Abstract:Traditional artificial ground freezing (AGF) temperature field monitoring methods are limited by insufficient spatial resolution, making it challenging to dynamically characterize the development process of the frozen wall across the entire region. This paper proposes a non-destructive monitoring method based on Rayleigh wave dispersion analysis. By exploiting the sensitivity of Rayleigh waves to variations in the wave velocity of the medium, this method inverts dispersion curves to derive the evolution of the stratum"s wave velocity structure during the freezing process, thereby realizing dynamic quantitative characterization of the spatial development morphology of the frozen wall. Taking the freezing gap project of Zhengzhou Metro Line 7 as an example, a hydrothermal coupling model for artificial ground freezing was established, and theoretical reference values for the evolution of frozen wall thickness were calculated. By comparing the frozen wall thickness obtained from on-site Rayleigh wave inversion with the simulation results, the feasibility of applying Rayleigh wave technology to monitor frozen wall development was verified. The research results indicate the following: 1) The formation of the frozen wall significantly alters the stratum"s wave velocity structure. At 16 days of freezing, the Rayleigh wave dispersion curve exhibits a sudden increase in wave velocity in the 5–15 Hz range, confirming the initial formation of the frozen wall. At 30 days of freezing, the wave velocity further increases and the dispersion characteristics tend to stabilize, indicating that the frozen wall continues to develop and enters a stable strengthening stage. 2) Based on the inversion analysis of Rayleigh wave dispersion curves, the dynamic development thickness of the frozen wall can be accurately obtained. At 16 and 30 days of freezing, the inverted thicknesses are 9.5 m and 12.3 m, respectively, with relative errors of 5.0% and 2.5% compared to the theoretical reference values, and root mean square errors (RMSEs) of 12% and 9%, respectively. 3) Compared with finite element calculations, the 85% confidence interval of the inversion deviation covers most samples, with an overall average relative error of less than 15% and an average monitoring accuracy of 0.5 m, enabling continuous and effective monitoring of the frozen wall development process. This study provides a novel high-resolution, non-destructive monitoring method for artificial ground freezing engineering.