Revealing Land Surface Deformation Over Backfilling Mining Area By Integrating Time Series InSAR Technology And Mining Subsidence Method

Backfill mining,a green subsidence reduction technology,prevents roof caving by using non-spontaneous combustion materials to fill the goaf.This method has been successfully applied in numerous mining areas.Yineng Coal mine in Shandong Province,China,has a designed production capacity of 450,000 tons per year,with coal deposits beneath the village in the mining area accounting for 71%of the basic reserves of the entire well field.The new ultra-high water material filling mining technology invented by China University of Mining and Technology has been adopted in the research area since late 2014.However,filling mining still generates surface deformation with certain magnitude and temporal-spatial characteristics,destabilizing surface structures and posing threats to the safety of mining residents.Therefore,regular monitoring of surface deformation in filling mining areas is essential for preventing and controlling geological disasters and protecting people’s lives and property.Compared with conventional monitoring methods for surface deformation in mining areas,Interferometric Synthetic Aperture Radar(InSAR)can rapidly acquire information on surface deformation and its timing in a wide range and with high resolution at a low cost.Integrating InSAR with prediction models of mining surface deformation,such as time function and Probability Integration Model(PIM),allows for the acquisition of complete and high-precision surface deformation in time and space.This provides technical support for extracting temporal and spatial characteristics of mining surface deformation and offers the possibility for remote sensing monitoring of subsidence reduction effects in filling mining areas.However,research on extracting temporal and spatial characteristics of surface deformation and remote sensing monitoring of subsidence reduction effects in filling and mining areas is still lacking.To address this issue,this paper focuses on the Yineng backfill mining area and its surrounding mining area,and covers the following main aspects:(1)Timing processing of DS InSAR technology in the study area:This paper employs advanced DS InSAR technology to process the time sequence of 26 Sentinel-1A images covering the study area from January 10,2019,to July 3,2020,addressing the issue of low-density monitoring points of SBAS InSAR results.The technology integrates the Hypothesis Test of Confidence Interval(HTCI),particle recognition algorithm,space adaptive filtering,EMI(Eigen-decomposition-based)Maximum-likelihood-estimator of the Interferometric phase,and SBAS method.Compared with the traditional SBAS InSAR(Small Baselines InSAR)method,the DS InSAR strategy used in this paper significantly improves the coherence point density(by about 2 times)and the precision of the deformation value(RMSE decreased by about 4.00mm).(2)Extraction of spatio-temporal evolution law of surface deformation in the study area:Based on the accuracy verification of DS InSAR results,the study finds that there are eight settlement areas in the mining area during the study period.Apart from the Yineng filling mining area,the area affected by subsidence in the surrounding mining area reaches 14.26km~2.The settlement area of Yineng mining area is primarily located in the first mining area,with surface deformation occurring above the surface of the six filling mining faces within the mining area,conforming to the mining subsidence law of the working face.(3)Construction of high-precision dynamic prediction model of surface deformation in filling and mining area:Considering the limitations of traditional dynamic prediction models for surface deformation in mining areas,which are not suitable for filling mining areas and have time function parameters overly dependent on measured parameters,this paper introduces the equivalent mining height theory.Combining this with the arctangent time function,which is independent of measured parameters in the mining area,a mining subsidence dynamic prediction model suitable for filling mining conditions is constructed based on the widely-used PIM model.(4)Monitoring of subsidence reduction effect in filling mining area by remote sensing:Using CG1312 working face as an example,the dynamic prediction model of surface deformation under filling mining conditions is employed to fully account for the residual deformation of the surface.The high-precision predicted value of DS InSAR,when the surface reaches a stable state,is taken as the boundary constraint condition of the model.In combination with artificial intelligence algorithms,the predicted complete surface deformation field under filling mining and non-filling mining conditions is compared to realize remote sensing monitoring of subsidence reduction effects of filling mining.The results demonstrate that the filling mining technology effectively reduces the range(22.00%)and magnitude(61.50%)of surface settlement,considerably mitigating the potential threat to surface houses.There are 33 pictures,6 tables and 104 references in this thesis.