Search

The longitudinal wave velocity (v L) in a rock bolt is a useful parameter for evaluating the conditions of the rock bolt and the surrounding rock mass. This study investigated the influence of rock properties on the prediction of v L in a rock bolt using numerical and machine learning approaches. Through numerical simulations, we obtained a dataset of the variations in v L according to rock properties, compressional wave velocity (v p), shear wave velocity (v s), density (ρ), Poisson's ratio (μ), porosity (η), uniaxial compressive strength (UCS), and slake durability index (I SD). This dataset was used to design a deep neural network, and the predicted v L was correlated with rock properties. Notably, v L is strongly correlated with v p , v s , ρ , η , UCS , and I SD. Principal component analysis was employed to characterize the relationship between the rock properties, and the retaining rock properties for random forest (RF) were determined. In the RF, the variable importance (VI) of rock properties was assessed. In particular, v s emerged as the most significant predictor of v L. However, relying on v s to predict v L is not sufficient because it accounts for approximately 60–70 % of the VI. For a more reliable prediction of v L , it is essential to incorporate both v s and v p , which collectively account for approximately 80 % of VI. Notably, the VI of physical properties (v p , v s , ρ , and η) accounts for more than 90 %, implying that these properties can be effectively used to predict v L even in the absence of data concerning mechanical properties. [ABSTRACT FROM AUTHOR]

Physico-mechanical rock properties are typically investigated via laboratory tests using core samples. However, sample coring is time consuming and expensive. The longitudinal wave velocity in rock bolts embedded in a rock mass depends on the surrounding rock properties. Hence, the longitudinal wave velocity in rock bolts can be utilized to predict rock properties. This study presents the relationship between granite rock properties and longitudinal wave velocity (v L) in rock bolts to predict rock properties using the longitudinal wave velocity in a rock bolt. Chemical (saline solution) and mechanical (slake durability test) weathering processes are employed to diversify the properties of the rock specimens. Laboratory tests are conducted on rock specimens to measure the physico-mechanical rock properties, including the velocities (v p and v s) associated with constrained and shear moduli, density (ρ), Young's modulus (E), Poisson's ratio (μ), and porosity (η), compressive strength (f c), and slake durability index (SDI). The measured rock properties are used in the rock mass model, and variations in v L in the rock bolt with different properties are investigated via numerical simulations. Results show that v p , v s , ρ , E , and f c are correlated significantly with v L (R 2 > 0.9), whereas η and SDI are moderately correlated with v L (R 2 > 0.7). However, a meaningful correlation between μ and v L is not obtained. The root mean square and mean absolute percentage errors are estimated to validate the correlation equations, and errors are not considered. Results of the t -test show that the calculated t -value is higher than the critical t -value, and the p -value is smaller than the significance level of 0.05, indicating that the correlation coefficient is significant. This study shows that the velocity of longitudinal waves in a rock bolt can be a useful indicator for predicting in-situ rock properties. [ABSTRACT FROM AUTHOR]