ARMA parameters are developed to assess the damage potential in seismic records. The parameters are chosen to fit acceleration time series of particular earthquake records using the maximum-likelihood method. For each event, a set of random accelerograms is generated and used to establish statistically-valid structural response spectra. Since the number of earthquake accelerograms in any seismic region is limited, non-stationary stochastic models are used to characterize earthquake ground motion. From a sample of earthquakes, the mean and variance of response spectral ordinates are obtained for damage predictors. Structural design spectra for earthquakes are based on smoothed linear response spectra obtained from different events scaled by their peak values. Such an approach does not incorporate other characteristics of the excitation represented by measured data. Samples of acceleration records are generated for each event. In this study, individual records for an earthquake are treated as one realization of an underlying non-stationary process that actually characterizes the earthquake. This research provides a reliable description of the information contained within acceleration records, and can also provide a reasonable estimate of the average nonlinear demand spectra.
Full Text (PDF)