A constrained back propagation neural network(C-BPNN)model for standard penetration test based soil liquefaction assessment with global applicability is developed,incorporating existing knowledge for liquefaction trig...A constrained back propagation neural network(C-BPNN)model for standard penetration test based soil liquefaction assessment with global applicability is developed,incorporating existing knowledge for liquefaction triggering mechanism and empirical relationships.For its development and validation,a comprehensive liquefaction data set is compiled,covering more than 600 liquefaction sites from 36 earthquakes in 10 countries over 50 years with 13 complete information entries.The C-BPNN model design procedure for liquefaction assessment is established by considering appropriate constraints,input data selection,and computation and calibration procedures.Existing empirical relationships for overburden correction and fines content adjustment are shown to be able to improve the prediction success rate of the neural network model,and are thus adopted as constraints for the C-BPNN model.The effectiveness of the C-BPNN method is validated using the liquefaction data set and compared with that of several liquefaction assessment methods currently adopted in engineering practice.The C-BPNN liquefaction model is shown to have improved prediction accuracy and high global adaptability.展开更多
Earthquake induced liquefaction is one of the main geo-disasters threating urban regions, which not only causes direct damages to buildings, but also delays both real-time disaster relief actions and reconstruction ac...Earthquake induced liquefaction is one of the main geo-disasters threating urban regions, which not only causes direct damages to buildings, but also delays both real-time disaster relief actions and reconstruction activities. It is thus important to assess liquefaction hazard of urban regions effectively and efficiently for disaster prevention and mitigation. Conventional assessment approaches rely on engineering indices such as the factor of safety(FS) against liquefaction, which cannot take into account directly the uncertainties of soils. In contrast, a physics simulation-based approach, by solving soil dynamics problems coupled with excess pore water pressure(EPWP) it is possible to model the uncertainties directly via Monte Carlo simulations. In this study, we demonstrate the capability of such an approach for assessing an urban region with over 10 000 sites. The permeability parameters are assumed to follow a base-10-lognormal distribution among 100 model analyses for each site. A dynamic simulation is conducted for each model analysis to obtain the EPWP results. Based on over 1 million EPWP analysis models, we obtained a probabilistic liquefaction assessment. Empowered by high performance computing, we present for the first time a probabilistic liquefaction hazard assessment for urban regions based on dynamics analysis, which consider soil uncertainties.展开更多
基金The authors would like to thank the National Natural Science Foundation of China(Grant Nos.51678346 and 51879141)Tsinghua University Initiative Scientific Research Program(2019Z08-QCX 01)for funding this work.
文摘A constrained back propagation neural network(C-BPNN)model for standard penetration test based soil liquefaction assessment with global applicability is developed,incorporating existing knowledge for liquefaction triggering mechanism and empirical relationships.For its development and validation,a comprehensive liquefaction data set is compiled,covering more than 600 liquefaction sites from 36 earthquakes in 10 countries over 50 years with 13 complete information entries.The C-BPNN model design procedure for liquefaction assessment is established by considering appropriate constraints,input data selection,and computation and calibration procedures.Existing empirical relationships for overburden correction and fines content adjustment are shown to be able to improve the prediction success rate of the neural network model,and are thus adopted as constraints for the C-BPNN model.The effectiveness of the C-BPNN method is validated using the liquefaction data set and compared with that of several liquefaction assessment methods currently adopted in engineering practice.The C-BPNN liquefaction model is shown to have improved prediction accuracy and high global adaptability.
基金This research was supported by the FOCUS Establishing Supercomputing Center of Excellence。
文摘Earthquake induced liquefaction is one of the main geo-disasters threating urban regions, which not only causes direct damages to buildings, but also delays both real-time disaster relief actions and reconstruction activities. It is thus important to assess liquefaction hazard of urban regions effectively and efficiently for disaster prevention and mitigation. Conventional assessment approaches rely on engineering indices such as the factor of safety(FS) against liquefaction, which cannot take into account directly the uncertainties of soils. In contrast, a physics simulation-based approach, by solving soil dynamics problems coupled with excess pore water pressure(EPWP) it is possible to model the uncertainties directly via Monte Carlo simulations. In this study, we demonstrate the capability of such an approach for assessing an urban region with over 10 000 sites. The permeability parameters are assumed to follow a base-10-lognormal distribution among 100 model analyses for each site. A dynamic simulation is conducted for each model analysis to obtain the EPWP results. Based on over 1 million EPWP analysis models, we obtained a probabilistic liquefaction assessment. Empowered by high performance computing, we present for the first time a probabilistic liquefaction hazard assessment for urban regions based on dynamics analysis, which consider soil uncertainties.