Seismic reliability functions for complex systems based on a secant-stiffness reduction index

摘要

The stiffness-reduction index proposed in this study can be efficiently used for the determination of second moment probabilistic descriptions of Z_F, the natural logarithm of the minimum value of the intensity required to produce collapse of a complex nonlinear system and, therefore, for the determination of seismic reliability functions of those systems. This is achieved while avoiding the need to depend on the determination of a lateral deformation capacity, which is very sensitive to the strength and stiffness degradation of the system due to its cyclic response, as well as to its lateral configuration at the instant of impending collapse.rnThe determination of the reliability index β is based on the determination of the mean and standard deviation of Z_F, on the basis of a statistical sample of values of Z, the natural logarithm of the intensity, and I_(SSR), the stiffness-reduction index, in the vicinity of I_(SSR) = 1.0. Information derived from sample points with I_(SSR) cannot be directly taken into account through a conventional least-squares regression analysis, because the corresponding intensities may be equal to or larger than the required minimum values. The maximum likelihood approach presented here is more efficient than one previously proposed by Esteva and Ismael (2004) when only the collapse failure mode is of interest in the evaluation of the seismic vulnerability of a system.