Pub Date : 2026-04-01Epub Date: 2025-12-10DOI: 10.1016/j.tust.2025.107370
Jiaqi Li , Yuxuan Wang , Jie Huang , Bowen Du , Zhouhong Zong , Minghong Li
Internal explosions in tunnels can cause severe casualties and structural damage, with end closures significantly amplifying shock wave effects. To investigate the influence of end closure on shock wave propagation, two large-scale internal explosion tests were conducted, and a validated finite element model was established for both open-ended and one-end closed tunnel conditions. The model was then employed to analyze the effects of end-face failure time, charge position, and explosive mass. The results indicate that reflected waves from the closed end produce a distinct secondary overpressure peak, which increases as the measurement point approaches the closed end. A demarcation point was identified that separates regions dominated by either the first or end-reflected peak overpressure. Both peak overpressures are primarily influenced by explosive mass, with location-induced average deviations of less than 8 %. While end-face failure time has a negligible influence on peak overpressures, it significantly affects pressure decay following reflection. Predictive formulas are proposed to estimate the first and end-reflected peak overpressures, as well as the demarcation point location, and have been rigorously validated with average errors below 7 %. This study provides a valuable reference for blast-resistant tunnel design.
{"title":"Blast wave propagation and overpressure prediction in a tunnel with one closed end","authors":"Jiaqi Li , Yuxuan Wang , Jie Huang , Bowen Du , Zhouhong Zong , Minghong Li","doi":"10.1016/j.tust.2025.107370","DOIUrl":"10.1016/j.tust.2025.107370","url":null,"abstract":"<div><div>Internal explosions in tunnels can cause severe casualties and structural damage, with end closures significantly amplifying shock wave effects. To investigate the influence of end closure on shock wave propagation, two large-scale internal explosion tests were conducted, and a validated finite element model was established for both open-ended and one-end closed tunnel conditions. The model was then employed to analyze the effects of end-face failure time, charge position, and explosive mass. The results indicate that reflected waves from the closed end produce a distinct secondary overpressure peak, which increases as the measurement point approaches the closed end. A demarcation point was identified that separates regions dominated by either the first or end-reflected peak overpressure. Both peak overpressures are primarily influenced by explosive mass, with location-induced average deviations of less than 8 %. While end-face failure time has a negligible influence on peak overpressures, it significantly affects pressure decay following reflection. Predictive formulas are proposed to estimate the first and end-reflected peak overpressures, as well as the demarcation point location, and have been rigorously validated with average errors below 7 %. This study provides a valuable reference for blast-resistant tunnel design.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107370"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738097","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}
Pub Date : 2026-04-01Epub Date: 2025-12-27DOI: 10.1016/j.tust.2025.107414
Peng Xie , Junling Bu , Yulin Yang , Bin Li
Frequent road collapses in Guangzhou, China, have been linked to disjoints in underground drainage pipes. Clarifying the mechanical behavior of disjointed pipes under complex service conditions is of critical significance for targeted rehabilitation. This study focuses on concrete drainage pipes during the initial stage of disjoint, wherein significant erosion of the surrounding soil has not yet developed. The combined effects of traffic loads and groundwater level fluctuations were considered. Full-scale model box tests and 3D refined numerical simulations were conducted to systematically investigate the influence of disjoint on the mechanical behavior of the concrete pipe. Further parametric analyses were conducted to examine the effects of traffic load magnitude, soil cover depth, groundwater level, pipe diameter, and dislocation length on the hoop bending moments at the bell and spigot. The results demonstrate that disjointing induces stress concentration at the bell and spigot joints, with the maximum vertical displacement and hoop bending moment increasing by 12 % and 837 % compared to intact pipes. Increases in traffic load from 0.5 to 1.0 MPa and pipe diameters from 400 to 600 mm significantly amplify the hoop bending moments at both the bell and spigot joints. In contrast, greater soil cover depth and elevated groundwater levels substantially mitigate these moments. Disjoint length has a nonlinear influence, with bell moment peaking and then declining, while the spigot moment continues to rise, reaching a 135 % increase.
{"title":"Mechanical behavior of disjointed concrete pipes under combined traffic loads and groundwater fluctuations","authors":"Peng Xie , Junling Bu , Yulin Yang , Bin Li","doi":"10.1016/j.tust.2025.107414","DOIUrl":"10.1016/j.tust.2025.107414","url":null,"abstract":"<div><div>Frequent road collapses in Guangzhou, China, have been linked to disjoints in underground drainage pipes. Clarifying the mechanical behavior of disjointed pipes under complex service conditions is of critical significance for targeted rehabilitation. This study focuses on concrete drainage pipes during the initial stage of disjoint, wherein significant erosion of the surrounding soil has not yet developed. The combined effects of traffic loads and groundwater level fluctuations were considered. Full-scale model box tests and 3D refined numerical simulations were conducted to systematically investigate the influence of disjoint on the mechanical behavior of the concrete pipe. Further parametric analyses were conducted to examine the effects of traffic load magnitude, soil cover depth, groundwater level, pipe diameter, and dislocation length on the hoop bending moments at the bell and spigot. The results demonstrate that disjointing induces stress concentration at the bell and spigot joints, with the maximum vertical displacement and hoop bending moment increasing by 12 % and 837 % compared to intact pipes. Increases in traffic load from 0.5 to 1.0 MPa and pipe diameters from 400 to 600 mm significantly amplify the hoop bending moments at both the bell and spigot joints. In contrast, greater soil cover depth and elevated groundwater levels substantially mitigate these moments. Disjoint length has a nonlinear influence, with bell moment peaking and then declining, while the spigot moment continues to rise, reaching a 135 % increase.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107414"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840520","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}
Pub Date : 2026-04-01Epub Date: 2025-12-02DOI: 10.1016/j.tust.2025.107336
Ruishan Cheng , Wensu Chen , Cheng Xu , Hong Hao
Ground-borne vibration induced by subway trains in urban areas has caused disturbance to inhabitants along subway lines. Its effective reduction, therefore, has become increasingly important and has attracted more and more attention. Metamaterial/metastructure-based approaches with excellent vibration attenuation capabilities have been recently proposed to mitigate train-induced vibrations around rail tracks. However, owing to the intrinsic difficulties of conventional metastructure designs, effectively mitigating the low-frequency components of train-induced vibrations by installing metastructures around train tracks remains a great challenge. In this study, an innovative design with metaconcrete-filled steel tubular column (MFSTC) is proposed to replace conventional concrete subgrades under rail tracks. The performance of the MFSTC in reducing low-frequency vibrations is first investigated analytically and numerically. It is found that the reduction of the low-frequency vibrations starting near 0 Hz is achievable with proper design of the bottom support of the MFSTC and the shear resistance between the concrete matrix and steel tube of the MFSTC. In addition, the effects of bottom support stiffness and MFSTC configurations on the MFSTC’s vibration reduction capacities are examined. The results show that high shear stiffness between the steel tube and concrete matrix and strong bottom support to the outer steel tube can effectively suppress low-frequency vibrations. A detailed design strategy of the MFSTC aimed at attenuating train vibrations is proposed, and its capability is demonstrated in a case study.
{"title":"Metaconcrete-filled steel tubular column for rail-induced vibration mitigation in tunnels","authors":"Ruishan Cheng , Wensu Chen , Cheng Xu , Hong Hao","doi":"10.1016/j.tust.2025.107336","DOIUrl":"10.1016/j.tust.2025.107336","url":null,"abstract":"<div><div>Ground-borne vibration induced by subway trains in urban areas has caused disturbance to inhabitants along subway lines. Its effective reduction, therefore, has become increasingly important and has attracted more and more attention. Metamaterial/metastructure-based approaches with excellent vibration attenuation capabilities have been recently proposed to mitigate train-induced vibrations around rail tracks. However, owing to the intrinsic difficulties of conventional metastructure designs, effectively mitigating the low-frequency components of train-induced vibrations by installing metastructures around train tracks remains a great challenge. In this study, an innovative design with metaconcrete-filled steel tubular column (MFSTC) is proposed to replace conventional concrete subgrades under rail tracks. The performance of the MFSTC in reducing low-frequency vibrations is first investigated analytically and numerically. It is found that the reduction of the low-frequency vibrations starting near 0 Hz is achievable with proper design of the bottom support of the MFSTC and the shear resistance between the concrete matrix and steel tube of the MFSTC. In addition, the effects of bottom support stiffness and MFSTC configurations on the MFSTC’s vibration reduction capacities are examined. The results show that high shear stiffness between the steel tube and concrete matrix and strong bottom support to the outer steel tube can effectively suppress low-frequency vibrations. A detailed design strategy of the MFSTC aimed at attenuating train vibrations is proposed, and its capability is demonstrated in a case study.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107336"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145651901","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}
Pub Date : 2026-04-01Epub Date: 2025-12-13DOI: 10.1016/j.tust.2025.107369
Chenghao Ye, Jiale Li, Jiaqi Bao, Xuejing Hu, Peihong Zhang
Research on vortex motion enhances the understanding of combustion dynamics and smoke transport patterns in tunnel spill fires under forced ventilation. This study combines model experiment and numerical simulations to examine the fuel layer diffusion process, fire plume characteristics, and smoke transportation of spill fires under longitudinal ventilation. Under forced convection, the temperature distribution on the bottom plate surface is divided into three regions: sub-boiling, boiling, and super-boiling. As longitudinal ventilation velocity increases, the leeward side diffusion length of the sub-boiling zone expands, allowing cold fuel to spread farther. The fuel diffusion pattern shifts from circular to teardrop-shaped, while cylindrical flow around the flame. Asymmetric fluid rotation compresses the leeward flame width, significantly enhancing vorticity in the wake region. Longitudinal ventilation extends the fire plume length but reduces the number of adherent and deflected vortices. It promotes the breakup of large vortex structures, increases small-scale vortices, disrupts continuous heat transfer paths, expands the contact area between hot fluid and cold air, and diminishes buoyant plume strength. Consequently, the fire plume behavior becomes increasingly governed by horizontal inertial forces. Strong airflow dilutes ceiling smoke temperature, reducing the upstream spread speed and distance of hot smoke. This alters airflow circulation, weakens cold-hot mixing, straightens streamlines, and narrows the ceiling temperature rise zone. In the windless environment, vorticity decays along the ceiling toward both sides. Under the weak ventilation, buoyant plumes compete with forced convection, expanding the vorticity distribution, induced flow creates a zero-vorticity zone downstream. At higher ventilation velocity, forced convection dominates, reducing the back-layering length nearly to zero, eliminating upstream vortices, concentrating vorticity downstream, and causing the zero-vorticity zone to disappear.
{"title":"The dynamic combustion process and smoke transport of tunnel spill fire under longitudinal ventilation: the evolution of vortex structures","authors":"Chenghao Ye, Jiale Li, Jiaqi Bao, Xuejing Hu, Peihong Zhang","doi":"10.1016/j.tust.2025.107369","DOIUrl":"10.1016/j.tust.2025.107369","url":null,"abstract":"<div><div>Research on vortex motion enhances the understanding of combustion dynamics and smoke transport patterns in tunnel spill fires under forced ventilation. This study combines model experiment and numerical simulations to examine the fuel layer diffusion process, fire plume characteristics, and smoke transportation of spill fires under longitudinal ventilation. Under forced convection, the temperature distribution on the bottom plate surface is divided into three regions: sub-boiling, boiling, and super-boiling. As longitudinal ventilation velocity increases, the leeward side diffusion length of the sub-boiling zone expands, allowing cold fuel to spread farther. The fuel diffusion pattern shifts from circular to teardrop-shaped, while cylindrical flow around the flame. Asymmetric fluid rotation compresses the leeward flame width, significantly enhancing vorticity in the wake region. Longitudinal ventilation extends the fire plume length but reduces the number of adherent and deflected vortices. It promotes the breakup of large vortex structures, increases small-scale vortices, disrupts continuous heat transfer paths, expands the contact area between hot fluid and cold air, and diminishes buoyant plume strength. Consequently, the fire plume behavior becomes increasingly governed by horizontal inertial forces. Strong airflow dilutes ceiling smoke temperature, reducing the upstream spread speed and distance of hot smoke. This alters airflow circulation, weakens cold-hot mixing, straightens streamlines, and narrows the ceiling temperature rise zone. In the windless environment, vorticity decays along the ceiling toward both sides. Under the weak ventilation, buoyant plumes compete with forced convection, expanding the vorticity distribution, induced flow creates a zero-vorticity zone downstream. At higher ventilation velocity, forced convection dominates, reducing the back-layering length nearly to zero, eliminating upstream vortices, concentrating vorticity downstream, and causing the zero-vorticity zone to disappear.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107369"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145732313","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}
Pub Date : 2026-04-01Epub Date: 2025-12-11DOI: 10.1016/j.tust.2025.107339
Wengang Zhang , Han Han , Weixin Sun , Zhihao Wu , Yang Yang , Yumiao Yan
The formation of mud cake on the cutterhead during shield tunneling is a critical issue that directly affects construction efficiency and operational safety. This phenomenon is governed by a combination of geological conditions and excavation parameters. Traditional approaches typically rely on single-point monitoring data or empirical judgments, which are inadequate for achieving dynamic and accurate prediction of mud cake formation. To address this challenge, this study uses a section of the Shenzhen–Shantou Longgang shield tunnel project as a case study and proposes an integrated approach that combines a Transformer–LSTM deep learning model with real-time shield monitoring and advanced geological prediction. Specifically, real-time excavation parameters collected by the shield monitoring system were combined with geological information obtained from advanced geological prediction to construct a comprehensive dataset containing 41 shield operation parameters and 4 geological parameters. Using the Transformer-LSTM model, the cutterhead mud cake rate was dynamically predicted. The results demonstrate that this approach effectively leverages the complementary strengths of geological forecasting and excavation data, achieving high predictive performance with a coefficient of determination () of 0.986. The proposed method offers a valuable reference for the risk assessment and control of mud cake formation during shield tunneling.
{"title":"Prediction of mud cake formation on shield cutterheads based on multi-source monitoring data integrated with deep learning method","authors":"Wengang Zhang , Han Han , Weixin Sun , Zhihao Wu , Yang Yang , Yumiao Yan","doi":"10.1016/j.tust.2025.107339","DOIUrl":"10.1016/j.tust.2025.107339","url":null,"abstract":"<div><div>The formation of mud cake on the cutterhead during shield tunneling is a critical issue that directly affects construction efficiency and operational safety. This phenomenon is governed by a combination of geological conditions and excavation parameters. Traditional approaches typically rely on single-point monitoring data or empirical judgments, which are inadequate for achieving dynamic and accurate prediction of mud cake formation. To address this challenge, this study uses a section of the Shenzhen–Shantou Longgang shield tunnel project as a case study and proposes an integrated approach that combines a Transformer–LSTM deep learning model with real-time shield monitoring and advanced geological prediction. Specifically, real-time excavation parameters collected by the shield monitoring system were combined with geological information obtained from advanced geological prediction to construct a comprehensive dataset containing 41 shield operation parameters and 4 geological parameters. Using the Transformer-LSTM model, the cutterhead mud cake rate was dynamically predicted. The results demonstrate that this approach effectively leverages the complementary strengths of geological forecasting and excavation data, achieving high predictive performance with a coefficient of determination (<span><math><mrow><msup><mrow><mi>R</mi></mrow><mn>2</mn></msup></mrow></math></span>) of 0.986. The proposed method offers a valuable reference for the risk assessment and control of mud cake formation during shield tunneling.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107339"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145731755","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}
This study presents a comprehensive investigation into the seismic behavior of a double-layer tunnel lining system, focusing on damage progression, failure mechanisms, and load transfer under varying seismic intensities. Using finite element modeling and theoretical formulations, the response of both primary and secondary linings was analyzed in terms of stiffness degradation, plastic hinge formation, crack propagation, dominant frequency shifts, and interface slip behavior. The findings reveal a progression of failure beginning with localized cracking and bending in the primary lining, followed by stiffness degradation and load redistribution to the secondary lining. Three performance levels including Primary Performance Level, Second Performance Level, and Critical Performance Limit, were identified. Interface slip conditions fully bonded, partially bonded, and fully debonded, were shown to significantly affect load transfer efficiency and structural integrity. Full-slip conditions reduce internal forces but increase displacement demands, while rigid connections concentrate stress but enhance structural capacity.
{"title":"Seismic performance of double-layer tunnel linings: a multi-performance-level framework","authors":"Bahram Salehi , Aliakbar Golshani , Jamal Rostami , Barbara Schneider-Muntau","doi":"10.1016/j.tust.2025.107394","DOIUrl":"10.1016/j.tust.2025.107394","url":null,"abstract":"<div><div>This study presents a comprehensive investigation into the seismic behavior of a double-layer tunnel lining system, focusing on damage progression, failure mechanisms, and load transfer under varying seismic intensities. Using finite element modeling and theoretical formulations, the response of both primary and secondary linings was analyzed in terms of stiffness degradation, plastic hinge formation, crack propagation, dominant frequency shifts, and interface slip behavior. The findings reveal a progression of failure beginning with localized cracking and bending in the primary lining, followed by stiffness degradation and load redistribution to the secondary lining. Three performance levels including Primary Performance Level, Second Performance Level, and Critical Performance Limit, were identified. Interface slip conditions fully bonded, partially bonded, and fully debonded, were shown to significantly affect load transfer efficiency and structural integrity. Full-slip conditions reduce internal forces but increase displacement demands, while rigid connections concentrate stress but enhance structural capacity.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107394"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145814020","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}
Pub Date : 2026-04-01Epub Date: 2025-12-17DOI: 10.1016/j.tust.2025.107379
Zhiyao Feng , Shuying Wang , Yufeng Shi , Yalong Jiang , Xiangsheng Chen , Tongming Qu
The seepage behaviour of excavated sand becomes increasingly complex due to foam injection. In this investigation, the impact of soil pressure on the hydraulic conductivity of sand-foam mixtures is examined through a newly developed permeameter. The findings show that the hydraulic conductivity is conspicuously affected by the soil pressure, and this influence largely relies on the amount of foam bubbles. Specifically, the initial hydraulic conductivity undergoes a sharp decrease with increasing soil pressure, followed by a gradual decline until it reaches a stable phase, while the initial stabilization duration increases with soil pressure. The blocking capacity and stability of the water-blocking structure are notably enhanced under soil pressure due to soil skeleton compression. This implies that standard permeability tests may overstate the permeability of sand-foam mixtures owing to the omission of soil pressure, potentially causing overdosing of chemical additives. Moreover, an effective void ratio calculation model is proposed to quantitatively characterize the ability of foam bubbles to obstruct seepage channels. The mechanisms through which soil pressure affects the mixture’s seepage behaviour are also revealed by balloon compression tests.
{"title":"Experimental investigation on permeability of sand-foam mixture under soil pressure in mechanized tunnelling","authors":"Zhiyao Feng , Shuying Wang , Yufeng Shi , Yalong Jiang , Xiangsheng Chen , Tongming Qu","doi":"10.1016/j.tust.2025.107379","DOIUrl":"10.1016/j.tust.2025.107379","url":null,"abstract":"<div><div>The seepage behaviour of excavated sand becomes increasingly complex due to foam injection. In this investigation, the impact of soil pressure on the hydraulic conductivity of sand-foam mixtures is examined through a newly developed permeameter. The findings show that the hydraulic conductivity is conspicuously affected by the soil pressure, and this influence largely relies on the amount of foam bubbles. Specifically, the initial hydraulic conductivity undergoes a sharp decrease with increasing soil pressure, followed by a gradual decline until it reaches a stable phase, while the initial stabilization duration increases with soil pressure. The blocking capacity and stability of the water-blocking structure are notably enhanced under soil pressure due to soil skeleton compression. This implies that standard permeability tests may overstate the permeability of sand-foam mixtures owing to the omission of soil pressure, potentially causing overdosing of chemical additives. Moreover, an effective void ratio calculation model is proposed to quantitatively characterize the ability of foam bubbles to obstruct seepage channels. The mechanisms through which soil pressure affects the mixture’s seepage behaviour are also revealed by balloon compression tests.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107379"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791365","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}
Pub Date : 2026-04-01Epub Date: 2025-12-18DOI: 10.1016/j.tust.2025.107377
Yuxuan Wang , Shengrong Xie , Yiyi Wu , Chenyang Liu , Jiaqi Hou
Aiming at the problem that superimposed surrounding rock stress during ultra-large mining height working face (ULMHWF) passing through the abandoned roadway (AR) is prone to inducing intense dynamic disasters, and existing evaluation methods fail to achieve full-process dynamic quantification of damage, this study constructed a dynamic evaluation system for surrounding rock damage based on the failure approaching index (FAI), whose accuracy was verified through experiments, numerical simulations, and on-site engineering applications. Uniaxial compression test results indicate that microcracks in low-strength rock masses are more susceptible to activation and damage, forming multiple sets of macroscopic cracks with low energy release intensity. A 3D numerical model with rigorously calibrated constitutive parameters was established, and a FISH-language calculation module based on FAI was developed to unveil the time-varying evolution characteristics of the damage zone as the distance advances. Specifically, ULMHWF excavation induces a “>”-shaped stress peak zone ahead of the coal wall; during advancement from 50 m to 5 m, the FAI peak increases by 45.6 %, and its distribution characteristics are consistent with those of the plastic zone. In addition, visual characterization was realized via a MATLAB-based 3D reconstruction algorithm. On-site borehole observation results show that the failure depth predicted by FAI is highly consistent with the measured data. Based on this, multi-dimensional synergistic reinforcement support measures and a four-stage optimized technology for passing through the AR were proposed, achieving safe and efficient mining successfully. This study confirms the engineering application value of FAI, providing a more effective method for quantitative evaluation of surrounding rock damage in underground engineering.
{"title":"A computational framework for dynamic quantitative assessment of surrounding rock damage based on failure approaching index in underground construction","authors":"Yuxuan Wang , Shengrong Xie , Yiyi Wu , Chenyang Liu , Jiaqi Hou","doi":"10.1016/j.tust.2025.107377","DOIUrl":"10.1016/j.tust.2025.107377","url":null,"abstract":"<div><div>Aiming at the problem that superimposed surrounding rock stress during ultra-large mining height working face (ULMHWF) passing through the abandoned roadway (AR) is prone to inducing intense dynamic disasters, and existing evaluation methods fail to achieve full-process dynamic quantification of damage, this study constructed a dynamic evaluation system for surrounding rock damage based on the failure approaching index (<em>FAI</em>), whose accuracy was verified through experiments, numerical simulations, and on-site engineering applications. Uniaxial compression test results indicate that microcracks in low-strength rock masses are more susceptible to activation and damage, forming multiple sets of macroscopic cracks with low energy release intensity. A 3D numerical model with rigorously calibrated constitutive parameters was established, and a <em>FISH</em>-language calculation module based on <em>FAI</em> was developed to unveil the time-varying evolution characteristics of the damage zone as the distance advances. Specifically, ULMHWF excavation induces a “>”-shaped stress peak zone ahead of the coal wall; during advancement from 50 m to 5 m, the <em>FAI</em> peak increases by 45.6 %, and its distribution characteristics are consistent with those of the plastic zone. In addition, visual characterization was realized via a MATLAB-based 3D reconstruction algorithm. On-site borehole observation results show that the failure depth predicted by <em>FAI</em> is highly consistent with the measured data. Based on this, multi-dimensional synergistic reinforcement support measures and a four-stage optimized technology for passing through the AR were proposed, achieving safe and efficient mining successfully. This study confirms the engineering application value of <em>FAI</em>, providing a more effective method for quantitative evaluation of surrounding rock damage in underground engineering.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107377"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786033","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}
Pub Date : 2026-04-01Epub Date: 2025-12-02DOI: 10.1016/j.tust.2025.107334
Xun Wu , Jun He , Kan Huang , Xiangsheng Chen , Sidong Feng , Bin Huang
As a key component of urban rail system, shield tunnels are susceptible to cracking and stiffness degradation, from which an urgent need for effective reinforcement arises. In this study, grouted channel steel (GCS) with bolted connections was assessed as a reinforcement strategy for damaged tunnel segments. Full-scale tests integrated with finite element simulations verified the reliability of the proposed modelling approach, as strong agreement was observed between simulated and experimental load–displacement responses and crack development. The deformation of reinforced segments proceeded through three stages: elastic, strengthening, and failure. The effectiveness of reinforcement was governed by the level of prior damage. Elastic stiffness declined progressively with increasing damage, whereas strengthening-stage stiffness remained stable. Stiffness enhancement was concentrated in the strengthening stage, reaching values up to sixteen times those of the elastic stage. Under service limit conditions, balanced stiffness improvements of 95–120 % were achieved across stages, and recovery after unloading reached 140–160 %. These results indicate that GCS reinforcement provides a robust means of restoring and extending the service life of shield tunnel segments in urban rail systems.
{"title":"Experimental and numerical investigation of shield tunnel segments reinforced with grouted channel steel under diverse damage scenarios","authors":"Xun Wu , Jun He , Kan Huang , Xiangsheng Chen , Sidong Feng , Bin Huang","doi":"10.1016/j.tust.2025.107334","DOIUrl":"10.1016/j.tust.2025.107334","url":null,"abstract":"<div><div>As a key component of urban rail system, shield tunnels are susceptible to cracking and stiffness degradation, from which an urgent need for effective reinforcement arises. In this study, grouted channel steel (GCS) with bolted connections was assessed as a reinforcement strategy for damaged tunnel segments. Full-scale tests integrated with finite element simulations verified the reliability of the proposed modelling approach, as strong agreement was observed between simulated and experimental load–displacement responses and crack development. The deformation of reinforced segments proceeded through three stages: elastic, strengthening, and failure. The effectiveness of reinforcement was governed by the level of prior damage. Elastic stiffness declined progressively with increasing damage, whereas strengthening-stage stiffness remained stable. Stiffness enhancement was concentrated in the strengthening stage, reaching values up to sixteen times those of the elastic stage. Under service limit conditions, balanced stiffness improvements of 95–120 % were achieved across stages, and recovery after unloading reached 140–160 %. These results indicate that GCS reinforcement provides a robust means of restoring and extending the service life of shield tunnel segments in urban rail systems.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107334"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145651896","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}
Pub Date : 2026-04-01Epub Date: 2026-01-02DOI: 10.1016/j.tust.2025.107424
Jinghuan Pan, Hang Lin, Jinbiao Wu
Deep learning models for TBM geological prediction suffer from two critical limitations: they provide only single-point estimates, which are unreliable when encountering unknown conditions, and they struggle with the multi-source heterogeneity between continuous operational data and discrete geological information. To address these challenges, this study proposes an uncertainty quantification framework that integrates a Gaussian Mixture Model (GMM) with a Probabilistic Bayesian Convolutional Neural Network (PBCNN). First, the GMM clusters TBM operational parameters to establish a data-driven labeling system, effectively resolving the data heterogeneity. The optimal number of clusters is determined using internal evaluation metrics. Subsequently, a PBCNN architecture, tailored for tabular time-series data, is constructed to decompose the model’s predictive uncertainty into its epistemic and aleatoric components. The framework’s effectiveness was validated using field data from the Bainikeng Station section of the Shenzhen-Dayawan Intercity Railway project. Four typical operational scenarios were designed—Unknown Geological Conditions, Changing Operational Status, Data Proportion Fluctuation, and Different Noise Environments—for systematic testing. The results demonstrate that our method not only achieves high predictive accuracy but also provides a quantitative assessment of prediction credibility and risk. This establishes a more robust and generalizable paradigm for geological identification and intelligent decision-making in TBM construction.
{"title":"Uncertainty-aware geological prediction in TBM tunneling: A probabilistic bayesian framework with exploratory multi-source label construction","authors":"Jinghuan Pan, Hang Lin, Jinbiao Wu","doi":"10.1016/j.tust.2025.107424","DOIUrl":"10.1016/j.tust.2025.107424","url":null,"abstract":"<div><div>Deep learning models for TBM geological prediction suffer from two critical limitations: they provide only single-point estimates, which are unreliable when encountering unknown conditions, and they struggle with the multi-source heterogeneity between continuous operational data and discrete geological information. To address these challenges, this study proposes an uncertainty quantification framework that integrates a Gaussian Mixture Model (GMM) with a Probabilistic Bayesian Convolutional Neural Network (PBCNN). First, the GMM clusters TBM operational parameters to establish a data-driven labeling system, effectively resolving the data heterogeneity. The optimal number of clusters is determined using internal evaluation metrics. Subsequently, a PBCNN architecture, tailored for tabular time-series data, is constructed to decompose the model’s predictive uncertainty into its epistemic and aleatoric components. The framework’s effectiveness was validated using field data from the Bainikeng Station section of the Shenzhen-Dayawan Intercity Railway project. Four typical operational scenarios were designed—Unknown Geological Conditions, Changing Operational Status, Data Proportion Fluctuation, and Different Noise Environments—for systematic testing. The results demonstrate that our method not only achieves high predictive accuracy but also provides a quantitative assessment of prediction credibility and risk. This establishes a more robust and generalizable paradigm for geological identification and intelligent decision-making in TBM construction.</div></div>","PeriodicalId":49414,"journal":{"name":"Tunnelling and Underground Space Technology","volume":"170 ","pages":"Article 107424"},"PeriodicalIF":7.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884067","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号