Estimation of Frequency‐Dependent Strong Motion Duration Via Wavelets and Its Influence on Nonlinear Seismic Response

Abstract

A procedure for estimation of frequencydependent strong motion duration (FDSMD) is developed. The proposed procedure utilizes the continuous wavelet transform and is based on the decomposition of the earthquake record into a number of component time histories (named "pseudo-details") with frequency content in a selected range. The "significant" strong motion duration of each pseudo-detail is calculated based on the accumulation of the Arias intensity (AI). Finally, the FDSMD of the earthquake record in different frequency ranges is defined as the strong motion duration of the corresponding pseudo-detail scaled by a weight factor that depends on the AI of each pseudo-detail. The efficiency of this new strong motion definition as an intensity measure is evaluated using incremental dynamic analysis (IDA). The results obtained show that the proposed FDSMD influence the peak response of short-period structures with stiffness and strength degradation.

Quantifying the Influence of Ground Motion Duration on Structural Collapse Capacity Using Spectrally Equivalent Records

“…In this study, the scarcity of available ground motions has been addressed by collecting and utilizing long duration ground motions recorded from recent large magnitude earthquakes, most notably the 2008 Wenchuan (China), 2010 Maule (Chile), and 2011 Tohoku (Japan) earthquakes. The second challenge of isolating the effect of duration from other ground motion characteristics has been previously addressed by Hancock and Bommer (2007), Montejo and Kowalsky (2008), and Ou et al (2014) by modifying the spectral content of recorded accelerograms to have similar response spectra. Sideras and Kramer (2012) used stochastically simulated accelerograms having similar amplitude and frequency characteristics, but different durations.…”

A Probabilistic Methodology to Determine Elastic Acceleration Response Spectra for Pulse-Type Records through Multi-Resolution Analyses

“…In previous analyses, response spectra were usually normalized with the peak ground motion or other similar parameters such as spectrum intensity, effective peak acceleration, etc., so the influence of peak ground motion on response spectra can be eliminated to an extent. Previous research work has shown that strong motion duration may largely influence the inelastic demand of short-period structures with stiffness and strength degradation [Bommer et al, 1999[Bommer et al, , 2004Hancock and Bommer., 2007;Montejo and Kowalsky, 2008]. But for elastic systems, the effects caused by duration of strong motion are inconspicuous.…”

Wavelet‐Based Method for Generating Nonstationary Artificial Pulse‐Like Near‐Fault Ground Motions

“…Wavelet analysis has been used widely in engineering to detect frequency content of signals; for example, in earthquake engineering (Zhou and Adeli, ; Pakrashi et al., ; Montejo and Kowalsky, ), structural engineering (Amini et al., ; Xiang and Liang, ; Jiang et al., , ; Adeli and Kim, , ; Kim and Adeli, , b; Spanos et al., ; Jiang and Adeli, , b), transportation engineering (Samant and Adeli, ; Karim and Adeli, , b, ; Ghosh‐Dastidar and Adeli, ; Xie et al., ; Boto‐Giralda et al., ), dam engineering (Su et al., ), medical science (Acharya et al., ), neural science (Lin et al., ; Kodogiannis et al, ), and image analysis (Zou et al., ; He et al., ; Tao et al., ; Hsu, ). One approach for generating spectrum‐compatible time histories is based on the wavelet transform (Iyama and Kuwamura, ; Mukherjee and Gupta, , b; Suarez and Montejo, ; Rajasekaran et al., ; Amiri et al., , ).…”