Time Series InSAR Modeling And Deformation Estimation Of Drilling Water Solution Rock Salt Mine Based On CT-PIM

The dynamic deformation caused by an omnidirectional advanced drilling solution mining seriously threatens the safety of the life and property of the rock salt mine infrastructure and local residents.In addition,the continuous mining of rock salt mines is easy to lead to the mechanical changes of underground rock formations and water systems,resulting in the leakage of brine and environmental pollution.Therefore,it is very significat to monitor the long-term spatial-temporal deformation of rock salt mines.Multi-Temporal Interferometric Synthetic Aperture Radar(MT-InSAR)is an advanced InSAR technology developed in recent years.This technology has huge advantages over traditional InSAR technology,and has shown great potential in the surface deformation monitoring of mining area.In the process of MT-InS AR data processing,the establishment of deformation model is a crucial link.That is to say,the time-series function relationship between the phase and the deformation parameters is modeled.The accuracy of the deformation model directly affects the accuracy of the deformation estimation,and can also provide a reference for the interpretation of subsequent deformation results.However,most of the current time-series InSAR models of mines are composed of one or several pure empirical mathematical functions.Most of the model parameters are mathematical coefficients whose physical meaning is not clear enough.The deformation mechanism of mining subsidence is not considered in the deformation model.The purpose of this paper is to introduce the coordinate-time function(CT)of the dynamic probability integration model(PIM)into the traditional SBAS-InSAR linear rate model.The time series function model between InSAR unwrapping phase and probability integral prediction parameters is constructed.On this basis,the unknown parameters of the deformation are estimated to realize the time series deformation estimation of the mining area.The main research work includes:(1)The traditional SBAS-InSAR technology is used to analyze the time-series deformation of 24 Sentinel-1A SAR image covering drilling solution rock salt mine.The subsidence rate and cumulative settlement amount of high coherence points in time series are obtained.(2)Combined with the coordinate-time function of dynamic surface deformation,the time series function model between InSAR unwrapping phase and probability integral parameters(q,tan ?,s1,s2,k,n)is modeled to improve the linear rate model in traditional SBAS technology.(3)Carry out research on nonlinear parameter estimation algorithms for new models.The genetic algorithm(GA)is used to obtain the initial value of the probability integral predicted parameters,and then the simplex method(SM)is used to optimize the initial value.(4)Based on the new model,a case study of deformation estimation in the drilling water-soluble rock salt mine is carried out,and the simulation data and real data are used to carry out the experiment.The simulation experiment results show that the root mean square error(RMSE)between the time series deformation and the simulated true value at the high coherence point is ± 13 mm.It proves that the InSAR-CTPIM model in this paper is theoretically feasible.In the real experiment,the new model was applied to a rock salt mine in Changde,Hunan,and the time-series deformation from June 15,2015 to January 11,2017 was obtained,with the accumulated maximum settlement of 211 mm.Compared with the external level deformation results,the RMSE is about ± 31 mm.