In the process of TBM tunneling in coal mine roadway, tunneling parameters have a significant impact on propulsion efficiency, energy consumption and tool wear, and there is a complex nonlinear coupling relationship among them. Traditional empirical parameter tuning or single-objective optimization methods are difficult to balance efficiency, cost and safety. Therefore, this paper proposes a multi-objective optimization method for TBM tunneling parameters based on intelligent integration algorithm, which integrates Grey Wolf Optimization (GWO), Radial Basis Function (RBF) neural network, Non-dominated Sorting Genetic Algorithm (NSGA-Ⅱ) and TOPSIS decision method. The nonlinear mapping relationship between tunneling parameters and propulsion speed, tunneling specific energy and tool wear is constructed by RBF neural network, and GWO algorithm is used to optimize its hyperparameters to improve the prediction accuracy. On this basis, NSGA-Ⅱ is used to achieve multi-objective optimization, and the Pareto optimal solution set is obtained. Finally, the optimal solution is screened by TOPSIS-entropy weight method. Based on the measured data of TBM tunneling process in Zhengtong Coal Industry, the verification is carried out. The results show that the propulsion speed is increased by 23.97 % under the optimal scheme. The specific energy of excavation decreased by 26.44 %; the tool wear is reduced by 43.67 %. The verification results show that the method can significantly reduce energy consumption and tool loss while improving tunneling efficiency, and realize the collaborative optimization of efficiency, cost and safety, which has good engineering applicability and promotion value.