Intelligent Approaches for Predicting the Intact Rock Mechanical Parameters and Crack Stress Thresholds

This study aims to make a unique contribution to the existing body of knowledge about rock strength and deformation parameters and crack stress thresholds through intelligent and statistical approaches applied to a database comprising various rock types (i.e., sedimentary, igneous, and metamorphic rocks). The database contains physical-mechanical and ultrasonic parameters. Six distinct machine learning (ML) algorithms- artificial neural network (ANN), random forest (RF), decision tree (DT), K-nearest neighbor (KNN), support vector regression (SVR), and bagging regressor (BR)- along with the conventional linear regression techniques, were employed to develop predictive models. These models estimate uniaxial compressive strength and Tangent Young's modulus based on bulk density and P-wave ultrasonic velocity. Furthermore, they predict crack stress thresholds (i.e., crack closure stress, crack initiation stress, and crack damage stress) as a function of unconfined compressive strength, Tangent Young's modulus, bulk density, P-wave ultrasonic velocity, axial strain at failure, and lateral strain at failure. Various performance indices were utilized to evaluate and compare the performance of these models. The results indicated that the RF method outperformed other ML-based and linear regression-based approaches in predicting the output parameters. Additionally, the multiple parametric sensitivity analysis (MPSA) was carried out to determine the significance of input parameters in predicting the output variables. This analysis revealed that P-wave ultrasonic velocity bulk density have the highest and lowest impact on predicting unconfined compressive strength and Tangent Young's modulus, respectively. On the other hand, unconfined compressive strength was identified as the most influential parameter in predicting crack initiation stress and crack damage stress, while radial strain at failure and P-wave ultrasonic velocity showed the least impact on the foregoing outputs, respectively. This is while axial strain at failure and bulk density were, respectively, found as the most important and least important factors in predicting crack closure stress. Finally, to facilitate easy access to the prediction results and enhance the practicality of the proposed RF model, a graphical user interface (GUI) was developed, which enables the practical application of the most performing developed prediction model.