Surface Settlement Prediction of Rectangular Pipe-Jacking Tunnel Based on the Machine-Learning Algorithm

• DOI: 10.1061/jpsea2.pseng-1453
• Publication Date: 2023-11-29T09:55:08Z
• Authors: Da Hu; Yongjia Hu; Shun Yi; Xiaoqiang Liang; Yongsuo Li; Xian Yang
• Keywords: Jacking; Settlement (finance)

The construction disturbance mechanism of rectangular pipe-jacking tunnels is more intricate than that circular tunnels, leading to potential issues such as excessive accumulation and deformation the surrounding formation, which can result in engineering disasters. However, there currently a lack reliable methods for predicting these disturbances. Machine-learning techniques have capability analyze influence multiple independent variables on dependent variable, offering new approach surface settlement tunnels. To address sensitivity existing machine-learning models initial parameters, an improved particle swarm optimization (IPSO) method employed. This incorporates adaptive mutation technique, inertia weight, postoptimization mutant particles enhance size determine probability obtaining optimal value. By leveraging strong mapping nonlinear fitting abilities backpropagation (BP) algorithm, IPSO-BP algorithm model developed compared with BP, support vector machine, random forest (RF) using actual monitoring data. findings indicate presence specific noise data, prediction demonstrates enhanced accuracy 26%, 25%, 10% left amplitude. serve valuable reference similar projects.