Wave velocity is used to determine rock material, porosity, degree of petrification, fluid type, and mechanical and behavioral properties. In this study, after assessing the relationship between the static elastic modulus (Es) and the dynamic elastic modulus (Ed), various models using statistical and intelligent methods were presented for predicting shear wave velocity (Vs) and compressional wave velocity (Vp) based on porosity (P), Brazilian tensile strength (BTS), density (D), point load index (PLI), and water absorption (A) of sedimentary rocks. The Vp and Vs were estimated using simple and multiple regression, back-propagation artificial neural network (BPANN), support vector regression (SVR), and adaptive neuro-fuzzy inference system (ANFIS) methods. The examination of necessary assumptions of the models such as analysis of variance (ANOVA), variance inflation factor (VIF), mean absolute percentage error (MAPE), root-mean-square error (RMSE), variance accounted for (VAF), and independence of errors showed the high accuracy of the obtained model using multiple linear regression. The SVR approach using the radial basis kernel function with R2=100% and 99% showed the best accuracy in estimating Vs and Vp, respectively. The average ratio of Ed/Es, dynamic-to-static Poisson ratio ( νd∕νs ) , and Vp/Vs were obtained as 2.52, 2.92, and 2.82, respectively. The most accurate relationship between Ed and Es was developed in the form of a power function with R2=0.88.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.