An adaptive operating parameters decision-making method for shield machine considering geological environment

Abstract

The performance of shield machine is sensitive to geological conditions and its operation depends on driver experience. Based on D-ResNet network and improved inertia weight adaptive constrained particle swarm optimization algorithm (IWACPSO), this study proposes an adaptive operating parameters decision-making method for shield machine considering geological environment. First, a novel D-ResNet network is proposed to establish the mapping models between geological type, operating parameters and tunneling speed, cutterhead torque, respectively. Then, we design the evaluation indicators of tunneling performance considering construction efficiency and considering both construction efficiency and energy consumption, respectively, and build the mapping model between operating parameters and performance evaluation indicators. On this basis, the optimization equation with operation parameters as control variables and improving tunneling performance as the goal is established. In order to solve the decision results of optimal operating parameters under various geological types, we present an enhanced inertia weight adaptive constrained particle swarm optimization algorithm. Finally, the proposed D-ResNet and IWACPSO-based decision-making method are verified by the actual construction data of subway project in Singapore. The results show that the R2 values of the proposed D-ResNet network for predicting the tunneling speed and cutterhead torque reach 0.98 and 0.96, respectively, which are much better than AdaBoost, support vector regression, multiple linear regression and deep neural network. The IWACPSO-based decision-making method could accurately solve the optimal solution of operating parameters, and the optimization time is much lower than that of ant colony algorithm, genetic algorithm and exhaustive method. Moreover, operation parameter decision-making method considering construction efficiency increased the tunneling speed by 61.02% on average, and the operation parameter decision-making method considering construction efficiency and energy consumption increased the tunneling speed by 59.55% on average, while the tunneling specific energy decreased by 20.50% on average compared with the former.