Prediction of EPB Shield Tunneling Advance Rate in Mixed Ground Condition Using Optimized BPNN Model

Tunneling in mixed ground often results severe torque fluctuations and a low advance rate. Therefore, choosing reasonable set of parameters for accurate rate prediction is paramount to reduce cutter wear improve tunneling efficiency. However, since the geological conditions are diverse uncertain, (AR) EPB shield significantly more difficult than that homogeneous (i.e., full-face hard-rock ground). In addition, operating can be subjective suboptimal, each them has some intricate influence on AR. this paper, an optimized back-propagation neural network by genetic algorithm (BPNN-GA) was proposed parameter selection AR prediction, four typical machine learning methods were used comparison. Five processing strategies with different input also compared determine optimum conditions. The models adopted case study Nanjing Metro Line S6 project, total 1188 rings datasets study. showed modified BPNN could effectively implemented prediction. It concluded Strategy B—i.e., using composite ratio layer as input—was best strategy Hence, high correlation between measured predicted observed coefficient (R2) 0.920.