Prediction of Excavation and Settlement of Shallow-buried Tunnel Based on Radial Basis Function Neural Network of Human Brain

Abstract

Artificial neural network technology is to simulate the neural network structure of the human brain so as to solve the nonlinear engineering problem information processing system by establishing artificial neurons and sensors. The ground surface settlement caused by the excavation of shallow-buried tunnels is a research hotspot in the field of tunnels and underground engineering. By analyzing the influence factors that cause ground surface settlement, this study selects seven factors such as the cohesion and internal friction angle of surrounding rock as input and takes the measured value as the output to establish a ground surface settlement prediction model based on radial basis function neural network (RBFNN). A genetic algorithm is introduced to eliminate the slow convergence speed and local optimum of RBFNN. RBFNN prediction model can predict the ground surface settlement caused by the excavation of shallow-buried tunnels. The relative error of prediction is controlled within ±9%, and the prediction results meet the actual needs of the project. This study can provide reference for the expansion and application of RBFNN of human brain cortex in the engineering field.