Search

Understanding the variation patterns of tunnel boring machine (TBM) operational parameters is crucial for assessing engineering geological conditions and quality grades of surrounding rock within tunnels. Studying the multifractal characteristics of the TBM operational parameters can help identify the patterns, but the relevant research has not yet been explored. This paper proposed a novel classification model for quality grades of surrounding rock in TBM tunnels based on multifractal analysis theory. Initially, the statistical characteristics of eight TBM cycle data with different grades of surrounding rock were explored. Subsequently, the method of calculating and analyzing the multifractal characteristic parameters of the TBM operational data was deduced and summarized. The research results showed that the TBM operational parameters of cutterhead torque, total thrust, advance rate, and cutterhead rotation speed have significant multifractal characteristics. Its multifractal dimension, midpoint slope of the generalized fractal spectrum, and singularity strength range can be used to evaluate the surrounding rock grades of the tunnel. Finally, a novel classification model for the tunnel surrounding rocks based on the multifractal characteristic parameters was proposed using the multiple linear regression method, and the model was verified through four TBM cycle data containing different surrounding rock grades. The results showed that the proposed multifractal-based classification model for tunnel surrounding rocks has high accuracy and applicability. This study not only achieves multifractal feature representation and surrounding rock classification for TBM operational parameters but also holds the potential for adaptive adjustment of TBM operational parameters and automated tunneling applications.

Blasting operation plays a crucial role in open-pit mining projects and significantly affects the mining efficiency and operational costs. However, blasting operations are usually accompanied by several side effects, such as backbreak and flyrock hazards, which result in wasting explosive energy, damage to the surrounding environment, and poor rock fragmentation. Due to the complex nonlinear relationship between the blast pattern parameters, rock characteristics, and foregoing hazards, the conventional criteria and simple regression analysis cannot provide highly accurate and reliable predictive models. In this study, based on the compiled 152 datasets from four different open-pit mines in Iran, six machine learning (ML) algorithms, including K-nearest neighbor (KNN), random forest (RF), eXtreme gradient boosting (XGBoost), support vector machine (SVM), decision tree (DT), and linear regression (LR), were used to develop robust models for predicting backbreak, flyrock, and rock fragmentation size. Three different datasets containing different combinations of inputs were defined for model development, and the prediction performance of the models was evaluated using R2and root mean square error (RMSE) indices. The results showed that KNN, RF, and XGBoost algorithms outperform others in predicting fragmentation, flyrock, and backbreak, respectively. Furthermore, the parameters of burden, spacing, powder factor, sub-drilling, hole depth, and uniaxial compressive strength were identified as the best set of inputs for ML-based model development. The sensitivity analyses also revealed that blast design parameters of stemming, hole diameter, and sub-drilling have the highest impact on the prediction of flyrock, rock fragmentation, and backbreak, respectively. Finally, the SHapley Additive exPlanations (SHAP) analysis significantly improved the interpretability of the developed ML models and provided more insight regarding the intricate relationships between the parameters.

Because of the release of in situ rock stress, rocks surrounding deep underground excavations are mostly driven into a post-peak damage state. Thus, investigations on cyclic loading-induced damage of rocks are of paramount significance in terms of the long-term stability of rock structures. In this study, using a damage-controlled testing method, the effect of post-peak cyclic loading on the failure behavior of different rock types was comprehensively evaluated. This investigation revealed that post-peak cyclic loading imposes no significant influence on the overall failure behavior of rocks. The evolution of damage variables also indicated that the large portion of axial strain damage (Da) occurs during the first cycle at the peak strength, whereas lateral strain damage (Dl) evolved more rapidly than Dadid in the post-peak regime under further loading and unloading cycles. The cumulative permanent axial strain (∑εap) evolved nonlinearly with the axial stress ratio (σi/σa−peak) in the post-peak regime, which was manifested as sparse-dense-sparse hysteretic loops. A rapid reduction in the crack damage stress ratio was observed for rocks in the initial sparse hysteretic loops, which was followed by a lower decrease rate in the subsequent dense hysteretic loops, and an almost constant value was finally reached in the last sparse hysteretic loops. Furthermore, tangent Young’s modulus (Etan) and Poisson’s ratio (υ) evolved in three main stages: One, Etanfirst increased and then decreased in the initial sparse hysteretic loops, while υcontinuously increased, representing significant deformation in the lateral direction. Two, Etandecreased in the dense hysteretic loops, while υincreased at a higher rate for stronger rocks because of the decrease in Daaccumulation. Three, the increase rate of υdeclined in the second set of sparse hysteretic loops, and Etandecreased until complete failure occurred.