Equivalent Linear Ground Response Analysis Considering Statistical Randomization of Shear Wave Velocity

The outcomes of ground response analysis (GRA) can significantly get affected by the variations of shallow soil properties. As the estimation of those properties are always associated with some inherent variabilities, it is quite important to consider those variabilities in the estimation of GRA. In this study, the effect of the variability associated with shear wave velocity (Vs) measurements on GRA is assessed for a site located in Kolkata. The variability in Vs is accounted through the statistical randomization. In the work, the parametric variations of standard deviation of Vs and inter-layer correlation coefficient, involved in the randomization process, are studied. In addition to that, the effect of intensity of input ground motion has also been investigated. Four different values of each parameter are considered in the analysis. In each case, 400 randomized profiles are analysed and a total of 4800 simulations are performed in the present study. The outcomes of GRA are presented in the form of transfer function (TF), spectral acceleration (SA) and PGA variation with depth. To quantify the variations for different adopted cases of parametric variations, the outcomes have also been quantified in the form of mean and standard deviations. The study reveals that the effect of standard deviation of Vs and intensity of input motion affect the GRA outcome substantially. In the study of the effect of standard deviation of Vs, TF exhibits higher standard deviations with a maximum value of similar to 3, whereas SA and PGA variations show a maximum value of similar to 0.45 and similar to 0.35, respectively. In case of intensity of input motion, the maximum values of standard deviation are observed to be similar to 3.5, 0.45 and 0.4 for TF, SA and PGA variations, respectively. Finally, the probability density functions and cumulative distribution functions of surface PGA variations are developed and proposed. To study the above, an equivalent linear ground response analysis program is developed in MATLAB environment along with the provisions of Vs randomization and repeated simulations of GRA.