Construction Risk Monitoring Method of Subway Foundation Pit Engineering Based on Simulated Annealing Neural Network

Abstract

Traditional subway foundation pit engineering construction risk monitoring methods have slow convergence speed when solving large-scale practical problems, which affects the accuracy of monitoring. Therefore, a subway foundation pit engineering construction risk monitoring method based on simulated annealing neural network is designed. By identifying the accident risk sources of foundation pit engineering, understanding the accident causes and prevention mechanism among human, machine and environment, the classification of risk sources of foundation pit engineering is obtained, and the safety risk monitoring index system is constructed. The monitoring indicators are analyzed in detail, and the annealing neural network is optimized, and the process of double-layer simulated annealing algorithm is designed to realize risk monitoring. In the case simulation experiment, the designed monitoring method and the traditional method are used to monitor the project. The monitoring experimental results show that the proposed method can accurately predict the deformation of the subway tunnel through the monitoring data of the deep foundation pit construction adjacent to the existing subway tunnel.