evaluation of uncertainty in shear-wave velocity based on cpt records using the robust optimization method

Shear-wave velocity (vs) is used to evaluate the soil shear modulus and classify the soil type in pseudo-static analysis. empirical correlations are developed to relate vs and cone penetration test (cpt) records. however, uncertainty in the input parameter measurements is always a major concern. therefore, the current research employs a novel method based on robust optimization to study the effect of such uncertainties. to measure the merits of the suggested method, 407 records were collected and categorized for several soil types. the identification procedure employed in this investigation is based on the robust model of least squares, solved using the interior point technique for second-order cone problems. the uncertainty definition is examined against correlation coefficients for empirical models, and optimum values are determined based on the frobenius norm of the data points. a diagram for calculating the shear wave velocity considering uncertainties is also presented. this study suggests that the robust method is the best pattern recognition tool for uncertain datasets compared to previous statistical models. other power models also have good accuracy compared to the polynomial model, but when uncertainty is taken into account, the accuracy of the other models is lower compared to the polynomial model.