Selection of Viscous Damping Model for Evaluation of Seismic Responses of Buildings

Abstract

Users of structural analysis software commonly apply the Rayleigh damping model as a default option for analyzing of structures. When a distributed plasticity model is utilized in the numerical models, the structural responses are not found to be sensitive to the damping models; even the Rayleigh damping model leads to the reasonable seismic results. Moreover, the lower bounds and upper bounds of the demands are sensitive to the type of damping model. In this paper, the effects of viscous damping models in the seismic demands of moment and concentrically braced steel frames are carefully investigated. For this purpose, seven viscous damping models are considered using the different forms of the stiffness matrices in the finite-element modeling. The declared structural systems are modeled against several strong ground motion records in OpenSees environment. Then, the seismic responses such as drifts, accelerations and base shears are compared for the aforementioned damping models. The main result of this research is that applying Rayleigh damping model, as a default option for structural analyzing software will produce underestimated responses. In addition, it is shown that no matter which type of stiffness matrix is used in the stiffness proportional damping model. Therefore, using initial stiffness matrix is cost efficient, particularly in time-consuming nonlinear analyses.