| Coal,as the main energy source for China,have made great contributions to the economic development,but meanwhile it is also a main pollution source,causing a series of environmental disasters.It is of great significance to monitor land subsidence accurately and master the laws of deformation for rationally exploiting underground coal resources,restoring ecological environment of mining area,predicting and preventing disasters.Some traditional situ measurements are ineffective,low security,high cost,consuming intensive work,and cannot satisfy the application requirements.Interferometric synthetic aperture radar(In SAR)has been a new method for land subsidence in mining area for its great imaging ability at all time and all weather with high resolution,wide coverage and convenient repeated observation.However,the application of In SAR technique is limited by the temporal and spatial baseline,atmospheric phase screen,orbital error and so on.Due to land subsidence in mining area characterized by large deformation gradient,rapid motion,there are still many problems in land subsidence monitoring in mining area using In SAR.In addition,many study areas collecting multi-sensor SAR datasets along with the development of SAR satellites,how to integrate these SAR datasets with different spatial and temporal resolution is a problem that In SAR monitoring in mining area need to face directly.Therefore aiming at the shortage of In SAR technique applied in mining area,this dissertation study the ground subsidence in mining area using multi spaceborne SAR images,according to the actual situation of Fengfeng and Xuzhou mining area in Hebei and Jiangsu Province.This dissertation solves the problems of multi-sensor SAR image fusion processing,non-linear dynamical movement monitoring in mining area.The main research works and results of the dissertation are summarized as follows:(1)The research development situation of In SAR technique was summarized.The basic theory of D-In SAR(differential In SAR)and time series In SAR techniques were presented.The components of interferometric phase,the sources of decorrelation and the corresponding solutions were also analyzed.(2)A multi-sensor SAR images fusion method was proposed.This method geocoded the multi-sensor SAR images on the same scene to a common geographical coordinate,unifying the spatial resolution,coordinate of different SAR system.The observation values in different line of sight(LOS)were transformed into the vertical dimension by three dimensional decomposition method,providing a common geometry system for the results calculation,solving the problem different SAR images in different observation directions.A method of multi-sensor SAR images fusion was studied based on the SBAS technology.Considering the different operating bands of different SAR system,a multi spaceborne SAR images fusion method according to the quality of the interferograms was proposed.The methods enriching the measurements,extending observation time,was proved to be available,efficient,accurate by the experiments.(3)A method based on the probability integral prediction and In SAR technique monitoring the ground subsidence in mining area was proposed.According to the subsidence gradient of In SAR technique,an implementation plan about extracting the subsidence basin in mining area was studied.Setting the upper and lower critical threshold,the subsidence values bigger than the upper threshold,the results from probability integral method(PIM)were treated as the final values,the subsidence values smaller than the lower threshold,the results from In SAR were treated as the final values.The fusion weight values were designed according to the errors between In SAR,PIM and levelling using to obtaining the final values during the upper and lower threshold.These issues that decorrelation easily occurred in mining area and PIM inaccurate around the subsidence basin were resolved.(4)Ground deformation after coal mining was studied based on the spaceborne SAR method.The small deformation after mining activities were analyzed using Sentinel images.The TOPS SAR data were used to obtain the spatial distribution of the ground subsidence rate and time series results after mining activities.Combining the classification result of optical imagery,the relation between the distribution of land subsidence and land use was studied.The spatial distribution characteristics of surface deformation after mining were revealed.Land subsidence was monitored accurately for long time which was necessary for stability evaluation of surface,rational exploitation and use of land,safety of building in mining districts.(5)The MFM method(Multi-function Fusion Method)was proposed to monitor the land subsidence in mining area for long term.The linear function,step function,polynomial function,exponential function,trigonometric function,probability integral etc.were used in the non-linear ground deformation.Tikhonov was use to solve the ill-conditioned system to obtain the ground deformation with complex variation characteristics for long term.The study on Fengfeng mining area shows that this method overcomes the disadvantages of the traditional In SAR time series method on a degree.Combined with the multi-sensor SAR images fusion method,long time series land subsidence results over mining area could be obtained with several SAR datasets.The temporal and spatial land movement characterizations in mining area over a full mining period were revealed. |
