Research On Time-series InSAR Mining Area Surface Deformation Prediction Combined With Kalman Filter

Since the early days of the founding of the people’s Republic of China,the demand for coal in China is increasing day by day,and the accompanying problem is the surface subsidence caused by the increasing intensity of mining,resulting in a series of irreversible geological disasters around the mining area.therefore,mastering the law of surface deformation of the mining face and its goaf and predicting the surface deformation is of great significance to the safety of people’s production and life.Taking the mined-out area of Zhalainuoer mining area as the research target area,a time series InSAR surface deformation prediction method combined with Kalman filter is proposed in this paper.Through the analysis of the technical principle of SBAS-InSAR,on the basis of the original Kalman filter model,more measured values are interpolated to filter,and the state equation and observation equation of the filter are derived.Through the comparison and analysis of different interpolation algorithms,the cubic polynomial interpolation algorithm is selected to improve the filter according to the results.To solve the problem that SBAS-InSAR technology can only provide surface deformation information but can not predict the deformation trend,and the current prediction methods are not suitable for large-scale prediction.The main research work and results are summarized as follows:(1)A total of 68 scenes of Sentinel-1B satellite descending orbit images from January 2019 to December 2021 were selected,and SBAS-InSAR technology was used to analyze the Zhalainuoer mining area and its surrounding goafs.Deformation monitoring.Aiming at the problem of low accuracy of manual selection of control points,the high-precision PS points made by PS-InSAR are used to replace manual selection of control points,which improves the monitoring accuracy.(2)In view of the fact that SBAS-InSAR technology can only provide surface deformation information but cannot predict the deformation trend,and the current prediction method is not suitable for large-scale prediction,this paper proposes a time-series InSAR surface deformation prediction method combined with Kalman filtering.Methods By analyzing the technical principle of SBAS-InSAR,the state equation and observation equation were derived,and the real monitoring results were used to analyze and verify the accuracy.(3)To solve the problem of insufficient experimental data for prediction,this paper uses interpolation method for data expansion.Since the 68 scene images cover three years of data,in order to make the interpolation data more reasonable,this paper sets the interval to 0.2,and the data The number expanded to about 350,which is close to the number of days corresponding to three years.In this paper,four interpolation methods including Hermitian interpolation,cubic spline interpolation,monolinear interpolation,and cubic polynomial interpolation were selected for comparative analysis.After comparison,the accuracy of the cubic polynomial interpolation method is the highest,and the number of points with a difference range below 3mm reaches81.78%.(4)The Kalman filter is used to predict the data set expanded by the interpolation algorithm,and the Knothe time function model prediction method is introduced to compare the results.In this paper,two groups are randomly selected from the predicted point results for accuracy evaluation.The results show that the average deviation between the predicted value and the observed value is less than 1mm.Compared with the Knothe time function model method,the accuracy of this algorithm is higher and the prediction effect is better.Finally,this paper selects four groups of prediction time corresponding to the surface results,the results are all less than 3mm,the trend is basically consistent with the prediction point results,the prediction results are reliable.The paper has 44 figures,9 tables,and 75 references.