Surface Deformation Analysis Of Iron Ore Mining Based On InSAR Technology

For the monitoring of iron ore area,although the conventional measurement method has high monitoring accuracy and flexible measurement method,it also has many shortcomings,such as point-based monitoring,unable to reflect the overall deformation trend of the mining area;long monitoring period,easily destroyed measurement points,resulting in late monitoring;consuming a lot of manpower and material resources,and high monitoring costs.In recent years,new measurement methods(IBIS-M,GB-InSAR,etc.)have emerged,although they have high measurement accuracy and can be used for face-to-face monitoring,but the instrument is expensive,the maintenance cost is high,and the monitoring area is small.Compared with the above two technologies,space-borne InSAR technology has the advantages of full-time,all-weather,large-scale,high precision and low cost,and can provide long-term monitoring of surface deformation in mining areas.Through this technology,the surface deformation characteristics of iron ore mining and the deformation of key deformation areas in each mining area are analyzed.After conventional D-InSAR processing,there are still some phase effects,such as atmospheric delay and linear residual,which seriously restrict the monitoring accuracy of D-InSAR in small deformation magnitude.In order to enable D-InSAR technology to monitor iron ore area more accurately,this paper uses GACOS atmospheric model to remove the non-linear atmospheric phase in the process of its processing,and extracts the linear residual phase of D-InSAR monitoring area by GPS point elevation interception method and high coherence continuous area fixed window median method,estimates the overall residual phase by these residual phase mean values,and aligns the line.Sex residual phase is removed.The processing results are compared with the measured leveling data,and the error reaches(+5.6mm).The GPS data in the test area are used to further verify the accuracy of the millimeter level monitoring.Aiming at the problem of large error in the conventional de-entanglement method of D-InSAR in mining area,the GVF-Snake boundary detection model is used to de-entanglement the winding phase in D-InSAR.In the process of de-entanglement,high-pass filtering and cubic spline interpolation are used to remove the "2?" jumping boundary points for the first time.By comparing with other conventional methods,it is found that the conventional quality chart guidance method and Goldstein branch-cutting method have large error or even can't be unwrapped for large deformation magnitude phase in open-pit mining.Only the method in this paper and 3D unwrapping can correctly unwrap this area.Compared with the measured leveling data after atmospheric denoising,linear residual phase removal and phase rotation vertical deformations,the final deformations obtained by this method and 3D unwrapping can achieve millimeter-level monitoring accuracy,but this method is superior to the 3D unwrapping method.The atmospheric phase removed by traditional PS-lnSAR technology is estimated based on the characteristics of vertical delayed stratification and strong terrain correlation.It does not consider the influence of the non-linear atmospheric phase caused by turbulent mixing delay.When the non-linear error is too large,the processing error will be too large.Based on the above problems,this paper obtains the atmospheric delay phases of four models by TRAIN software,and compares them with the real atmospheric delay data assessed by GPS points.It is found that GACOS atmospheric delay data has the highest accuracy.GACOS atmospheric delay(taking into account the non-linear atmosphere)is used to denoise the atmosphere instead of assessing the non-linear atmosphere.Using GPS value in U direction in NEU coordinate system to analyze the results of taking into account and not considering the non-linear atmosphere,it is found that both of them can achieve the monitoring accuracy of millimeter level,but the results of taking into account the non-linear atmosphere are more accurate.When the precise D-InSAR processing results are taken as true values,the errors of PS-InSAR monitoring results considering and not considering the non-linear atmospheric effects are(+2.7mm)and(+8.6mm),respectively.The PS-InSAR monitoring results considering the non-linear atmospheric effects are more accurate and reliable.