Font Size: a A A

Hilbert-Huang Transform(HHT) And Its Appli-Cation In Structural Damage Identification

Posted on:2007-04-13Degree:MasterType:Thesis
Country:ChinaCandidate:L Z HeFull Text:PDF
GTID:2132360185986297Subject:Structural engineering
Abstract/Summary:PDF Full Text Request
Natural corrosion and lots of other factors inevitably cause damage in structures under complex environment. The damage identification and security research of structures play an important role in greatly reducing the loss of lives and properties.The study state of art and development trend of structural damage identification research is simply summarized in this paper.Hilbert-Huang Transform (HHT) is applied to the damage identification and intrinsic frequency confirmation of time-variable structures.Furthermore, a kind of inverse calculation is presented to calculate structural stiffness.This method can solve three main problems: the time that damage occurred, the location and the degree of damage.At the same time, the dynamic response of structure under environmental excitation such as ground micro-earthquake, vehicle vibration and wind vibration, is usually disturbed with noise. Thereby, the Signal-to-Noise ratio of data collected by sensors is rather low. To solve this problem, some researches on denoising for the nonstationary and noisy-strong signal are carried out in this paper. Combining Hilbert-Huang transform (HHT) which has been proved to be more valid in the aspect of time-frequency analysis, with mature Wavelet transform (WT) threshold theory, an EMD-based wavelet threshold denoising algorithm is presented.Besides, during the process of empirical mode decomposition (EMD), the error resulting from the boundary problem is becoming more.A semi-periodical—semi-symmetrical method is presented to reduce the error.Finally, based on above findings, a set of technique is presented towards the problem of structural damage identification. First of all, the EMD-based wavelet threshold denoising algorithm is apllied to denoise noisy structural response data to reduce the effect resulting from noise. During the process of the empirical mode decomposition (EMD), the two boundaries of the response signal are processed with semi-periodical—semi-symmetrical method. Subsequently, Hilbert-Huang transform (HHT) is used in identifying structural intrinsic frequency. Structural stiffness with Newton iteration method is calculated in the end.
Keywords/Search Tags:Structural damage identification, Hilbert-Huang transform (HHT), Empirical mode decomposition (EMD), Mean-squared Signal-to-noise ratio(SQRT_SNR)
PDF Full Text Request
Related items