Implementation And Application Study Of Lead-rubber Seismic Isolation Bearings

Seismic isolation,which has been extensively employed and developed in engineering structures,plays a vital role in the seismic improvement of structural performance,safety and function,and lead-rubber bearing?LRB?is one of widely-utilized seismically-isolated devices.The focus of this disseration is placed on the implementation and application investigation of a numerical model for LRBs.The main tasks involve the numerical study of ultimate behavior of LRBs under combined compression and shear,the implementation and validation of a numerical model for LRBs,stochastic modeling and synthesis of near-fault forward-directivity ground motions,and seismic fragility assessment of base-isolated structures under that ground motions.Main conclusions are drawn as follows:?1?Numerical study on the ultimate behavior of LRBs under combined compression and shearPredicting the ultimate behavior of LRBs subjected to combined compression-shear loading is an indispensable part of quantitatively evaluating the performance of seismically-isolated structures during earthquakes.First,the initial critical load P_cr,0,which is derived from the effective shear modulus G_effff and three different approximations for bending modulus E_b,is analyzed and evaluated by existing experimental data.Then,the finite element models of bearings are developed and validated by the critical load-lateral displacement results.Based on that,the criterion of shear failure is quantitatively determined and a parametric finite-element study is performed to investigate the failure mode of bearings under this combined loading.It is demonstrated that,by virtue of the concept of G_eff,the E_b based on the pressure solution and further simplification provides a more reasonable estimate of P_cr,0.The ratio of the inner to outer diameter of a bearing has a negligible impact on the critical load results,and the overlapping area method is capable of capturing the general trend of critical load in the deformed configuration.Furthermore,the dividing line between stability limit and shear failure is dominated by the second shape factor S₂;as the first shape factor S₁ increases,the ultimate shear deformation capacity corresponding to the shear failure slightly increases.?2?Implementation and validation of a numerical model for lead-rubber bearingsTo accurately analyze the performance of seismically-isolated structure,a numerical model,which is implemented in OpenSees,accounts for the mechanical characteristics of bearings as follows:first,the bi-lateral interaction effect of hysteretic behaviors,as well as the variation in post-yield shear stiffness due to vertical load,is considered;second,the reduced vertical stiffness and critical load with increasing horizontal displacement are accounted for,moreover,the cavitation and permanent damage effects in bearings are mathematically included.To validate the numerical model,simulation analyses are performed for a series of static and dynamic loading tests,including bi-lateral displacement-controlled tests,cyclic tensile loading tests and three-dimensional shaking table tests.It is demonstrated that this model reasonably captures the mechanical behaviors of bearings,and accounting for the variation in the meahanical characteristics of bearings is beneficial to accurately representing the dynamical responses of bearings during earthquakes,it is indicated that the proposed model provides an effective tool for the failure mode analyses of bearings and the seismic fragility analyses of seismically isolated structures.?3?Stochastic modeling and synthesis of near-fault forward-directivity ground motionsNear-fault forward-directivity ground motions have unfavorable impacts on long-period structures,especially seismically isolated structures.Accounting for the scarcity of that records,the stochastic modeling and synthesis of near-fault forward-directivity ground motions for prescribed seismic scenario is presented.This model,which is established in the orientation containing the strongest pulse,combines the directivity pulse part represented by the Gabor wavelet pulse model with the high-frequency part characterized by the modulated filtered white-noise model.This stochastic model reasonably incorporates the directivity effects,the pulse behavior,and the temporal and spectral nonstationarity.Based on the database of recorded near-fault ground motions,the empirical predictive equations for the model parameters are constructed in terms of focal mechanism,moment magnitude,source-to-site distance,and site conditions,the correlations between model parameters are considered as well.Based on that,the procedure for synthesizing near-fault forward-directivity ground motions for the given seismic scenario is presented.These scenario-specified generated ground motions are capable of providing dynamic excitations to the vulnerability assessment of seismically isolated structures.?4?Copula-based seismic fragility assessment of base-isolated structures under near-fault forward-directivity ground motionsBased on the proposed numerical model and the stochastic ground motion model,the seismic fragility assessment of base-isolated structures under near-fault forward-directivity ground motions is performed.First,a general framework for generating system-level fragility curves,which accounts for the vulnerability contributions of multiple components,is developed based on the joint probabilistic seismic demand model?JPSDM?and the capacity model of various components;in this framework,the dependence structure of multiple component demands is modelled and simulated via the copula technique,the best-fit copula function is quantitatively identified and is used to construct the sampling-based JPSDM.This framework is applied in the case of a typical base-isolation RC frame,and the maximum shear displacement in the isolation layer and the maximum inter-storey drift of the superstructure are considered as the two main component demands;it is demonstrated that the dependence among those two component demands can be reasonably characterized by the t copula,which is quantitatively identified as the best-fit copula function,and the lognormal marginal distribution underlying the corresponding PSDM;the whole system is more fragile than any components,and the impact of choosing different copula functions on the system-level fragility depends on the relative fragility contribution of different components.