Prediction of shear‐wave velocity from CPT data at Eskisehir (Turkey), using a polynomial model

Abstract

ABSTRACTIt is widely known that seismic shear‐wave velocity is one of the most important parameters in site characterization studies. In some instances, it is necessary to determine shear‐wave velocity indirectly from common in situ tests, such as the cone penetration test. There are numerous earlier studies showing this possibility. In this study, the relation between shear‐wave velocity and subsoil geotechnical properties, e.g., cone‐tip resistance and sleeve friction from cone penetration test is reinvestigated in a study area located in Eskisehir, Turkey. New polynomial models are proposed for the correlation. A total of 437 samples extracted from 37 sites made of clay, sand, sand‐clay mixture, and miscellaneous soil types have been used. We compare our results involving polynomial fitting with earlier results of statistical correlation using power‐law or logarithmic relations between shear‐wave velocity and cone‐tip resistance or sleeve friction. The predicted values using our model are checked against the measured ones to evaluate the performance of the polynomial model. The results suggest that the newly proposed approach provides a means for recognizing more efficiently the patterns in the data and reliably predicting the shear‐wave velocity. Additionally, the sensitivity analysis reveals the influence of parameters and the contribution of each coefficient in the polynomial model. Cone penetration test cone‐tip resistance relates more strongly than sleeve friction to shear‐wave velocity. The intercept term in the polynomial is of primary importance in such correlation, for all soil types.