) based on rock–machine interaction principles and evaluates its performance using historical excavation sequences. The framework incorporates physical relationships into feature engineering through the Colorado School of Mines (CSM) model and the adaptive piecewise constant approximation (APCA) method. A TPE-LightGBM voting model, which employs a tree-structured Parzen estimator to optimize the light gradient boosting machine, is used to achieve SRC prediction at the boring cycle level. Case studies on two deep-buried, long tunnel projects demonstrate that the proposed framework achieves over 89 % accuracy, with a recall exceeding 0.94 under adverse geological conditions. When dynamically updating the prediction model through excavation history, a strong negative linear correlation was observed between dataset distribution discrepancies and prediction performance, confirming that model generalization is influenced by similarities in feature distribution. Cross-project learning demonstrates that joint multi-project training improves prediction accuracy during the early stages of construction, and SHAP analysis validates the model’s credibility by aligning learned patterns with rock-machine interaction principles. Overall, this study not only improves the accuracy and efficiency of SRC prediction but also offers insights for developing more robust cross-project applications.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107378"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AGFNN: A smart platform for uncertainty-aware prediction of freezing time in artificial ground freezing","authors":"Kai-Qi Li ,&nbsp;Zhen-Yu Yin ,&nbsp;Yong Liu","doi":"10.1016/j.tust.2025.107331","DOIUrl":"10.1016/j.tust.2025.107331","url":null,"abstract":"<div><div>Accurate and efficient prediction of freezing time (<em>t</em>_f) is essential for the design and safety evaluation of artificial ground freezing (AGF) systems. Traditional numerical simulations, while accurate, are computationally intensive and often impractical for iterative design or real-time decision-making. This study proposes a surrogate modeling framework based on a neural network (<em>NN</em>) architecture to predict <em>t</em>_f across a wide range of soil conditions and design parameters. The surrogate model is trained on a large dataset generated from transient finite element simulations and incorporates Monte Carlo dropout (MCD) to enable uncertainty quantification. The <em>NN</em>-based surrogate model demonstrates high predictive accuracy and rapid inference, enabling reliable estimation of <em>t</em>_f with associated confidence intervals. To enhance model transparency, the Shapley Additive Explanations (SHAP) method is also employed to evaluate the relative importance of each input parameter in influencing the predicted output. The trained model is deployed as a standalone desktop application named <em>AGFNN</em> with an intuitive graphical user interface (GUI), supporting both single-case manual input and batch-mode Excel-based predictions. The platform streamlines the prediction workflow for engineering users, enabling uncertainty-aware decision-making without the need for numerical modeling expertise. Results demonstrate that the proposed method significantly reduces computational cost while maintaining interpretability, offering a robust and accessible tool for design optimization and risk-informed decision support in AGF projects.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107331"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unsupervised deep-learning model for quantitative diagnosis of tunnel lining cavity condition using interfering GPR data","authors":"Tess Xianghuan Luo ,&nbsp;Qingzhou Zheng ,&nbsp;Feifei Hou","doi":"10.1016/j.tust.2025.107373","DOIUrl":"10.1016/j.tust.2025.107373","url":null,"abstract":"<div><div>Accurate diagnosis of cavity conditions in aging metro tunnel linings is essential for effective maintenance. This study introduces a novel three-stage method for quantitatively diagnosing cavity conditions in tunnel linings using interfering GPR radargrams. Firstly, defective tunnel lining sections are identified using an enhanced EfficientNetV2s model, preventing false alarms caused by artifacts generated during noise suppression. Secondly, suppresses reinforcement noise and enable accurate reconstruction of radargram reflections related to size and position. It is by employing the unsupervised Contrastive Unpaired Translation (CUT) model as a baseline, with further improvements enhance feature extraction and stability. Finally, cavity conditions are quantitatively characterised with Faster-RCNN from the reconstructed radargrams. Experiments demonstrate that cavity size identification in primary lining and surrounding rock achieved an accuracy of 93.3 % and 79.24 %, respectively, while defective lining classification reached 92.9 % accuracy. The reconstruction also achieved an SSIM of 0.898 and MS-SSIM of 0.762, respectively. The proposed method is unsupervised, requiring no paired training data, but demonstrates superior performance.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107373"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145732323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mechanical performance evaluation of buried pipelines with corrugated flexible joints under fault displacements","authors":"Rulin Zhang ,&nbsp;Chen Wang ,&nbsp;Yiming Lang ,&nbsp;Siying Chen ,&nbsp;Liyun Zhang ,&nbsp;Yunhao Wen","doi":"10.1016/j.tust.2025.107350","DOIUrl":"10.1016/j.tust.2025.107350","url":null,"abstract":"<div><div>Buried pipelines crossing active faults are highly susceptible to permanent ground deformation induced by earthquakes, which can lead to severe structural damage. To enhance their seismic performance, this study investigates the mechanical behavior of buried steel pipelines equipped with corrugated flexible joints (PFJ). A refined three-dimensional (3D) nonlinear finite element model was developed in ABAQUS, explicitly incorporating the corrugated geometry of flexible joints and realistic pipe–soil interaction. The model enables a comprehensive evaluation of the coupled deformation mechanisms among joints, pipe segments, and surrounding soil. The study first examines the effects of joint number and layout, followed by a comparative analysis between PFJ and conventional continuous pipelines (CP). A series of parametric studies were then performed considering fault type, dip angle, internal pressure, and the diameter-to-thickness ratio (<span><math><mrow><mi>D</mi><mo>/</mo><mi>t</mi></mrow></math></span>). The results indicate that PFJ effectively confines deformation within the joints and significantly reduces stress and strain in adjacent pipe segments. The two joints closest to the fault plane dominate the overall deformation pattern, with the tensile-side joint identified as the most critical location for potential failure. Among all the parameters, fault type and <span><math><mrow><mi>D</mi><mo>/</mo><mi>t</mi></mrow></math></span> ratio have a significant impact on the overall deformation of the PFJ system. The reverse fault causes the most severe joint deformations, and an excessively large <span><math><mrow><mi>D</mi><mo>/</mo><mi>t</mi></mrow></math></span> ratio may lead to the failure of the joint’s protective effect, causing buckling of the pipe segments. In contrast, the fault dip angle and internal pressure have relatively minor effects on the overall deformation of PFJ. Changes in fault dip angle primarily affect the intensity of the deformation, rather than altering the overall deformation pattern of the PFJ. Internal pressure influences only the stress and strain values, with negligible impact on the overall deformation of the PFJ system.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107350"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145732352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation and analysis of the ductile dynamic response of mountain tunnels based on shaking table tests","authors":"Guo Xu ,&nbsp;Feng Gao ,&nbsp;Zihao Zhang ,&nbsp;Mengzhu Peng ,&nbsp;Xukai Tan ,&nbsp;Weiwei Wu ,&nbsp;Qiang Hui","doi":"10.1016/j.tust.2025.107328","DOIUrl":"10.1016/j.tust.2025.107328","url":null,"abstract":"<div><div>Dynamic response tests using a shaking table were conducted to directly measure the time-history relationship of ductile displacements in cast-in-place horseshoe-shaped mountain tunnels. These tests provided displacement and maximum deformation rate (MDR) control indices for the tunnels. Shaking table tests on such tunnels have typically focused on seismic mechanisms, primarily through measurements of acceleration, strain, strength, and internal forces. However, research on ductile displacements and deformation rate control indices remains limited. In this study, a novel experimental methodology was developed and utilized to directly quantify the time-history relationship of ductile displacements in mountain tunnels. A series of shaking table tests were conducted to evaluate the seismic performance, influence extent, and impact range of the tunnels. The seismic response characteristics and damage assessment under various conditions were systematically investigated. The primary conclusions are summarized as follows: (1) The practicality and effectiveness of the method for directly measuring the time-history relationship of ductile displacements were validated through testing and analysis. (2) Recommended MDR thresholds for ductility indicators in mountain tunnels are proposed, with values of 8.27 ‰, 6.09 ‰, and 4.85 ‰ for shallow-buried, medium-buried, and deep-buried tunnels, respectively. (3) Burial depth is identified as a critical factor influencing the seismic performance of mountain tunnels, with both displacement and MDR showing a decreasing trend as burial depth increases. (4) Based on the seismic response characteristics and damage patterns of cast-in-place horseshoe-shaped mountain tunnels, four distinct damage patterns were identified: elastic, elastoplastic, plastic-critical, and critical-collapse. These findings provide a foundation for predicting damage and failure, as well as for developing probabilistic seismic demand models for cast-in-place horseshoe-shaped tunnels under seismic loading, thus offering valuable references for the establishment of relevant industry standards and guidelines.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107328"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145651972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Failure mechanism and stability control of fault induced by underground cavern excavation: insights from theoretical analysis and numerical modeling","authors":"Guangzhe Tao ,&nbsp;Zongqing Zhou ,&nbsp;Bing Shao ,&nbsp;Ning Liu ,&nbsp;Chenglu Gao ,&nbsp;Daosheng Zhang","doi":"10.1016/j.tust.2025.107410","DOIUrl":"10.1016/j.tust.2025.107410","url":null,"abstract":"<div><div>The fault is one of the key factors that restricts the stability of surrounding rock masses in underground caverns. A comprehensive method including theoretical analysis, numerical simulation, and field investigation was conducted to investigate the failure mechanisms and control strategy of fault based on the underground cavern of the Kala (KL) hydropower station in Southwest China. Firstly, a mechanical failure criterion of fault failure was derived based on Anderson’s fault stress model and the Mohr-Coulomb criterion. Then, discrete element numerical software 3DEC was used to reveal the evolution process of fault slip and the influence mechanisms of system parameters, including shear stiffness, fault dip angle, and shear strength. Finally, a support and control strategy for fault stability was proposed based on the fault instability failure mechanism, which was implemented in the field. The results indicate that fault stability increases with increasing internal friction angle, cohesion, and minimum principal stress, but decreases with increasing maximum principal stress. The fault slip process exhibits three distinct successive stages: slow growth, rapid evolution, and stable equilibrium. Numerical results show that fault slip displacement follows a nonlinear trend with increasing dip angle (increasing first and then decreasing) and decreases significantly with higher shear stiffness and internal friction angle. A synergistic support strategy including “precision blasting + timely support” and “shotcrete + prestressed rockbolt + prestressed cables” was proposed based on the fault failure mechanism, which significantly reduces the risk of fault instability activation. Field monitoring indicates that the support scheme significantly reduces the tendency for fault instability, enhancing the stability of the surrounding rock.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107410"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Urban underground space for resilient cities","authors":"Wout Broere ,&nbsp;Monika Mitew-Czajewska","doi":"10.1016/j.tust.2025.107412","DOIUrl":"10.1016/j.tust.2025.107412","url":null,"abstract":"<div><div>Urban resilience is the ability of an urban environment, including its infrastructure and inhabitants, to remain functioning in the face of acute shocks and chronic stresses. This paper explores the role underground space plays in urban resilience, considering both short-term events and long-term changes. Underground structures are inherently resilient against damage caused by natural disasters, such as earthquakes, but they can be at higher risk of damage due to sudden flooding. The risk of flooding requires special attention in the context of climate change, as the risk of sudden flood events rises. At the same time underground space offers opportunities to help reduce climate change impacts, and use of underground space is essential in realizing compact and energy efficient cities, and as such is an essential component of a resilient city.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107412"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}