Application of Gray-Markov Model to Land Subsidence Monitoring of a Mining Area

Debao Yuan,Ling Zhang,Chengxing Geng,Zhenchao Zhang

doi:10.1109/access.2021.3106144

Abstract

Land subsidence monitoring in mining areas is one of the main applications of surface deformation monitoring, which is of great significance for safety production. Using the IPTA (Interferometric Point Target Analysis) time-series InSAR (Interferometry Synthetic Aperture Radar) method, land subsidence data from the new exploration area in the Weizhou mining area were analyzed and compared with static GPS (Global Positioning System) monitoring data for 2017–2020. Gray-Markov model was established by combining the gray prediction model with the Markov model to predict the surface subsidence of the mining area. The results show that (1) InSAR data have high accuracy and application potential in prediction of long-term surface deformation in mining areas; (2) The Gray-Markov model can better reflect the volatility and practicality of subsidence data in mining areas; (3) The prediction results have high accuracy, and the Gray-Markov model can serve as an effective guide for long-term surface deformation monitoring and safety management.

Highlights

Coal has always occupied a high proportion in China's energy consumption
Land subsidence monitoring in mining areas is one of the main applications of surface deformation monitoring, which is of great significance for safety production
The results show that (1) InSAR data have high accuracy and application potential in prediction of long-term surface deformation in mining areas; (2) The Gray-Markov model can better reflect the volatility and practicality of subsidence data in mining areas; (3) The prediction results have high accuracy, and the Gray-Markov model can serve as an effective guide for long-term surface deformation monitoring and safety management

Summary

INTRODUCTION

Coal has always occupied a high proportion in China's energy consumption. China is rich in coal resources, mainly distributed in Shanxi, Inner Mongolia and Shaanxi. Studies have shown that the deformation monitoring data in mining areas has a certain gray characteristics,but most of these studies are based on observatory leveling or GPS measurements It is independent of the subsequent evolution links, and cannot form a complete and systematic monitoring and prediction model [15]. This paper takes the new exploration area of Weizhou mining area as the research area, considering the dynamic nature of the actual surface subsidence, we collected InSAR technology and Gray-Markov model, proposed a new dynamic prediction method, processed the Sentinel-1 satellite data from 2017.01 to 2020.12 with IPTA time series InSAR technology, and compared the result with the GPS data to verify the accuracy of the InSAR data. The IPTA method uses the two-dimensional linear phase model, completes the separation and removal of various errors by constantly returning to the iterative scheme, and obtains the linear deformation rate of the coherent point target [16].

Markov Prediction Model

State Division

Determine The Optimal Value

Findings

CONCLUSION