• Structural Engineering and Mechanics
• /
• v.73 no.6
• /
• pp.651-666
• /
• 2020

In this study, the spatial variation mechanisms of large far-field earthquakes at engineering scales are first investigated with data from the 2008 Ms 8.0 Wenchuan earthquake. And a novel 'coherency cut-off frequency' is proposed to distinguish the spatial variations in ground motions in the low-frequency and high-frequency ranges. Then, a practical piecewise coherency model is developed to estimate and characterize the spatial variation in earthquake ground motions, including the effects of source-to-site distances, site conditions and neighboring topography on these variations. Four particular earthquake records from dense seismograph arrays are used to investigate values of the coherency cut-off frequency for different source-to-site distances. On the basis of this analysis, the model is established to simulate the spatial variations, whose parameters are suitable for both near- and far-field earthquake conditions. Simulations are conducted to validate the proposed model and method. The results show that compared to the existing models, the proposed model provides an effective method for simulating the spatial correlations of ground motions at local sites with known source-to-site distances.

• Smart Structures and Systems
• /
• v.16 no.5
• /
• pp.919-935
• /
• 2015

The Hilbert-Huang transform (HHT) consists of empirical mode decomposition (EMD) and Hilbert spectral analysis. EMD has been successfully applied for identification of mode shapes of structures based on input-output approaches. This paper aims to extend application of EMD for output-only identification of mode shapes of linear structures. In this regard, a new simple and efficient method based on band-pass filtering and EMD is proposed. Having rather accurate estimates of modal frequencies from measured responses, the proposed method is capable to extract the corresponding mode shapes. In order to evaluate the accuracy and performance of the proposed identification method, two case studies are considered. In the first case, the performance of the method is validated through the analysis of simulated responses obtained from an analytical structural model with known dynamical properties. The low-amplitude responses recorded from the UCLA Factor Building during the 2004 Parkfield earthquake are used in the second case to identify the first three mode shapes of the building in three different directions. The results demonstrate the remarkable ability of the proposed method in correct estimation of mode shapes of the linear structures based on rather accurate modal frequencies.