| Strong ground motion records are the signals that contain a lot of structural response information and motion information. They are basic data to investigate earthquake engineering and the design of aseismatic engineering. In order to gain important information of strong ground motion records, we use Fourier transform traditionally. With the development of information technology, all kinds of signal management technologies come into being and flourish in recent 20 years, such as Wavelet, Hilbert-Huang transform. The paper will investigate their characteristics by comparing their physical essences. We are going to 'find their foreground in strong ground motion records management and analysis, and to discuss possible influence in earthquake functionThe main achievements of this study could be summarized as follows:1. systemic expatiation of the three transforms and their physical meanings in signal management.From the aspect of physical meanings, Fourier transform, Wavelet transform and Hilbert-Huang transform are explained . Their foundations and basic characteristics will be discussed. The differences in dealing with frequency conversion signals will be introduced. At the same time, the differences on their physical meanings will be said.2. the examples on analyzing their applied fields and their contactions.From the aspect of application, the differences of their applied fields in point are listed out .And Wavelet's wave filter function to earthquake signals is given a demonstration. We also talk about its future and HHT's shortcoming in this aspect. Of the basic measures: instantaneous frequency and original frequency, their relationship is given qualitative analysis in earthquake data. Resolution to differential coefficient equation will be discussed. In the same time ,their application future is noted.3. statistic analysis to Chi-Chi earthquake,to talk about HHT transform etc.To Chi-Chi earthquake ,we choose B,C,D,E sites of Taiwan standard,39 issues for D site. Epicenter distance and site will be discussed with Fourier transform, mostly in energy density spectrum. We will compare instantaneous energy spectrum and tune-frequency spectrum. Places that HHT is superior to Fourier transform in dealing with earthquake data will be seen, especially in distance and site. The changing significanceof HHT in dealing with nonstationary signals is affirmed. At the same time the aspects that need improving would be pointed out. We find out by analyzing:? Duration variety makes little change to Fourier spectrum.? Fourier transform suits to HHT in analyzing the trends of site and distance, especially of energy density spectrum.? Fourier transform behaves far from as well as HHT in earthquake, information. HHT's shortcoming is imperfect theory and non-standard program.? In site condition, HHT marginal spectrum and energy density spectrum (EDS) put up same developing trends with Fourier transform, but predominant frequency is different. We can see more sameness in EDS.EDS can tell us weakened characteristics of site to earthquake, and instantaneous EDS can tell us that main duration becomes longer as site is softer. C and D site's frequencies concentrate on 15Hz below. In weak shock time ,high frequencies are more and irregular. Soft ground records concentrate on 5 Hz below, such as E site. In the same site ,vertical vibration has more high frequencies than the level of the same record.? In order to discuss distance influence, we analyze C and D site. We find that instantaneous EDS's main duration become longer relatively with distance. Response spectrum from linear acceleration supposition is not corresponding to Fourier spectrum. Near-field site, D site condition ,has more high frequencies and the proportion of HHT predominant frequency to high frequency is higher than Fourier's..
|
PDF Full Text Request