Directivity pulses in near-fault ground motions—I: Identification, extraction and modeling

In ensuring structural safety against near-fault pulse-type ground motions it is of critical importance that sufficient recorded motions of this type are available and that suitable synthetic motions can be simulated for those situations where the recorded motions are insufficient or non-existent. To this end, an objective criterion is developed for the directivity pulse-type motions, to identify such motions from the available suite of recorded ground motions. This criterion is based on the value of maximum fractional signal energy contribution by any half-cycle of the velocity time-history. For the simulation of directivity pulse-type motions, it is proposed to estimate and superimpose directivity pulse(s) on a non-pulse type motion. An algorithm is developed to extract the directivity pulse features from a pulse-type accelerogram via its repetitive smoothening, and based on the pulses extracted from a database of 91 records two mathematical functions are proposed to model the directivity pulses. It is found that the amplitude and dominant Fourier period of the expected velocity pulse may be sufficient to generate the shapes of the expected acceleration and velocity pulses.

► An energy-based pulse identifier for identifying directivity-pulse type motions is proposed.
► An algorithm for extracting directivity pulses from pulse-type ground motions is proposed.
► Two types of directivity pulses are identified from 91 pulse-type motions.
► Analytical models based on the Mexican Hat function are proposed for the two pulse types.
► Amplitude and dominant Fourier period of velocity pulse are inputs for the proposed models.