Application of wavelet theory with denoising to estimate the parameters of an earthquake

Abstract

In this paper, Strong Ground Motion (SGM) parameters are calculated using Discrete Wavelet Transform (DWT) in different kinds of soil with different magnitudes. The Main Earthquake Record (MER) is divided into approximation and detailed signals using wavelet transform with denoising. The high and low frequencies of MER are separated from each other. Previous studies have shown that the approximation signal has the greatest effect on dynamic response and is very similar to the main signal. Then, SGM parameters of the new signal are calculated by DWT decomposition. This process continues over five levels and, at each level, SGM parameters are calculated and compared with the MER. Its error percentage is then presented. In DWT with the denoising method, the curve becomes softer such that the calculation time reduces. Results show that the error percentage in the first two levels is less than 1% and for the third level, this index is less than 3%. In addition, the reduction percentage of calculation time is 1%, 4%, and 8%, respectively, in the first to third levels. The best result is relative to the third decomposition level in which error value as well as computational time reduction is nearly 3% and 8%.