Error Analysis And Research On Processing Method Based On Monitoring Data Of GB-SAR Open-pit Mine Slope

Ground-based synthetic aperture radar is the main monitoring equipment in deformation monitoring fields such as open-pit mine slopes,landslides,and dam stability.It has the advantages of all-day monitoring,high accuracy and large monitoring range.However,the monitoring accuracy of GB-SAR is greatly affected by various error factors such as speckle noise,orbit,and atmosphere.In order to improve the accuracy of GB-SAR deformation monitoring,the following researches are carried out in the study:(1)In order to solve the speckle noise problem in GB-SAR interference images.The adaptive threshold wavelet denoising algorithm based on Stein unbiased likelihood estimation is used to filter the interference image.By comparing and analyzing the interference image before processing and the interference image after processing.The result shows that the method can not only effectively remove the noise phase in the interference image,but also can maintain the effective phase signal of the interference image.(2)A method of extracting high-quality PS(Permanent scatterer)points using quadruple threshold is proposed to solve the problem of inaccurate extraction of high-quality PS points in GB-SAR interference images.The results show that the method can accurately extract high quality PS points.The accuracy of the model has been improved.(3)Aiming at the phenomenon of orbit error and atmospheric delay error in GB-SAR interference image of Malanzhuang open-pit mine.The method for correcting the phase error in the azimuth direction and the delay error caused by the atmospheric change in the distance direction based on the multiple regression model was proposed.By comparing and analyzing the interference image before processing and the interference image after processing.Compare and analyze the results before and after processing.The results show that the method can accurately correct the monitoring error of GB-SAR,and the accuracy is improved to 83.3%compared with the model before correction.(4)In order to solve the problem that it is difficult to correct the atmospheric delay error of the GB-SAR interference image due to the complex atmospheric convection effect existing in the deep concave open-pit mine.The characteristics of meteorological changes in deep concave open-pit mines during the sunrise and sunset periods were analyzed based on data from the Dagushan weather station.The reason for the abnormal distribution of atmospheric delay errors in GB-SAR interference images was revealed.Meteorological models,multiple regression models,and random forest models were established to correct the atmospheric delay error of GB-SAR.The accuracy and universality of each model were compared and evaluated based on multi-function angular reflector displacement experiment.A method based on the combination of multiple regression models and random forest models was proposed which was based on comprehensive analysis.The results show that the method can accurately correct the atmospheric delay error of GB-SAR in deep concave open-pit pits which can ensure the monitoring accuracy better than 0.3mm.In summary,the GB-SAR small-period monitoring data processing algorithm was proposed which was based on the in-depth analysis of the GB-SAR open-pit mine slope monitoring data processing and the determination of the causes and correction methods of GBSAR monitoring errors.The method successfully monitor the abnormal areas of two open-pit mine test sites by GB-SAR monitoring data real-time fast processing and high-precision deformation acquisition.It lays a foundation for the accurate warning of geological disasters in open pit mines.