DOI: 10.11799/ce202509018
Research on multi-objective optimization method of TBM tunneling parameters based on intelligent integration algorithm
SHEN Yuanwei
ZHU Hao
CAO Mengxuan
FENG Changru
ZHU Mengyuan
ZHANG Chaofan
Abstract：During TBM tunneling in coal mine roadways,tunneling parameters significantly influence on the advancement efficiency,energy consumption and cutter wear,with complex nonlinear coupling existing among these three factors.Traditional empirical tuning or single-objective optimization methods often fail to balance efficiency,cost,and safety.To address this,we proposed a multi-level intelligent optimization method integrating GWO-RBF-NSGAⅡ-TOPSIS,combining metaheuristic optimization,nonlinear modeling,multi-objective evolution,and decision ranking to achieve full-process control of TBM parameters.An RBF neural network was used to model the nonlinear relationship between tunneling parameters and advancement speed,specific energy,and cutter wear,with its hyperparameters optimized by the Grey Wolf Optimizer(GWO)to improve prediction accuracy.NSGA Ⅱ was then applied for multi-objective optimization to obtain a Pareto-optimal solution set,from which the optimal scheme was selected using the TOPSIS method with entropy weighting.According to the field data from TBM tunneling at Zhengtong Coal Mine,the optimal scheme increased the advancement speed by 23.97%,reduced specific energy by 26.44%,and decreased cutter wear by 43.67%.This method can significantly improve efficiency while reducing energy consumption and tool wear,achieving coordinated optimization of efficiency,cost,and safety,and demonstrating strong engineering applicability and promotion potential.
Keywords：TBMGWORBF neural networkNSGAⅡTOPSISmulti-objective optimizationoptimization of tunneling parameterstunneling specific energy
Publication Date：2025-09-20
Online Publishing Date：2025-11-03(First online date of this platform, not the publication date of the document)
Pages：9( 129-137 )