Optimization of TBM Tunneling Parameters for Deep Buried Tunnel Based on Rock Cluster Grading and Strata Intelligent Identification

The intelligent decision-making of tunnel boring machine (TBM) tunneling parameters is of great significance for the construction of deep buried long tunnels. Providing the TBM main drivers with the optimal quantitative construction scheme under different surrounding rock conditions can significantly improve the tunneling efficiency and geological adaptability of TBM, and reduce the excavation energy consumption of TBM. This study first removes outliers from TBM tunneling data based on the Local Outlier Factor (LOF) algorithm. Then, a TBM tunnel surrounding rock clustering and grading system considering tunneling performance was constructed using Deep Subspace Clustering (DSC) algorithm. On this basis, a strata identification model based on Bayesian Optimization (BO)-Extreme Gradient Boosting (XGBoost) was constructed, with an identification accuracy of 92.5%, achieving accurate identification of the strata. Subsequently, a prediction model for the tunneling performance of surrounding rocks at all grades was constructed based on the Whale Optimization Algorithm (WOA)-Gated Recurrent Unit Neural Network (GRU) algorithm. The average R², MAPE, and RMSE of the model were 0.9475, 5.7132%, and 2.1041, respectively, indicating that the model has high prediction accuracy. Subsequently, the Non-dominated Sorting Genetic-III (NSGA-III) model for TBM tunneling parameters classification optimization was constructed using the WOA-GRU model as the fitness function. The Pareto optimality obtained by the NSGA-III model increased by an average of 37.67% and 23.32% compared to the original value and sample mean, respectively, significantly improving the tunneling performance of TBM. Finally, the difference between the proposed clustering classification method and the traditional BQ (basic quality indicators of rock mass) classification system in model prediction and optimization is quantified. The results show that the prediction and optimization results of the model under the condition of cluster classification are significantly better than that of BQ classification. The classification and decision method of TBM tunneling parameters proposed in this paper can provide quantitative guidance for TBM's main drivers to make construction decisions under different surrounding rock grades and improve TBM tunneling performance.