| description abstract | Blasting parameters are crucial factors that directly affect the quality of tunnel excavation. To achieve optimal blasting results, it is necessary to continuously optimize the blasting parameters throughout the construction process, taking into account geological conditions. However, current research mainly focuses on optimizing single-type borehole parameters and fails to simultaneously address the requirements for minimizing overexcavation, underexcavation, and fragment size. This study proposes an intelligent optimization method for blasting construction parameters that combines support vector regression (SVR) with the Non-Dominated Sorting Genetic Algorithm II (NSGA-II). Through grid search and genetic optimization algorithms, the SVR regression model is refined, establishing an accurate nonlinear mapping relationship between the borehole parameters of peripheral holes, auxiliary holes, and slot holes, and the resulting blasting effects. The NSGA-II algorithm is then employed to search for the Pareto optimal set of blasting construction parameters, with the goal of minimizing average linear overexcavation and the maximum fragment diameter. The technique for order of preference by similarity to the ideal solution (TOPSIS) method is used for multiattribute decision-making to identify the optimal blasting plan. The results show that the SVR model, optimized by the genetic algorithm, provides high prediction accuracy for blasting construction parameters, with determination coefficients of 0.89 and 0.97. Multiobjective optimization of blasting parameters using NSGA-II explores the effects of different parameter combinations on tunnel blasting outcomes. In designing and optimizing blasting parameters, particular attention should be paid to the optimization of peripheral and slot hole parameters to effectively control overexcavation, underexcavation, and fragment size. The intelligent optimization method proposed in this study, which integrates advanced intelligent algorithms with professional blasting construction knowledge, forms an efficient and intelligent optimization system. This system enhances the feasibility and accuracy of blasting construction plans. | |
