Monitoring Of Surface Fracture Activity In Goaf Based On Multi-source Data

Resource development is an important part of China's western development.Jinchuan copper nickel mine in Gansu Province,as the largest large-scale polymetallic symbiotic copper sulfide mine in China,has a decisive strategic position.With the development of underground resource exploitation,the overlying surrounding rock in the West Second Mining Area of Longshou Mine in Jinchuan copper nickel mine has become unstable,and a large number of circular and intricate fracture groups and large collapse pits have appeared on the surface of the goaf.The monitoring of surface deformation in this area was originally realized through the deployment of GNSS network.Due to the serious surface deformation in the goaf,the UAV aerial survey technology with high safety was used for dynamic visual monitoring of surface cracks,but the accuracy of deformation information obtained was low.InSAR technology has the advantages of continuous,large area and high precision,and is widely used in various deformation monitoring fields.To solve the above problems,this paper uses UAV aerial survey technology and InSAR technology as the main monitoring means,and uses GNSS technology to do comparative verification and analysis.In the same time span,the object-oriented information extraction technology is used to automatically extract the surface fracture information from two UAV images.Combined with the two-dimensional surface deformation field obtained from SAR images based on the difference between time domain and space domain,the activity characteristics of surface fractures are analyzed.Using multisource data for monitoring,we can not only obtain the distribution and change law of surface cracks from the macro perspective,but also conduct subtle deformation analysis to comprehensively analyze the activity law of surface cracks.The main research work and results are as follows:(1)UAV image data in the study area on August 19,2020 and June 27,2021 were obtained through UAV aerial survey technology.The improved unsharp masking obtained by Laplace operator,Sobel gradient operator,GHP filter and Gamma transform is used to enhance the UAV image.The extraction rule set is constructed according to the spectral and shape features of the surface cracks to extract the object-oriented surface cracks.The accuracy of surface crack extraction is evaluated by five indexes: the number of automatically extracted pieces,the number of pieces extraction rate,the total length of automatically extracted pieces,the length extraction rate and the position overlap rate.After statistical analysis,the overall extraction rate of surface cracks based on object-oriented technology has reached more than80%,and good results have been achieved.(2)Using SBAS-InSAR technology,the sentinel-1A data of 25 scene orbit ascent and 27 scene orbit descent are processed,and the deformation information of the orbit ascent and descent in the study area from August 19,2020 to June 27,2021 is obtained.In order to improve the accuracy of two-dimensional solution,Pchip method is used for time registration of SAR images of lift orbit,and the two-dimensional surface deformation information is obtained through geometric solution.Then the empirical Bayes Kriging method is used to encrypt the two-dimensional surface deformation,and finally the two-dimensional surface deformation field based on time domain and space domain interpolation is obtained.(3)Combined with multi-source data,this paper analyzes the active characteristics of surface fractures.Firstly,the UAV image is used to automatically extract the information of two phases of surface fractures and analyze their distribution and change characteristics;Secondly,the GNSS data is used to compare and analyze the two-dimensional surface deformation field after interpolation in this paper.Through comparison,it is found that the maximum absolute error of the two-dimensional deformation field calculated in this paper is30.62 mm and the minimum absolute error is 0.69 mm.Then,the two sides of the main surface fractures and the vertical section lines are extracted from the two-dimensional surface deformation field for deformation analysis.Finally,based on the comprehensive analysis of multi-source data,the activity law of surface fractures in the study area is obtained.It is found that the surface fractures are relatively active near the edge steep slope of the irregular deformation basin,and there is obvious uneven deformation on both sides of the surface fractures with strong activity.Similar to the vertical deformation,the horizontal displacement also has important indication significance for the surface fractures.