Engineering Mechanics Institute Conference 2013

Full Program »

Implementation of new mathematical models of elastomeric bearings for analysis under extreme loadings

PDF Abstract
View File

Manish Kumar
University at Buffalo, The State University of New York
United States

Andrew Whittaker
University at Buffalo, The State University of New York
United States

Michael Constantinou
University at Buffalo, The State University of New York
United States

The United States Nuclear Regulatory Commission (USNRC) is sponsoring a research project that will quantify the response of low damping rubber (LDR) and lead-rubber (LR) bearings under loadings associated with extreme earthquakes. One of the goals of this project is to develop robust mathematical models of the isolators and to implement them in contemporary structural analysis programs. Under design basis loadings, the mechanical properties of elastomeric bearings are not expected to vary substantially and bearings are not expected to experience net tension. However, under extended or beyond design basis shaking, elastomer shear strains may exceed 300% in regions of high seismic hazard, bearings may experience net tension, the compression and tension stiffness will be affected by isolator lateral displacement, and the properties of the lead core in LR bearings will degrade due to substantial energy dissipation. New mathematical models have been proposed, which consider all the major properties that are expected to have effect on the response of seismically isolated structure subjected to beyond design basis shaking. The implementation of new mathematical models of isolators is presented here.
The mathematical models are implemented through creation of User Elements (UELs). The formulation of the UELs for the LDR and LR bearings is discussed in the framework of OpenSees and Abaqus. A systematic approach for the development of computational models from mathematical models is described. Comparative performance of OpenSees and Abaqus in its ability of modeling and analysis of isolators and base isolated structures is also discussed.
It is important that a high confidence level is established in the models used for prediction of the outcomes of high-consequence events to enable the system to perform as intended. The design basis and beyond design shaking of Nuclear Power Plants (NPPs) are example of such high-consequence events. The models used to predict the outcome of such events need to be verified and validated to establish the confidence level required in them for the prediction of response. A comprehensive exercise of verification and validation, consistent with ASME recommendations , is undertaken to quantify and remove error sources and establish robustness of the user elements.


Powered by OpenConf®
Copyright ©2002-2012 Zakon Group LLC