Comparative Experiment on Intelligent Monitoring Methods to Improve the Stability of Mountain Tunnel Construction

Aiming at the problem of poor real-time performance and insufficient precision of traditional monitoring methods in mountain tunnel construction, which leads to delayed warning of deformation of support structures, FBG fiber optic sensor arrays were arranged every 10 m along the tunnel axis, and 12 groups of three-dimensional laser scanning stations were set up at the same time; second-level data acquisition and transmission were achieved through edge computing gateway; LSTM time series prediction model including geological radar data and construction parameters was constructed; and 72 hours of continuous monitoring was carried out in three high-risk sections. The results show that the displacement monitoring accuracy of the fiber optic sensing system reaches 0.3 mm, the false alarm rate of the proposed model is reduced, and the optimal support decision response time is 8.7 s, providing a real-time dynamic stability control solution for complex geological tunnel construction.