Identification of pulse periods in near-fault ground motions using the HHT method

Large-amplitude and long-period pulses are observed in velocity time histories of near-fault ground-motion records. The pulses in these records have significant damage effect on flexible structures due to their long-period property; therefore, more attention should be paid to the frequency components in the ground motion. Based on the identification of frequency components in the original record, a new method based on the Hilbert-Huang transform (HHT) is proposed here. A ground-motion record can be decomposed into several intrinsic mode functions (IMFs) that carry different frequency components by the HHT without contamination from any a prior function. Only two fixed parameters, the peak ground velocity (PGV)/peak ground acceleration (PGA) ratio and the energy change of every IMF, are used to classify pulse-like ground-motion records. The inherent pulses of these records can also be extracted, based on the selection of IMFs for which PGV/PGA ratios are larger than 0.12 and energy changes that are greater than 0.1. For multipulse cases, all the pulses can be captured after extracting once, and the time course of inherent pulses can also be obtained. Then, pulse periods are calculated based on the solutions of instantaneous frequency of the peak for the extracted pulses. All the periods obtained using the HHT method can be verified by the results obtained from Baker's wavelet method. The 24 controversial records that are discussed in previous studies are examined here as well. The HHT method is a complete procedure that includes the classification of pulse-like ground motions, the extraction of velocity pulses, and the solution of pulse periods. It works well for multipulse records, especially because it can provide the exact timing of all the inherent pulses.