Modeling of Characteristics of Complex Microstructure and Heterogeneity at the Core Scale

Complex pore structures and strong matrix heterogeneity distinguish carbonate rocks, but there is a lack of comprehensive methods to describe these characteristics. In this study, a integrated approach is proposed to improve the accuracy and adaptability of velocity prediction methods, using a modified squirt flow model based on microcrack structures to characterize complicated pore structures, and a mixed random medium model to represent significant heterogeneity. In addition, the microcrack structure is obtained by inversion, but different from the D-Z method, each group of microcracks corresponds to a different equivalent medium model, so as to improve the accuracy of the inversion results. And the modified squirt flow model takes into account the attenuation caused by local flow between microcracks. The random medium model simulates the inhomogeneous body in the core by adjusting the autocorrelation length a and b, the rounding coefficient n, and the angle $\theta$. A comparative study of the measured data of five limestone and dolomite samples reveals that the P-wave prediction error of the new model is less than 5%, whereas the Biot model is less than 10%, implying that the prediction accuracy of the new model is better.