A new procedure for estimation of shear wave velocity profiles using multi station spectral analysis of surface waves, regression line slope, and genetic algorithm methods.

摘要

A new approach to estimate shear wave velocity using multi station spectral analysis of surface waves (MSASW), a genetic algorithm (GA), and regression line slope (RLS) methods is presented. MSASW is a nondestructive soil characterization method based on surface waves propagation. Spectral analysis of time series data by the MSASW method will result in a plot of phase velocity versus wave frequencies or wavelengths, or dispersion curve. The obtained dispersion curve then is inverted to estimate the shear wave velocity profile. To invert the dispersion curve and to estimate shear wave velocity (VS) from phase velocity (VR), a GA method is used. A GA is a global optimization technique to replicate the Darwinian evolutionary procedure used by Mother Nature. A summary of the properties and the application of the MSASW along with an overview of GA are presented.; One of the goals of this research is to use the multi-receiver version of spectral analysis of surface waves (SASW) as a method that uses several receivers as opposed to two receivers, which is used in the traditional SASW method. The experiments have shown better performances with the multi-receiver method when compared to the two-receiver SASW method (Foti 2000).; The frequencies used in this research range from 3.75 Hz to 100 Hz and the frequency increments range from 0.625 to 2.5 Hz. Computer programs have been developed to run and control the waveform generator and the data acquisition function module, to record the responses of the receivers, and to invert the dispersion curve to estimate the shear wave velocity profile.; The RLS method has been developed and used to find the dispersion curve. The method outlined by Matthews et al. (1996) has been used here as a guideline. The RLS method is simple to use and independent of expensive instruments and equipment, while it provides good results as compared to the frequency-wavenumber (f-k) method. The RLS method essentially calculates the phase difference angle between any two receivers (accelerometers), for any frequency swept by the wave generator during a test. The slope of the best line passing through all calculated phase angles for any frequency is the phase velocity for that frequency. The final result of the RLS method is the experimental dispersion curve. To analyze and process the data, a computer program in the Mathcad platform has been developed.; The estimated experimental dispersion curve is used in an inversion process using a GA to determine the theoretical dispersion curve and the resulting soil shear wave velocity (Vs) profile. Programs in the MATLAB platform have been developed for the GA-based inversion procedure. The results of the programs have been checked for accuracy by comparing them with results of other investigators (Rix et al. 2002; Pezeshk et al., 1998). The comparison showed a good agreement between the results of this study and the results of the above investigators.