Subject:
Correction
of Earthquake Records Using Wavelet Conversion with Case Study
Abstract
Time history analysis method is a powerful and effective method in
assessing structures response to earthquake due to the influence of structural
reflections at each time point during earthquake. To perform this analysis, there
is a need for accelerations that are consistent with the region's seismic
characteristics. But in some areas, seismic data is not well recorded. One way
to solve this problem is to use Acceleration of artificial mappings for the
seismic properties of the area. One of the problems with this path is the noise
in the earthquake signals. One of the best ways to remove noise from earthquake
signal data is wavelet transform. In this thesis, a wavelet transform-based
method for generating spectral-compatible mappings is introduced. In this method,
a record is decomposed by wavelet transform, for this purpose, the discrete
wavelet transform of the Elsentro earthquake mapping is obtained as the input
parameter and the details and approximations are derived from these mapping
acceleration data, which are general, low frequency and detail, high frequency
(noise). Now, if the Elsentro earthquake signal is subtracted from the details
resulting from the discrete violet function output, the noise is eliminated
from the earthquake mapping acceleration function. At the filtering stage,
careful analysis of the data obtained from the analysis of different waves
using wavelet transform is performed Will exist. Detail coefficients are small
and include high frequency signals, while approximation coefficients are almost
similar to the original wave and include low frequencies. By using discrete
wavelet transform and noise removal, one can obtain the time of each frequency
and its intensity. Also, by examining the frequency-time diagram it was
observed that what frequencies are applied to the structure by earthquakes.
This helps us to avoid resonance or resonance in the structure if the
earthquake frequency is equal to the intrinsic frequency of the structure.