| The critical structural response caused by near-fault ground motions is often influenced by the presence of a velocity pulse in the fault-normal component. A procedure for simulating fault-normal near-fault ground motions for a specified seismic environment (i.e., the magnitude, distance and faulting mechanism of an earthquake and the soil conditions at the site) is described. The proposed model combines an analytical model for the velocity pulse with a realization of a stochastic process that represents the high frequency content of a ground motion. The proposed ground motion velocity model is defined by a number of parameters that, for a specified near-fault record, are determined by a nonlinear regression. Predictive relationships for these parameters are derived from a series of regression analyses performed on an ensemble of recorded fault-normal near-fault ground motions. The results of time-history analyses performed on linear and nonlinear single-degree-of-freedom systems indicate that the proposed simulation procedure generates acceleration time histories that, on average, predict the displacement demands at all periods and ductility demand levels with sufficient accuracy for earthquake engineering applications. The simulated fault-normal ground motions generated by the proposed model are used to supplement the available database of fault-normal records to better understand the displacement response of linear and nonlinear single-degree-of-freedom (SDOF) and multiple-degree-of-freedom (MDOF) systems to such ground motions. In general, the results of these studies indicate that the displacement responses of SDOF and MDOF systems to fault-normal ground motions are dependent upon the period of the velocity pulse present in the record relative to the fundamental period of the system. The time-history results are also compared to those available for far-field ground motions in an effort to identify how seismic design procedures commonly used in the current practice should be modified for structures located in near-field environments. |
