11.23: Influence of the uniaxial material model of steel on the seismic response of steel structures

Abstract

ABSTRACTThe model proposed by Menegotto and Pinto [1] is widely used to simulate the dynamic response of steel structures and steel bars of reinforced concrete structures. In 1983, Filippou et al. [2] highlighted a flaw in the original formulation of this material model. They observed that, if partial unloading takes place at strains lower than the maximum recorded value, the reloading path provides forces that are higher than those expected. Filippou et al. deemed that such errors were not particularly significant for reinforced concrete members and did not propose any modification to the analytical formulation of the model.More recently, some authors have independently investigated this issue and highlighted that the above‐mentioned flaw may affect the prediction of the seismic response of steel structures. Thus, some modifications have been proposed [3–5] and implemented in OpenSees computer program.In this study, the seismic response of a steel moment resisting frame and that of a braced frame are analysed by nonlinear dynamic analysis by means of the OpenSees program. For each structure, the stress–strain relationship of steel is separately modelled by means of the original formulation of the model by Menegotto and Pinto and by the modified versions proposed by Zsarnóczay and Budaházy [3], Kolozvari et al. [4] and Bosco et al. [5]. The seismic response predicted by the considered models is compared. Attention is focused on the moment–curvature time‐history at the ends of the members of the framed structure and on the axial force‐axial deformation time‐history of braces. Global response parameters are also considered in terms of the maximum storey drifts.