An Automatic Spatial-Temporal Evolution Inversion Method of Mining Goaf Based on the Improved Hotspot Analysis and Probability Integral Method

Abstract

Mining activities may cause severe ground subsidence, endangering surface structures, and farmlands. Therefore, the acquisition of spatial-temporal evolution of the mining goaf is of great significance. Hotspot analysis (HSA) based on the Getis-Ord Gi statistics has been utilized to identify the areas with a rapid deformation rate. In this article, we propose an improved HSA (IHSA) method for automatic extraction of the surface subsidence caused by the mining goaf. In addition, we design a comprehensive workflow for the automatic spatial-temporal evolution inversion of surface deformation induced by the mining goaf. First, time series interferometric synthetic aperture radar (InSAR) is utilized to generate the surface deformation of the mining area. Then, the IHSA method is used for the automatic identification of the mining goaf. Finally, the total least-squares probability integral method (TLS-PIM) is applied for goaf inversion based on the extracted deformation information. For this study, the Wuxiang is selected as the study area. We have compared the IHSA method with four methods using five indicators, and likewise, we have compared the TLS-PIM method with four methods in terms of their correlation with the InSAR results in the strike and dip directions. The experimental results demonstrate the superiority of our method as a support for the geological hazard investigation and mine safety supervision department.