Contemporary demands necessitate the swift and accurate detection of cracks in critical infrastructures,including tunnels and pavements.This study proposed a transfer learning-based encoder-decoder method with visual ...Contemporary demands necessitate the swift and accurate detection of cracks in critical infrastructures,including tunnels and pavements.This study proposed a transfer learning-based encoder-decoder method with visual explanations for infrastructure crack segmentation.Firstly,a vast dataset containing 7089 images was developed,comprising diverse conditions—simple and complex crack patterns as well as clean and rough backgrounds.Secondly,leveraging transfer learning,an encoder-decoder model with visual explanations was formulated,utilizing varied pre-trained convolutional neural network(CNN)as the encoder.Visual explanations were achieved through gradient-weighted class activation mapping(Grad-CAM)to interpret the CNN segmentation model.Thirdly,accuracy,complexity(computation and model),and memory usage assessed CNN feasibility in practical engineering.Model performance was gauged via prediction and visual explanation.The investigation encompassed hyperparameters,data augmentation,deep learning from scratch vs.transfer learning,segmentation model architectures,segmentation model encoders,and encoder pre-training strategies.Results underscored transfer learning’s potency in enhancing CNN accuracy for crack segmentation,surpassing deep learning from scratch.Notably,encoder classification accuracy bore no significant correlation with CNN segmentation accuracy.Among all tested models,UNet-EfficientNet_B7 excelled in crack segmentation,harmonizing accuracy,complexity,memory usage,prediction,and visual explanation.展开更多
A circular shaft is often used to access a working well for deep underground space utilization.As the depth of underground space increases,the excavation depth of the shaft increases.In this study,the deformation char...A circular shaft is often used to access a working well for deep underground space utilization.As the depth of underground space increases,the excavation depth of the shaft increases.In this study,the deformation characteristics of a circular shaft with a depth of 56.3 m were presented and analysed.The main monitoring contents included:(1)wall deflection;(2)vertical wall movement;(3)horizontal soil movement;(4)vertical surface movement;and(5)basal heave.Horizontally,the maximum wall deflection was only 7.7 mm.Compared with the wall deflection data collected for another 29 circular excavations,the ratio of maximum wall deflection to excavation depth of this shaft was smaller due to a smaller ratio of diameter to excavation depth.The wall deflection underwent two stages of deformation:the first stage was mainly circumferential compression caused by the mutual extrusion of joints between walls,and the second stage was typical vertical deflection deformation.The horizontal soil movement outside the shaft was greater than the wall deflection and the deep soil caused great horizontal movement because of dewatering at confined water layers.Vertically,a basal heave of 203.8 mm occurred in the pit centre near the bottom.Meanwhile,the shaft was uplifted over time and showed 3 stages of vertical movement.The surface outside the shaft exhibited settlement and uplift deformation at different locations due to different effects.The basal heave caused by excavation was the dominant factor,driving the vertical movement of the shaft as well as the surrounding surface.The correlation between the wall deflection and the surface settlement outside the shaft was weak.展开更多
Corrugated steel–concrete(CSC)composite structures are increasingly used in tunnel and culvert projects due to their good mechanical properties.The design of CSC composite structures is often governed by deflection l...Corrugated steel–concrete(CSC)composite structures are increasingly used in tunnel and culvert projects due to their good mechanical properties.The design of CSC composite structures is often governed by deflection limits in service,hence it becomes crucial to evaluate accurately their flexural stiffness.In this work,the deflection deformation mechanism of CSC composite structure is studied by experimental and numerical methods,and a simplified formula for calculating the flexural stiffness is established.In addition,the deflection results obtained by different methods are compared and analyzed.It is found that:(1)the flexural stiffness of the CSC composite structure is constant only when the load is small,and after the bending moment exceeds a certain value,the flexural stiffness will gradually become smaller as the bending moment increases.(2)The value of the bending moment corresponding to the end of the elastic stage of the bending moment-deflection curve increases with the increase of the axial force in the composite structure.(3)As the axial force of the composite structure increases,the flexural bearing capacity of the structure increases first and then decreases.展开更多
The tensile behavior of hybrid fiber reinforced concrete(HFRC)is important to the design of HFRC and HFRC structure.This study used an artificial neural network(ANN)model to describe the tensile behavior of HFRC.This ...The tensile behavior of hybrid fiber reinforced concrete(HFRC)is important to the design of HFRC and HFRC structure.This study used an artificial neural network(ANN)model to describe the tensile behavior of HFRC.This ANN model can describe well the tensile stress-strain curve of HFRC with the consideration of 23 features of HFRC.In the model,three methods to process output features(no-processed,mid-processed,and processed)are discussed and the mid-processed method is recommended to achieve a better reproduction of the experimental data.This means the strain should be normalized while the stress doesn’t need normalization.To prepare the database of the model,both many direct tensile test results and the relevant literature data are collected.Moreover,a traditional equation-based model is also established and compared with the ANN model.The results show that the ANN model has a better prediction than the equation-based model in terms of the tensile stress-strain curve,tensile strength,and strain corresponding to tensile strength of HFRC.Finally,the sensitivity analysis of the ANN model is also performed to analyze the contribution of each input feature to the tensile strength and strain corresponding to tensile strength.The mechanical properties of plain concrete make the main contribution to the tensile strength and strain corresponding to tensile strength,while steel fibers tend to make more contributions to these two items than PVA fibers.展开更多
To improve the understanding on the failure behavior and its anchoring effect of weak-broken rock slope,the rock of grade IV according to China is taken as reference prototype,and a series of model tests were carried ...To improve the understanding on the failure behavior and its anchoring effect of weak-broken rock slope,the rock of grade IV according to China is taken as reference prototype,and a series of model tests were carried out in laboratory.These tests can be divided into two categories,that is,with bolt reinforcement and without bolt reinforcement.In which,the stability of slope reinforced with different bolt diameter,different anchor length and different space are studied.The test results show that the collapse of slope is the combination of tension failure at the top and the compression-shearing failure at the bottom of the slope,and its failure process presents progressive characteristics.The contributions of bolt reinforcement are mainly reflected by the aspects of shear resistance,crack resistance and anti-extension.The reinforcement of blot not only can improve the vertical bearing capacity before failure,but also can reduce the vertical settlement and allow greater lateral rock wall deformation;what is more,the stress concentration degree in rock mass can be dispersed,which do help to improve the stability of slope rock mass.展开更多
The waterproofing capacity of segmental joints is an essential indicator for the long-term service performance of shield tunnels.The segmental joints with double gaskets have been adopted to improve the waterproof per...The waterproofing capacity of segmental joints is an essential indicator for the long-term service performance of shield tunnels.The segmental joints with double gaskets have been adopted to improve the waterproof performance for certain tunnels with high water pressure,especially water-conveyance tunnels.This paper presents a series of waterproofing tests on joints with different settings of double gaskets via an improved test apparatus.Four cases are investigated,including(i)double gaskets with equal high/low waterproof capacity,(ii)the outer gasket with higher waterproof capacity,and(iii)the outer gasket with lower waterproof capacity.Different joint openings are also considered.Moreover,the water pressures at two cavities are continuously monitored to highlight the waterproofing mechanism and failure pattern of double gaskets.The results demonstrate that the gasket with the higher waterproofing capacity dominates the overall waterproofing capacity of joint with double gaskets;only a little enhancement of the waterproofing capacity is found for the joint with the same double gaskets.The waterproofing capacity of joints with double gaskets decreases with the increase of the joint opening.The failure pattern depends on the arrangement of double gaskets,and four stages can be identified during the whole failure process.The advantages of double gaskets in case of joint rotation are also discussed.展开更多
In this study,a deep excavation in an under-consolidated deposit in Zhuhai,China,was reported and investigated via plane strain finite element analysis(FEA).First,the project was simulated via FEA(under-consolidated d...In this study,a deep excavation in an under-consolidated deposit in Zhuhai,China,was reported and investigated via plane strain finite element analysis(FEA).First,the project was simulated via FEA(under-consolidated deposit),and a reasonable agreement between the lateral displacement of the measured and simulated retaining wall was obtained.Another FEA was then conducted under the assumption that the deposit was in a normally consolidated state.The numerical results indicate that the under-consolidated case resulted in a 25% increase in maximum lateral displacement of the contiguous pile-formed retaining wall,a 32% increase in bending moment in the wall,and approximately twice the maximum surface settlement behind the wall,when compared with those of the normally consolidated case.The main reasons for this are as follows:(1)the under-consolidated deposit was weaker,and(2)the ongoing consolidation of the under-consolidated deposit induced green-field settlement(approximately 4 mm)during the project period,thereby enhancing the bending deformation of the wall.Therefore,when designing deep excavation in an under-consolidated deposit,not only its weaker strength but also the negative effect of green-field settlement during the project period should be considered.展开更多
基金the National Natural Science Foundation of China(Grant Nos.52090083 and 52378405)Key Technology R&D Plan of Yunnan Provincial Department of Science and Technology(Grant No.202303AA080003)for their financial support.
文摘Contemporary demands necessitate the swift and accurate detection of cracks in critical infrastructures,including tunnels and pavements.This study proposed a transfer learning-based encoder-decoder method with visual explanations for infrastructure crack segmentation.Firstly,a vast dataset containing 7089 images was developed,comprising diverse conditions—simple and complex crack patterns as well as clean and rough backgrounds.Secondly,leveraging transfer learning,an encoder-decoder model with visual explanations was formulated,utilizing varied pre-trained convolutional neural network(CNN)as the encoder.Visual explanations were achieved through gradient-weighted class activation mapping(Grad-CAM)to interpret the CNN segmentation model.Thirdly,accuracy,complexity(computation and model),and memory usage assessed CNN feasibility in practical engineering.Model performance was gauged via prediction and visual explanation.The investigation encompassed hyperparameters,data augmentation,deep learning from scratch vs.transfer learning,segmentation model architectures,segmentation model encoders,and encoder pre-training strategies.Results underscored transfer learning’s potency in enhancing CNN accuracy for crack segmentation,surpassing deep learning from scratch.Notably,encoder classification accuracy bore no significant correlation with CNN segmentation accuracy.Among all tested models,UNet-EfficientNet_B7 excelled in crack segmentation,harmonizing accuracy,complexity,memory usage,prediction,and visual explanation.
基金the National Natural Science Foundation of China(Grant No.52090083)Shanghai Chenguang Program(Grant No.20CG26)Consulting project on Shanghai deep tunnel project.
文摘A circular shaft is often used to access a working well for deep underground space utilization.As the depth of underground space increases,the excavation depth of the shaft increases.In this study,the deformation characteristics of a circular shaft with a depth of 56.3 m were presented and analysed.The main monitoring contents included:(1)wall deflection;(2)vertical wall movement;(3)horizontal soil movement;(4)vertical surface movement;and(5)basal heave.Horizontally,the maximum wall deflection was only 7.7 mm.Compared with the wall deflection data collected for another 29 circular excavations,the ratio of maximum wall deflection to excavation depth of this shaft was smaller due to a smaller ratio of diameter to excavation depth.The wall deflection underwent two stages of deformation:the first stage was mainly circumferential compression caused by the mutual extrusion of joints between walls,and the second stage was typical vertical deflection deformation.The horizontal soil movement outside the shaft was greater than the wall deflection and the deep soil caused great horizontal movement because of dewatering at confined water layers.Vertically,a basal heave of 203.8 mm occurred in the pit centre near the bottom.Meanwhile,the shaft was uplifted over time and showed 3 stages of vertical movement.The surface outside the shaft exhibited settlement and uplift deformation at different locations due to different effects.The basal heave caused by excavation was the dominant factor,driving the vertical movement of the shaft as well as the surrounding surface.The correlation between the wall deflection and the surface settlement outside the shaft was weak.
基金funded by Shanghai 2020“Science and Technology Innovation Action Plan”Project,China(Grant No.20dz1202600)National Natural Science Foundation of China(Grant No.52090083)+1 种基金Science and Technology Innovation and Demonstration Project of Department of Transport of Yunnan Province,China(Grant No.[2020]106)The authors gratefully acknowledge their financial supports.
文摘Corrugated steel–concrete(CSC)composite structures are increasingly used in tunnel and culvert projects due to their good mechanical properties.The design of CSC composite structures is often governed by deflection limits in service,hence it becomes crucial to evaluate accurately their flexural stiffness.In this work,the deflection deformation mechanism of CSC composite structure is studied by experimental and numerical methods,and a simplified formula for calculating the flexural stiffness is established.In addition,the deflection results obtained by different methods are compared and analyzed.It is found that:(1)the flexural stiffness of the CSC composite structure is constant only when the load is small,and after the bending moment exceeds a certain value,the flexural stiffness will gradually become smaller as the bending moment increases.(2)The value of the bending moment corresponding to the end of the elastic stage of the bending moment-deflection curve increases with the increase of the axial force in the composite structure.(3)As the axial force of the composite structure increases,the flexural bearing capacity of the structure increases first and then decreases.
基金The authors would like to acknowledge the National Natural Science Foundation of China(Grant Nos.51978515,41941018)Shanghai Sailing Program(19YF1451400)Shanghai Municipal Science and Technology Major Project(2017SHZDZX02)for their financial support.
文摘The tensile behavior of hybrid fiber reinforced concrete(HFRC)is important to the design of HFRC and HFRC structure.This study used an artificial neural network(ANN)model to describe the tensile behavior of HFRC.This ANN model can describe well the tensile stress-strain curve of HFRC with the consideration of 23 features of HFRC.In the model,three methods to process output features(no-processed,mid-processed,and processed)are discussed and the mid-processed method is recommended to achieve a better reproduction of the experimental data.This means the strain should be normalized while the stress doesn’t need normalization.To prepare the database of the model,both many direct tensile test results and the relevant literature data are collected.Moreover,a traditional equation-based model is also established and compared with the ANN model.The results show that the ANN model has a better prediction than the equation-based model in terms of the tensile stress-strain curve,tensile strength,and strain corresponding to tensile strength of HFRC.Finally,the sensitivity analysis of the ANN model is also performed to analyze the contribution of each input feature to the tensile strength and strain corresponding to tensile strength.The mechanical properties of plain concrete make the main contribution to the tensile strength and strain corresponding to tensile strength,while steel fibers tend to make more contributions to these two items than PVA fibers.
基金supported by the National Natural Science Foundation of China(Grant No.40672184)the Open Foundation of the Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education of China.
文摘To improve the understanding on the failure behavior and its anchoring effect of weak-broken rock slope,the rock of grade IV according to China is taken as reference prototype,and a series of model tests were carried out in laboratory.These tests can be divided into two categories,that is,with bolt reinforcement and without bolt reinforcement.In which,the stability of slope reinforced with different bolt diameter,different anchor length and different space are studied.The test results show that the collapse of slope is the combination of tension failure at the top and the compression-shearing failure at the bottom of the slope,and its failure process presents progressive characteristics.The contributions of bolt reinforcement are mainly reflected by the aspects of shear resistance,crack resistance and anti-extension.The reinforcement of blot not only can improve the vertical bearing capacity before failure,but also can reduce the vertical settlement and allow greater lateral rock wall deformation;what is more,the stress concentration degree in rock mass can be dispersed,which do help to improve the stability of slope rock mass.
基金financially supported by National Natural Science Foundation of China(52090083)Shanghai Committee of Science and Technology(20DZ1202600)China Railway Southern Investment Group Co.Ltd(Project No.ZTNF-2020-1).
文摘The waterproofing capacity of segmental joints is an essential indicator for the long-term service performance of shield tunnels.The segmental joints with double gaskets have been adopted to improve the waterproof performance for certain tunnels with high water pressure,especially water-conveyance tunnels.This paper presents a series of waterproofing tests on joints with different settings of double gaskets via an improved test apparatus.Four cases are investigated,including(i)double gaskets with equal high/low waterproof capacity,(ii)the outer gasket with higher waterproof capacity,and(iii)the outer gasket with lower waterproof capacity.Different joint openings are also considered.Moreover,the water pressures at two cavities are continuously monitored to highlight the waterproofing mechanism and failure pattern of double gaskets.The results demonstrate that the gasket with the higher waterproofing capacity dominates the overall waterproofing capacity of joint with double gaskets;only a little enhancement of the waterproofing capacity is found for the joint with the same double gaskets.The waterproofing capacity of joints with double gaskets decreases with the increase of the joint opening.The failure pattern depends on the arrangement of double gaskets,and four stages can be identified during the whole failure process.The advantages of double gaskets in case of joint rotation are also discussed.
基金financially sponsored by National Foreign Experts and Talents Project of China(G20190009128)the Consulting Project on Hengqin Foundation Pits(CSCEC2B-SZ-HQKA-ZY-080)Shanghai Municipal Science and Technology Major Project(2017SHZDZX02),China.
文摘In this study,a deep excavation in an under-consolidated deposit in Zhuhai,China,was reported and investigated via plane strain finite element analysis(FEA).First,the project was simulated via FEA(under-consolidated deposit),and a reasonable agreement between the lateral displacement of the measured and simulated retaining wall was obtained.Another FEA was then conducted under the assumption that the deposit was in a normally consolidated state.The numerical results indicate that the under-consolidated case resulted in a 25% increase in maximum lateral displacement of the contiguous pile-formed retaining wall,a 32% increase in bending moment in the wall,and approximately twice the maximum surface settlement behind the wall,when compared with those of the normally consolidated case.The main reasons for this are as follows:(1)the under-consolidated deposit was weaker,and(2)the ongoing consolidation of the under-consolidated deposit induced green-field settlement(approximately 4 mm)during the project period,thereby enhancing the bending deformation of the wall.Therefore,when designing deep excavation in an under-consolidated deposit,not only its weaker strength but also the negative effect of green-field settlement during the project period should be considered.