A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Reliability analysis of tunnels considering the rock-support interaction using a metamodel-based reliability method
Abstract The convergence-confinement method (CCM) is an important approach to analysis the rock-support interaction and to design tunnel and its support. To make an effort on efficient reliability analysis of tunnels integrated with CCM, of which performance functions are mostly highly nonlinear and implicit, a hybrid intelligent algorithm combing uniform design (UD) and Gaussian process regression (GPR) is proposed. Based on the generated UD samples, GPR-parameter metamodels to predict the important parameters of CCM and GPR-probability metamodels to approximate the real performance functions are constructed consequently. Then sensitivity analysis, determination of the optimal preliminary support timing and failure probabilities calculations can be performed in conjunction with Monte Carlo simulation (MCS) based on the constructed GPR-parameter metamodels and GPR-probability metamodels. To verify the efficiency and accuracy of the proposed method, a typical circle tunnel is analyzed and good results are achieved. Finally, a practical hydraulic tunnel is analyzed using the proposed method and some beneficial conclusions are drawn.
Reliability analysis of tunnels considering the rock-support interaction using a metamodel-based reliability method
Abstract The convergence-confinement method (CCM) is an important approach to analysis the rock-support interaction and to design tunnel and its support. To make an effort on efficient reliability analysis of tunnels integrated with CCM, of which performance functions are mostly highly nonlinear and implicit, a hybrid intelligent algorithm combing uniform design (UD) and Gaussian process regression (GPR) is proposed. Based on the generated UD samples, GPR-parameter metamodels to predict the important parameters of CCM and GPR-probability metamodels to approximate the real performance functions are constructed consequently. Then sensitivity analysis, determination of the optimal preliminary support timing and failure probabilities calculations can be performed in conjunction with Monte Carlo simulation (MCS) based on the constructed GPR-parameter metamodels and GPR-probability metamodels. To verify the efficiency and accuracy of the proposed method, a typical circle tunnel is analyzed and good results are achieved. Finally, a practical hydraulic tunnel is analyzed using the proposed method and some beneficial conclusions are drawn.
Reliability analysis of tunnels considering the rock-support interaction using a metamodel-based reliability method
Li, Mengyao (author) / Wang, Gang (author) / Ma, Zhenyue (author) / Kang, Fei (author) / Li, Junjie (author)
2024-04-08
Article (Journal)
Electronic Resource
English
The Reliability-Based Design Optimization of considering Rock-Support Interaction for Rock Tunnels
| DOAJ | 2021
Reliability analysis with Metamodel Line Sampling
| Online Contents | 2016
Reliability analysis with Metamodel Line Sampling
| Elsevier | 2015
Seismic Reliability Analysis of Structures by an Adaptive Support Vector Regression-Based Metamodel
| Taylor & Francis Verlag | 2024