Monitoring And Recognition Of Datong Coalfield Subsidence Area Based On InSAR Technology

In order to carry out modernization,a large amount of underground coal resources have been mined and utilized,and large-scale goaf has emerged.This disrupted the original stress balance on the ground,and eventually resulted in large-scale surface subsidence in the mining area.The emergence of coal mining subsidence areas has a huge impact on the natural environment and people’s production activities and living environment.Therefore,the monitoring,identification and research of coal mining subsidence areas are particularly important.Its main research work and contents are as follows:(1)The basic principles of synthetic aperture radar interferometry and coherence are introduced in detail.From the principle of coherence it is shown that the incoherence can be used to identify the subsidence area.Radarsat-2 and sentinel-1data were prepared,and the deformation result graph and coherence image were obtained by using the method of small baseline subset.And the results of the two data were compared and found to have a certain consistency,indicating the correctness of the data processing.(2)According to the deformation results of radarsat-2,the basic conditions of the overall subsidence of Datong coalfield and eight key deformation mining areas were analyzed.At the same time,the time change maps of the largest deformation points in the eight mining areas were obtained.It can be seen that from 2016 to 2019,the total subsidence area of Datong coalfield is 417.58 km~2,and 75.91%of the subsidence area is below 20 mm,and 4.38%of the land subsidence is over 40 mm.It is concluded that the overall subsidence of Datong coalfield is mild,and a small part of the subsidence in the mining area is quite severe.The results showed that the subsidence area of the eight mining areas accounted for 47.01%of the total,indicating that the subsidence area was mainly concentrated in these eight mining areas.In addition,the cumulative settlements of the eight monitoring points in the mining area all increase with time.Among them,the largest settlement occurred in Yongdingzhuang Coal Industry,with a settlement of up to 94.77 mm.(3)In order to analyze the deformation characteristics of the subsidence area,the multi-temporal surface deformation information of four typical coal mines were obtained.The deformation characteristics of the subsidence area are summarized from the four maps.Another mine area with particularly obvious deformation characteristics was used to further analyze and verify the deformation characteristics of the subsidence area from the perspective of the plane deformation area and profile deformation information.(4)The method of deep learning is used to classify the subsidence area based on the deformation rate diagram and the coherence image,so as to realize the purpose of extracting the subsidence area automatically.