The random finite difference method(RFDM) is a popular approach to quantitatively evaluate the influence of inherent spatial variability of soil on the deformation of embedded tunnels.However,the high computational co...The random finite difference method(RFDM) is a popular approach to quantitatively evaluate the influence of inherent spatial variability of soil on the deformation of embedded tunnels.However,the high computational cost is an ongoing challenge for its application in complex scenarios.To address this limitation,a deep learning-based method for efficient prediction of tunnel deformation in spatially variable soil is proposed.The proposed method uses one-dimensional convolutional neural network(CNN) to identify the pattern between random field input and factor of safety of tunnel deformation output.The mean squared error and correlation coefficient of the CNN model applied to the newly untrained dataset was less than 0.02 and larger than 0.96,respectively.It means that the trained CNN model can replace RFDM analysis for Monte Carlo simulations with a small but sufficient number of random field samples(about 40 samples for each case in this study).It is well known that the machine learning or deep learning model has a common limitation that the confidence of predicted result is unknown and only a deterministic outcome is given.This calls for an approach to gauge the model’s confidence interval.It is achieved by applying dropout to all layers of the original model to retrain the model and using the dropout technique when performing inference.The excellent agreement between the CNN model prediction and the RFDM calculated results demonstrated that the proposed deep learning-based method has potential for tunnel performance analysis in spatially variable soils.展开更多
This study presents a practical design strategy for a large-size Submerged Floating Tunnel(SFT)under different target environments through global-performance simulations.A coupled time-domain simulation model for SFT ...This study presents a practical design strategy for a large-size Submerged Floating Tunnel(SFT)under different target environments through global-performance simulations.A coupled time-domain simulation model for SFT is established to check hydro-elastic behaviors under the design random wave and earthquake excitations.The tunnel and mooring lines are modeled with a finite-element line model based on a series of lumped masses connected by axial,bending,and torsional springs,and thus the dynamic/structural deformability of the entire SFT is fully considered.The dummy-connection-mass method and constraint boundary conditions are employed to connect the tunnel and mooring lines in a convenient manner.Wave-and earthquake-induced hydrodynamic forces are evaluated by the Morison equation at instantaneous node positions.Several wave and earthquake conditions are selected to evaluate its global performance and sensitivity at different system parameters.Different BuoyancyWeight Ratios(BWRs),submergence depths,and tunnel lengths(and mooring intervals)are chosen to establish a design strategy for reducing the maximum mooring tension.Both static and dynamic tensions are critical to find an acceptable design depending on the given target environmental condition.BWR plays a crucial role in preventing snap loading,and the corresponding static tension is a primary factor if the environmental condition is mild.The tunnel length can significantly be extended by reducing BWR when environmental force is not that substantial.Dynamic tension becomes important in harsh environmental conditions,for which high BWR and short mooring interval are required.It is underscored that the wet natural frequencies with mooring are located away from the spectral peaks of design waves or earthquakes.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.52130805 and 52022070)Shanghai Science and Technology Committee Program(Grant No.20dz1202200)。
文摘The random finite difference method(RFDM) is a popular approach to quantitatively evaluate the influence of inherent spatial variability of soil on the deformation of embedded tunnels.However,the high computational cost is an ongoing challenge for its application in complex scenarios.To address this limitation,a deep learning-based method for efficient prediction of tunnel deformation in spatially variable soil is proposed.The proposed method uses one-dimensional convolutional neural network(CNN) to identify the pattern between random field input and factor of safety of tunnel deformation output.The mean squared error and correlation coefficient of the CNN model applied to the newly untrained dataset was less than 0.02 and larger than 0.96,respectively.It means that the trained CNN model can replace RFDM analysis for Monte Carlo simulations with a small but sufficient number of random field samples(about 40 samples for each case in this study).It is well known that the machine learning or deep learning model has a common limitation that the confidence of predicted result is unknown and only a deterministic outcome is given.This calls for an approach to gauge the model’s confidence interval.It is achieved by applying dropout to all layers of the original model to retrain the model and using the dropout technique when performing inference.The excellent agreement between the CNN model prediction and the RFDM calculated results demonstrated that the proposed deep learning-based method has potential for tunnel performance analysis in spatially variable soils.
基金This work was supported by the National Research Foundation of Korea(NRF)Grant funded by the Korean Government(MSIT)(No.2017R1A5A1014883).
文摘This study presents a practical design strategy for a large-size Submerged Floating Tunnel(SFT)under different target environments through global-performance simulations.A coupled time-domain simulation model for SFT is established to check hydro-elastic behaviors under the design random wave and earthquake excitations.The tunnel and mooring lines are modeled with a finite-element line model based on a series of lumped masses connected by axial,bending,and torsional springs,and thus the dynamic/structural deformability of the entire SFT is fully considered.The dummy-connection-mass method and constraint boundary conditions are employed to connect the tunnel and mooring lines in a convenient manner.Wave-and earthquake-induced hydrodynamic forces are evaluated by the Morison equation at instantaneous node positions.Several wave and earthquake conditions are selected to evaluate its global performance and sensitivity at different system parameters.Different BuoyancyWeight Ratios(BWRs),submergence depths,and tunnel lengths(and mooring intervals)are chosen to establish a design strategy for reducing the maximum mooring tension.Both static and dynamic tensions are critical to find an acceptable design depending on the given target environmental condition.BWR plays a crucial role in preventing snap loading,and the corresponding static tension is a primary factor if the environmental condition is mild.The tunnel length can significantly be extended by reducing BWR when environmental force is not that substantial.Dynamic tension becomes important in harsh environmental conditions,for which high BWR and short mooring interval are required.It is underscored that the wet natural frequencies with mooring are located away from the spectral peaks of design waves or earthquakes.
