CN-122015769-A - Underground mining settlement monitoring method and system based on InSAR technology

CN122015769ACN 122015769 ACN122015769 ACN 122015769ACN-122015769-A

Abstract

The invention relates to the technical field of surveying and mapping and remote sensing, and discloses an underground mining subsidence monitoring method and system based on an InSAR technology. The method comprises the steps of obtaining multi-time-phase SAR images, GNSS observation data, DEM and exploitation logs, generating a differential interferogram sequence, utilizing GNSS to build an atmospheric phase screen model for correction, building a theoretical sedimentation prior model based on exploitation parameters, building the theoretical sedimentation prior model as constraint and embedding an improved minimum cost flow phase unwrapping algorithm for the exploitation parameters, carrying out space-time filtering and time sequence stacking on unwrapping phases, and combining singular value decomposition with self-adaptive deformation model fitting to generate a ground surface three-dimensional deformation field time sequence. The system comprises modules of multi-source data acquisition, differential interferogram generation, atmosphere correction, prior modeling, constraint unwrapping, deformation reconstruction and the like. The invention realizes dynamic monitoring of mining subsidence with millimeter precision, high frequency and large range through multisource fusion and physical mechanism constraint.

Inventors

LI KAIXUAN
ZHANG YANFEI
WANG JIN
DUAN PINZHANG
LIU HUALIN
BAI YAWEN
LIANG WENTAO
LIU TIEJUN
JI GANG
PENG KAI
FENG YARU
HAN ZHENHUA
JIAO RUI
LI ZIJING

Assignees

水利部牧区水利科学研究所

Dates

Publication Date: 20260512
Application Date: 20260305

Claims (10)

1. An InSAR technology-based underground mining settlement monitoring method is characterized by comprising the following steps: acquiring a multi-time phase synthesis aperture radar image dataset covering a target mining area; synchronously acquiring global navigation satellite system observation data, digital elevation model data and mining area exploitation progress logs which are matched with the multi-time phase synthetic aperture radar image data set in time; performing accurate registration and baseline estimation on the multi-time-phase synthetic aperture radar image dataset to generate a differential interferogram sequence; constructing an atmospheric phase screen model by using the global navigation satellite system observation data, and carrying out atmospheric delay phase correction on the differential interferogram sequence; Based on the digital elevation model data and mining area mining progress logs, establishing a theoretical sedimentation priori model caused by mining; taking the theoretical sedimentation prior model as an external constraint condition, embedding the theoretical sedimentation prior model into an improved minimum cost flow phase unwrapping algorithm, and generating an unwrapped interference phase sequence; And performing space-time filtering and time sequence stacking processing on the unwrapped interference phase sequence, extracting a main deformation mode by adopting a singular value decomposition method, and performing parameter fitting by combining a linear deformation rate model and a nonlinear deformation rate model to finally generate a surface three-dimensional deformation field time sequence product.
2. The method for monitoring subsidence of an underground mining based on the InSAR technology according to claim 1, wherein synchronously acquiring the global navigation satellite system observation data time-matched with the multi-time-phase synthetic aperture radar image dataset comprises: And (3) arranging at least three permanent global navigation satellite system monitoring stations in the mining area, recording carrier phases and pseudo-range observation values in two hours before and after each synthetic aperture radar image imaging moment, and calculating three-dimensional coordinates and covariance matrixes of the monitoring stations at the imaging moment through a precise single-point positioning algorithm.
3. The InSAR technology-based underground mining subsidence monitoring method of claim 2, wherein constructing an atmospheric phase screen model using the global navigation satellite system observation data comprises: Converting zenith total delay amount obtained by solving a global navigation satellite system into a large air path delay phase along the radar sight direction; Adopting a Kriging interpolation method, and taking the position of a monitoring station of a global navigation satellite system as a control point to generate an atmosphere position two-dimensional space distribution field covering the whole differential interferogram area; And subtracting the two-dimensional spatial distribution field of the atmosphere phase from the original differential interference diagram pixel by pixel to finish the atmosphere phase correction.
4. The method for monitoring sedimentation in underground mining based on InSAR technology as set forth in claim 3, wherein establishing a theoretical sedimentation prior model caused by mining comprises: Calculating the expected sinking value, the inclination value and the curvature value of any point on the earth surface by adopting a probability integration method according to the working face advancing position, the mining thickness, the mining depth and the rock stratum movement angle parameters in the mining progress log of the mining area; and (3) rasterizing the calculation result into a sedimentation rate priori graph with the same resolution as the synthetic aperture radar image, and endowing the sedimentation rate priori graph with a space weight coefficient which decays exponentially along with the distance from the center of the working surface.
5. The method for monitoring underground mining subsidence based on InSAR technology as set forth in claim 4, wherein embedding the theoretical subsidence prior model as an external constraint into an improved minimum cost flow phase unwrapping algorithm comprises: when constructing a phase gradient network diagram, taking a phase gradient corresponding to a theoretical sedimentation priori diagram as an initial flow reference value of the edge; a priori constraint penalty term is added to the minimum cost flow optimization objective function, and the weight of the priori constraint penalty term is dynamically adjusted by the local coherence coefficient.
6. The InSAR technology-based underground mining subsidence monitoring method of claim 5, wherein performing a spatio-temporal filtering and timing stacking process on the unwrapped interferometric phase sequence comprises: Carrying out spatial domain filtering on each scene unwrapping phase map by adopting a wavelet threshold denoising method; in the time dimension, sliding window median filtering is applied to the phase sequence of the same geographic coordinate, and the window length is 5-scene images; all the filtered phase maps are then stacked in time order to form a three-dimensional phase cube.
7. The method for monitoring sedimentation in underground mining based on InSAR technology as set forth in claim 6, wherein the steps of extracting the main deformation mode by singular value decomposition and performing parameter fitting in combination with linear and nonlinear deformation rate models include: singular value decomposition is carried out on the three-dimensional phase cube, and the first plurality of main components with the accumulated contribution rate exceeding 95% are reserved; performing least square fitting on the principal component time sequence and a linear function, a quadratic polynomial function and an exponential decay function respectively; According to the root mean square minimum principle of the fitting residual error, automatically selecting the optimal deformation model type; Finally inverting the parameters of the selected model into deformation rate fields of the earth surface in the east, north and vertical directions.
8. The method for monitoring subsidence of underground mining based on InSAR technology as set forth in claim 7, wherein in the process of generating the differential interferogram sequence, after conjugate multiplication of each pair of interference combinations, the digital elevation model data is utilized to simulate and remove the topographic phase composition, so as to obtain a differential interferogram only comprising deformation, atmosphere and noise information.
9. The method for monitoring the subsidence of an underground mining based on the InSAR technology according to claim 8, wherein the spatial weight coefficient of the subsidence rate prior map is defined as an exponential decay function taking the center of the current working surface as an origin, and is used for representing the spatial distribution characteristic of the prediction reliability of the theoretical model.
10. An InSAR technology-based underground mining subsidence monitoring system, characterized in that the underground mining subsidence monitoring is realized using an InSAR technology-based underground mining subsidence monitoring method as claimed in any one of claims 1 to 9, the system comprising: the multi-source remote sensing data acquisition unit is used for acquiring a multi-time phase synthetic aperture radar image dataset covering a target mining area, global navigation satellite system observation data, digital elevation model data and mining area exploitation progress logs; The differential interferogram generating unit is used for carrying out accurate registration and baseline estimation on the multi-time-phase synthetic aperture radar image dataset to generate a differential interferogram sequence; the atmospheric phase correction unit is used for constructing an atmospheric phase screen model by utilizing the global navigation satellite system observation data and carrying out atmospheric delay phase correction on the differential interferogram sequence; the theoretical sedimentation priori modeling unit is used for establishing a theoretical sedimentation priori model caused by mining based on the digital elevation model data and mining area mining progress logs; The constraint phase unwrapping unit is used for taking the theoretical sedimentation prior model as an external constraint condition, embedding the theoretical sedimentation prior model into an improved minimum cost flow phase unwrapping algorithm, and generating an unwrapped interference phase sequence; And the deformation time sequence reconstruction unit is used for performing space-time filtering and time sequence stacking processing on the unwrapped interference phase sequence, extracting a main deformation mode by adopting a singular value decomposition method, and performing parameter fitting by combining a linear deformation rate model and a nonlinear deformation rate model to finally generate a surface three-dimensional deformation field time sequence product.

Description

Underground mining settlement monitoring method and system based on InSAR technology Technical Field The invention belongs to the technical field of surveying and mapping and remote sensing, and particularly relates to an underground mining subsidence monitoring method and system based on an InSAR technology. Background With the acceleration of urbanization and resource development, surface subsidence caused by underground mining activities has become an important geological disaster threatening ecological safety and infrastructure stability of mining areas. Synthetic aperture radar interferometry (InSAR) technology has been widely used in the field of surface deformation monitoring by virtue of its advantages of large range, high precision and non-contact, and can acquire millimeter-level surface subsidence time sequence information. InSAR can only reflect macroscopic results of surface deformation, cannot reveal a hydrogeological driving mechanism behind subsidence, is difficult to distinguish subsidence response caused by different causes such as goaf collapse, aquifer drainage or construction activities, and limits deep application in risk early warning and cause tracing. Dynamic changes in groundwater level serve as an important factor in inducing or exacerbating ground subsidence. The effective stress of the aquifer is increased due to drainage, so that soil layer compression and surface subsidence are induced. The traditional monitoring means are mostly dependent on the arrangement of sparse underground water level observation wells, and although local water level data can be provided, the spatial coverage is limited, the updating frequency is low, the spatial resolution is not matched with the InSAR deformation field, and the dynamic coupling relation between the two is difficult to establish. In the prior art, inSAR deformation data and hydrological data are simply overlapped or subjected to post-hoc correlation analysis, and depth fusion of the InSAR deformation data and the hydrological data on time synchronism, spatial correspondence and physical mechanism consistency is lacking, so that a settlement-water level response model with prediction capability cannot be constructed. In the prior art, no method has been available for realizing the accurate space-time registration of InSAR earth surface deformation sequences and automatic high-frequency ground water level monitoring data, quantifying the contribution degree of aquifer drainage to ground settlement on the basis of the accurate space-time registration, and setting a dynamic threshold value according to real-time water level change to trigger settlement risk early warning. The technical blank causes that a mining area manager is difficult to identify a high-risk area before settlement occurs, and a differential prevention and control strategy cannot be implemented aiming at different hydrogeological conditions, so that the initiative and timeliness of ecological restoration and infrastructure protection are restricted. What is needed is an underground mining subsidence monitoring method and system that incorporates multi-source heterogeneous monitoring data and has mechanism interpretation capability and early warning function. Disclosure of Invention The invention provides an underground mining subsidence monitoring method and system based on a synthetic aperture radar interferometry technology, and aims to solve the problems of insufficient subsidence inversion precision, long monitoring period, low spatial resolution, weak dynamic response capability and the like caused by atmospheric delay errors, space-time coherence effects and a single data source in the prior art. According to the invention, by constructing a multisource heterogeneous remote sensing data fusion frame, introducing an external physical constraint model and designing a self-adaptive phase unwrapping and deformation time sequence reconstruction algorithm, high-precision, high-frequency and large-range continuous monitoring of the mining area surface millimeter-level sedimentation field is realized. The invention provides an InSAR technology-based underground mining settlement monitoring method, which comprises the following steps: acquiring a multi-time phase synthesis aperture radar image dataset covering a target mining area; synchronously acquiring global navigation satellite system observation data, digital elevation model data and mining area exploitation progress logs which are matched with the multi-time phase synthetic aperture radar image data set in time; performing accurate registration and baseline estimation on the multi-time-phase synthetic aperture radar image dataset to generate a differential interferogram sequence; constructing an atmospheric phase screen model by using the global navigation satellite system observation data, and carrying out atmospheric delay phase correction on the differential interferogram sequence; Based on the digital elevation m