InSAR-based inversion method and system for spatial geometric parameters of underground excavation

Abstract

The present invention relates to an inversion method and system for geometric parameters of underground excavation space based on InSAR, and the method comprises: obtaining the surface deformation field of the study area where the underground excavation space is located based on the InSAR technology; collecting underground space positioning prior information according to the surface deformation field; collecting underground space geological prior information; fusing to obtain a prior information set of underground space geometric parameters; constructing a functional relationship between InSAR observed deformation and underground space-induced three-dimensional surface deformation through a probability integral method according to the prior information set; constructing a likelihood function based on InSAR observation data and surface deformation data; inverting underground space geometric parameters based on Bayesian theorem and likelihood function to obtain posterior distribution of underground space geometric parameters; solving the posterior distribution of underground space geometric parameters using a differential evolution adaptive Metropolis algorithm to obtain an inversion result. The accuracy and precision of the inversion result are improved.

Description

Technical Field

[0001] The present invention relates to the field of geographic monitoring, and in particular to an InSAR-based underground excavation spatial geometric parameter inversion method and system. Background Art

[0002] For a long time, underground excavation space detection has been mainly based on contact methods such as geophysical prospecting, and has been widely used in urban roads, subways, mines and karst cave detection. However, due to the influence of topography, detection efficiency and engineering cost factors, contact detection methods cannot carry out high-precision, high-resolution, all-weather detection work over a wide area.

[0003] To address the above problems, InSAR integrates synthetic aperture radar and interferometry technology to observe the surface deformation induced by underground excavation on a large scale through non-contact means, and to deduce the spatial geometric parameters of underground excavation (such as position and geometric shape) through parameter inversion methods, providing a new opportunity for regional three-dimensional precise monitoring and situation awareness.

[0004] Most of the existing underground space inversion methods use nonlinear search algorithms based on deterministic inversion to calculate relevant parameters based on InSAR deformation observations, without considering the uncertainty of the inversion results. However, due to factors such as inaccurate correlation models, significant spatial variability of rock and soil masses, and strong uncertainty in geological structures, the existing underground space inversion methods have key bottlenecks such as unstable inversion results, low accuracy, and contradictions with geological conditions, which greatly limit the practical application scope of the inversion results of underground excavation space geometric parameters. Summary of the invention

[0005] The present invention provides an InSAR-based underground excavation spatial geometric parameter inversion method and system, which improves the accuracy and precision of the inversion result.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] In a first aspect, a method for inverting spatial geometric parameters of underground excavation based on InSAR is provided, comprising:

[0008] Based on InSAR technology, the surface deformation field of the study area where the underground excavation space is located is obtained;

[0009] Collecting prior information on underground spatial positioning of underground excavation space based on the surface deformation field;

[0010] Collect regional geological survey data of the study area and obtain prior geological information of underground space based on the regional geological survey data;

[0011] The prior information of underground space positioning and the prior information of underground space geology are integrated to obtain the prior information set of underground space geometric parameters;

[0012] Based on the prior information set, the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is constructed by the probability integration method.

[0013] According to the InSAR observed deformation, InSAR observation data is obtained, surface deformation data is calculated according to the functional relationship and the InSAR observation data, and a likelihood function is constructed based on the InSAR observation data and the surface deformation data;

[0014] Based on Bayesian theorem and likelihood function, the geometric parameters of underground space are inverted to obtain the posterior distribution of the geometric parameters of underground space.

[0015] The differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of underground space geometric parameters and obtain the inversion results.

[0016] Furthermore, the underground space positioning prior information of the underground excavation space is collected according to the surface deformation field, including:

[0017] Identify surface subsidence basins induced by underground excavation space based on the surface deformation field;

[0018] The boundary function of the surface subsidence basin is fitted by using elliptical analytical equation to divide the surface subsidence area and the stable area.

[0019] Select a section within 10% of the length corresponding to the major axis of the ellipse analytical equation as the potential main section; select a section within 10% of the length corresponding to the minor axis of the ellipse analytical equation as the potential secondary section;

[0020] The settlement values ​​of the potential main section and the potential secondary section are extracted and fitted using the probability integral settlement curve respectively;

[0021] The inflection point of the probability integral settlement curve is taken as the underground space boundary of the underground excavation space;

[0022] The spatial distribution information of the inflection points is used as the length and width of the underground excavation space, and the center point of the rectangle obtained by connecting the inflection points of the probability integral settlement curve is used as the coordinate of the center point of the underground excavation space;

[0023] The length, width and center point coordinates of the underground excavation space are used as prior information for underground space positioning.

[0024] Furthermore, the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is expressed as:

[0025] ;

[0026] is the sight line deformation value, is the radar incident angle, is the azimuth of the radar line of sight, is the vertical deformation component of the three-dimensional deformation of the surface, is the northward deformation component of the three-dimensional surface deformation, is the eastward deformation component of the three-dimensional surface deformation;

[0027] , and The calculation formula is:

[0028] ;

[0029] It indicates the angle between the direction of underground excavation space and the clockwise direction to the east. It indicates the angle between the direction of underground excavation space and the clockwise direction of north. is the ground deformation value of the underground excavation space direction, is the ground deformation value in the inclination direction of the underground excavation space, is the excavation thickness, is the inclination angle, is the sedimentation coefficient.

[0030] Furthermore, the method also includes:

[0031] The quadtree downsampling method is used to downsample the surface deformation field and collect the surface deformation values.

[0032] Furthermore, the expression of the likelihood function is:

[0033] ;

[0034] is the surface deformation data, is the InSAR observation data, are independent Gaussian distributed residuals with constant variance.

[0035] Furthermore, based on the Bayesian theorem and the likelihood function, the geometric parameters of the underground space are inverted to obtain the posterior distribution of the geometric parameters of the underground space, including:

[0036] The inversion formula for inverting the geometric parameters of underground space based on Bayesian theorem and likelihood function is:

[0037] ;

[0038] is the posterior distribution of underground space geometric parameters, is the likelihood distribution based on InSAR observation data, is the parameter prior distribution, is the marginal likelihood function;

[0039] According to the inversion formula, the posterior distribution of the geometric parameters of the underground space is obtained .

[0040] In the second aspect, an InSAR-based underground excavation spatial geometric parameter inversion system is provided, including:

[0041] An acquisition module is used to obtain the surface deformation field of the study area where the underground excavation space is located based on InSAR technology;

[0042] A positioning prior information acquisition module is used to acquire underground space positioning prior information of underground excavation space according to the surface deformation field;

[0043] A geological prior information collection module is used to collect regional geological survey data of the study area and obtain underground space geological prior information based on the regional geological survey data;

[0044] A priori information fusion module is used to fuse the underground space positioning priori information and the underground space geological priori information to obtain a priori information set of underground space geometric parameters;

[0045] The functional relationship building module is used to build the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space through the probability integration method based on the prior information set;

[0046] A likelihood function construction module is used to obtain InSAR observation data according to InSAR observation deformation, calculate surface deformation data according to the functional relationship and InSAR observation data, and construct a likelihood function based on InSAR observation data and surface deformation data;

[0047] A posterior distribution calculation module is used to invert the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the geometric parameters of the underground space;

[0048] The inversion module is used to solve the posterior distribution of underground space geometric parameters using the differential evolution adaptive Metropolis algorithm to obtain the inversion results.

[0049] Furthermore, the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is expressed as:

[0050] ;

[0051] is the sight line deformation value, is the radar incident angle, is the azimuth of the radar line of sight, is the vertical deformation component of the three-dimensional deformation of the surface, is the northward deformation component of the three-dimensional surface deformation, is the eastward deformation component of the three-dimensional surface deformation;

[0052] , and The calculation formula is:

[0053] ;

[0054] It indicates the angle between the direction of underground excavation space and the clockwise direction to the east. It indicates the angle between the direction of underground excavation space and the clockwise direction of north. is the ground deformation value of the underground excavation space direction, is the ground deformation value in the inclination direction of the underground excavation space, is the excavation thickness, is the inclination angle, is the sedimentation coefficient.

[0055] Furthermore, the expression of the likelihood function is:

[0056] ;

[0057] is the surface deformation data, is the InSAR observation data, are independent Gaussian distributed residuals with constant variance.

[0058] Furthermore, the posterior distribution calculation module is specifically used to invert the formula of the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function:

[0059] ;

[0060] is the posterior distribution of underground space geometric parameters, is the likelihood distribution based on InSAR observation data, is the parameter prior distribution, is the marginal likelihood function; the posterior distribution of underground space geometric parameters is obtained according to the inversion formula .

[0061] The beneficial effects achieved by the present invention are:

[0062] Based on the InSAR technology, the surface deformation field of the study area where the underground excavation space is located is obtained; based on the surface deformation field, the underground space positioning prior information of the underground excavation space is collected; the regional geological survey data of the study area is collected, and the underground space geological prior information is obtained based on the regional geological survey data; the underground space positioning prior information and the underground space geological prior information are integrated to obtain the prior information set of underground space geometric parameters; based on the prior information set, the functional relationship between the InSAR observed deformation and the underground space induced three-dimensional surface deformation is constructed by the probability integral method; the InSAR observation data is obtained based on the InSAR observed deformation, the surface deformation data is calculated based on the functional relationship and the InSAR observation data, and the likelihood function is constructed based on the InSAR observation data and the surface deformation data; the underground space geometric parameters are inverted based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the underground space geometric parameters; the differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of the underground space geometric parameters to obtain the inversion results;

[0063] A new idea of ​​introducing geological prior information to constrain inversion parameters is proposed. Different from the existing inversion method that only uses InSAR observed deformation and analytical models to directly establish a simple mathematical relationship, the present invention can improve the accuracy of evaluation.

[0064] The differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of geometric parameters, which solves the problem that traditional methods cannot accurately solve multi-peak posterior distributions, improves the multi-solution problem of underground space positioning, and improves the calculation accuracy of the inversion algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a flow chart of the method for inverting the spatial geometric parameters of underground excavation based on InSAR of the present invention;

[0066] Figure 2 It is a structural diagram of the InSAR-based underground excavation spatial geometric parameter inversion system of the present invention. DETAILED DESCRIPTION

[0067] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0068] like Figure 1 As shown, an embodiment of the present invention provides an InSAR-based underground excavation space geometric parameter inversion method, comprising:

[0069] 101. Obtain the surface deformation field of the study area where the underground excavation space is located based on InSAR technology;

[0070] The InSAR technology, which integrates SAR and interferometry, is used to extract the deformation of the ground objects induced by the underground excavation space by complex conjugate multiplication of the SAR images that pass through the study area twice or more times, and the surface deformation field of the study area where the underground excavation space is located is obtained.

[0071] 102, collecting underground space positioning prior information of the underground excavation space according to the surface deformation field;

[0072] The specific calculation process is:

[0073] S1, identifying the surface subsidence basins induced by underground excavation space based on the surface deformation field;

[0074] S2, using elliptical analytical equation to fit the boundary function of the surface subsidence basin to divide the surface subsidence area and the stable area;

[0075] S3, select a section within 10% of the length corresponding to the major axis of the ellipse analytical equation as the potential main section; select a section within 10% of the length corresponding to the minor axis of the ellipse analytical equation as the potential secondary section;

[0076] S4, extract the settlement values ​​of the potential main section and the potential secondary section and fit them using the probability integral settlement curve respectively; the expression of the probability integral settlement curve is:

[0077] ;

[0078] in, , , and is the parameter to be determined in the fitting process, erf is the Gaussian error function, ;

[0079] S5, taking the inflection point of the probability integral settlement curve as the underground space boundary of the underground excavation space;

[0080] S6, using the spatial distribution information of the inflection points as the length and width of the underground excavation space, and connecting the inflection points of the probability integral settlement curve to obtain the center point of the rectangle as the coordinate of the center point of the underground excavation space;

[0081] S7, based on the length, width and center point coordinates of the underground excavation space, is used as prior information for underground space positioning.

[0082] 103. Collect regional geological survey data of the study area and obtain prior geological information of underground space based on the regional geological survey data;

[0083] The regional geological survey data includes the data preserved from previous survey work on the study area, which determines the potential excavation strata and their corresponding occurrences that may be distributed in the underground excavation space, outputs the geological prior knowledge of the burial depth, direction and inclination of the potential distribution of the underground space, and obtains the geological prior information of the underground space after sorting.

[0084] 104, integrating the underground space positioning prior information and the underground space geological prior information to obtain a prior information set of underground space geometric parameters;

[0085] 105. Based on the prior information set, the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is constructed by the probability integration method;

[0086] The functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is expressed as:

[0087] ;

[0088] is the sight line deformation value, is the radar incident angle, is the azimuth of the radar line of sight, is the vertical deformation component of the three-dimensional deformation of the surface, is the northward deformation component of the three-dimensional surface deformation, is the eastward deformation component of the three-dimensional surface deformation;

[0089] , and The calculation formula is:

[0090] ;

[0091] It indicates the angle between the direction of underground excavation space and the clockwise direction to the east. It indicates the angle between the direction of underground excavation space and the clockwise direction of north. is the ground deformation value of the underground excavation space direction, is the ground deformation value in the inclination direction of the underground excavation space, is the excavation thickness, is the inclination angle, is the sedimentation coefficient.

[0092] 106, obtaining InSAR observation data according to the InSAR observation deformation, calculating the surface deformation data according to the functional relationship and the InSAR observation data, and constructing a likelihood function based on the InSAR observation data and the surface deformation data;

[0093] The expression of the likelihood function is:

[0094] ;

[0095] is the surface deformation data, is the InSAR observation data, are independent Gaussian distributed residuals with constant variance.

[0096] It should be noted that the surface deformation field is downsampled using the quadtree downsampling method to collect the surface deformation values. Dense sampling points are set for areas with large deformation gradients, while sparse sampling is performed for areas with small deformation gradients to reduce the data dimension in parameter inversion and improve analysis efficiency.

[0097] 107. Based on Bayesian theorem and likelihood function, the geometric parameters of underground space are inverted to obtain the posterior distribution of the geometric parameters of underground space;

[0098] The inversion formula for inverting the geometric parameters of underground space based on Bayesian theorem and likelihood function is:

[0099] ;

[0100] is the posterior distribution of underground space geometric parameters, is the likelihood distribution based on InSAR observation data, is the parameter prior distribution, is the marginal likelihood function;

[0101] According to the inversion formula, the posterior distribution of the geometric parameters of the underground space is obtained .

[0102] 108. The differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of underground space geometric parameters and obtain the inversion results.

[0103] In view of the difficulty of integrating complex posterior distribution, the differential evolution adaptive Metropolis algorithm in Markov chain Monte Carlo method is used to solve the posterior distribution of underground space geometric parameters. Before solving the posterior distribution, the iterative calculation sample size needs to be determined.

[0104] The default posterior distribution calculation iteration number is used as the convergence criterion indicator of the calculation result. If the calculated result index is less than 1.2, it can be considered that the posterior distribution has converged and the result can be output as the posterior distribution of the geometric parameters of the underground space. If it is greater than 1.2, the number of iterations needs to be increased to avoid the contradiction between the calculated results and the posterior distribution law of the actual underground space geometric parameters.

[0105] The beneficial effects of the embodiments of the present invention are:

[0106] Based on the InSAR technology, the surface deformation field of the study area where the underground excavation space is located is obtained; based on the surface deformation field, the underground space positioning prior information of the underground excavation space is collected; the regional geological survey data of the study area is collected, and the underground space geological prior information is obtained based on the regional geological survey data; the underground space positioning prior information and the underground space geological prior information are integrated to obtain the prior information set of underground space geometric parameters; based on the prior information set, the functional relationship between the InSAR observed deformation and the underground space induced three-dimensional surface deformation is constructed by the probability integral method; the InSAR observation data is obtained based on the InSAR observed deformation, the surface deformation data is calculated based on the functional relationship and the InSAR observation data, and the likelihood function is constructed based on the InSAR observation data and the surface deformation data; the underground space geometric parameters are inverted based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the underground space geometric parameters; the differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of the underground space geometric parameters to obtain the inversion results;

[0107] A new idea of ​​introducing geological prior information to constrain inversion parameters is proposed. Different from the existing inversion method that only uses InSAR observed deformation and analytical models to directly establish a simple mathematical relationship, the present invention can improve the accuracy of evaluation.

[0108] The differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of geometric parameters, which solves the problem that traditional methods cannot accurately solve multi-peak posterior distributions, improves the multi-solution problem of underground space positioning, and improves the calculation accuracy of the inversion algorithm.

[0109] In combination with the InSAR-based underground excavation space geometric parameter inversion method described in the above embodiments, the InSAR-based underground excavation space geometric parameter inversion system is described below through embodiments.

[0110] like Figure 2 As shown, an embodiment of the present invention provides an InSAR-based underground excavation spatial geometric parameter inversion system, comprising:

[0111] An acquisition module 201 is used to acquire the surface deformation field of the study area where the underground excavation space is located based on the InSAR technology;

[0112] A positioning prior information collection module 202 is used to collect underground space positioning prior information of the underground excavation space according to the surface deformation field;

[0113] A geological prior information collection module 203 is used to collect regional geological survey data of the study area and obtain underground space geological prior information based on the regional geological survey data;

[0114] A priori information fusion module 204 is used to fuse the underground space positioning priori information and the underground space geological priori information to obtain a priori information set of underground space geometric parameters;

[0115] Functional relationship construction module 205, used to construct the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space through the probability integration method according to the prior information set;

[0116] A likelihood function construction module 206 is used to obtain InSAR observation data according to the InSAR observation deformation, calculate the surface deformation data according to the functional relationship and the InSAR observation data, and construct a likelihood function based on the InSAR observation data and the surface deformation data;

[0117] The posterior distribution calculation module 207 is used to invert the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the geometric parameters of the underground space;

[0118] The inversion module 208 is used to solve the posterior distribution of the geometric parameters of the underground space using the differential evolution adaptive Metropolis algorithm to obtain an inversion result.

[0119] Preferably, in some embodiments of the present invention, the functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is expressed as:

[0120] ;

[0121] is the sight line deformation value, is the radar incident angle, is the azimuth of the radar line of sight, is the vertical deformation component of the three-dimensional deformation of the surface, is the northward deformation component of the three-dimensional surface deformation, is the eastward deformation component of the three-dimensional surface deformation;

[0122] , and The calculation formula is:

[0123] ;

[0124] It indicates the angle between the direction of underground excavation space and the clockwise direction to the east. It indicates the angle between the direction of underground excavation space and the clockwise direction of north. is the ground deformation value of the underground excavation space direction, is the ground deformation value in the inclination direction of the underground excavation space, is the excavation thickness, is the inclination angle, is the sedimentation coefficient.

[0125] Preferably, in some embodiments of the present invention, the expression of the likelihood function is:

[0126] ;

[0127] is the surface deformation data, is the InSAR observation data, are independent Gaussian distributed residuals with constant variance.

[0128] Preferably, in some embodiments of the present invention,

[0129] The posterior distribution calculation module 207 is specifically used to invert the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function. The formula is:

[0130] ;

[0131] is the posterior distribution of underground space geometric parameters, is the likelihood distribution based on InSAR observation data, is the parameter prior distribution, is the marginal likelihood function; the posterior distribution of underground space geometric parameters is obtained according to the inversion formula .

[0132] In summary, the beneficial effects of the InSAR-based underground excavation spatial geometric parameter inversion system are:

[0133] A new idea of ​​introducing geological prior information to constrain inversion parameters is proposed. Different from the existing inversion method that only uses InSAR observed deformation and analytical models to directly establish a simple mathematical relationship, the present invention can improve the accuracy of evaluation.

[0134] The differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of geometric parameters, which solves the problem that traditional methods cannot accurately solve multi-peak posterior distributions, improves the multi-solution problem of underground space positioning, and improves the calculation accuracy of the inversion algorithm.

[0135] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A method for inversion of spatial geometric parameters of underground excavation based on InSAR, characterized in that: include: Based on InSAR technology, the surface deformation field of the study area where the underground excavation space is located is obtained; collecting underground space positioning prior information of the underground excavation space according to the surface deformation field; Collecting regional geological survey data of the study area, and obtaining underground space geological prior information based on the regional geological survey data; The underground space positioning prior information and the underground space geological prior information are integrated to obtain a prior information set of underground space geometric parameters; According to the prior information set, a functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is constructed by a probability integral method; Obtaining InSAR observation data according to the InSAR observed deformation, calculating surface deformation data according to the functional relationship and the InSAR observation data, and constructing a likelihood function based on the InSAR observation data and the surface deformation data; Inverting the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the geometric parameters of the underground space; The differential evolution adaptive Metropolis algorithm is used to solve the posterior distribution of the geometric parameters of the underground space to obtain an inversion result.

2. The InSAR-based underground excavation spatial geometric parameter inversion method according to claim 1, characterized in that: The collecting of underground space positioning prior information of the underground excavation space according to the surface deformation field includes: identifying a surface subsidence basin induced by the underground excavation space according to the surface deformation field; Using an elliptic analytical equation to fit the boundary function of the surface subsidence basin to divide the surface subsidence area and the stable area; A cross section within 10% of the length corresponding to the major axis of the ellipse analytical equation is selected as a potential main cross section; a cross section within 10% of the length corresponding to the minor axis of the ellipse analytical equation is selected as a potential secondary cross section; Extracting the settlement values ​​of the potential main section and the potential secondary section and fitting them respectively using probability integral settlement curves; taking the inflection point of the probability integral settlement curve as the underground space boundary of the underground excavation space; The spatial distribution information of the inflection points is used as the length and width of the underground excavation space, and the center point of the rectangle obtained by connecting the inflection points of the probability integral settlement curve is used as the coordinate of the center point of the underground excavation space; The length, width and center point coordinates of the underground excavation space are used as prior information for underground space positioning.

3. The InSAR-based underground excavation spatial geometric parameter inversion method according to claim 1, characterized in that: The functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is expressed as: ； Said is the line-of-sight deformation value, is the radar incident angle, is the azimuth of the radar line of sight, is the vertical deformation component of the three-dimensional deformation of the surface. is the northward deformation component of the three-dimensional surface deformation, is the eastward deformation component of the three-dimensional surface deformation; Said , and The calculation formula is: ； Said represents the angle between the direction of the underground excavation space and the clockwise direction of the east. represents the angle between the direction of the underground excavation space and the clockwise direction of the north. is the ground deformation value of the underground excavation space direction, is the ground deformation value in the inclination direction of the underground excavation space, is the excavation thickness, is the inclination angle, is the sedimentation coefficient.

4. The InSAR-based underground excavation spatial geometric parameter inversion method according to claim 3, characterized in that: The method further comprises: The surface deformation field is downsampled by using a quadtree downsampling method to collect the surface deformation value.

5. The InSAR-based underground excavation spatial geometric parameter inversion method according to claim 1, characterized in that: The expression of the likelihood function is: ； Said is the surface deformation data, is the InSAR observation data, are independent Gaussian distributed residuals with constant variance.

6. The InSAR-based underground excavation spatial geometric parameter inversion method according to claim 5, characterized in that: The inversion of the underground space geometric parameters based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the underground space geometric parameters includes: The inversion formula for inverting the geometric parameters of underground space based on Bayesian theorem and the likelihood function is: ； Said is the posterior distribution of underground space geometric parameters, is the likelihood distribution based on InSAR observation data, is the parameter prior distribution, is the marginal likelihood function; According to the inversion formula, the posterior distribution of the geometric parameters of the underground space is obtained. .

7. An InSAR-based underground excavation spatial geometric parameter inversion system, characterized in that: include: An acquisition module is used to obtain the surface deformation field of the study area where the underground excavation space is located based on InSAR technology; A positioning priori information acquisition module, used to acquire underground space positioning priori information of the underground excavation space according to the surface deformation field; A geological prior information collection module is used to collect regional geological survey data of the study area and obtain underground space geological prior information based on the regional geological survey data; A priori information fusion module, used to fuse the underground space positioning priori information and the underground space geological priori information to obtain a priori information set of underground space geometric parameters; A functional relationship building module is used to build a functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space through a probability integration method according to the prior information set; A likelihood function construction module is used to obtain InSAR observation data according to the InSAR observation deformation, calculate the surface deformation data according to the functional relationship and the InSAR observation data, and construct a likelihood function based on the InSAR observation data and the surface deformation data; A posterior distribution calculation module is used to invert the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function to obtain the posterior distribution of the geometric parameters of the underground space; The inversion module is used to solve the posterior distribution of the geometric parameters of the underground space by using the differential evolution adaptive Metropolis algorithm to obtain an inversion result.

8. The InSAR-based underground excavation spatial geometric parameter inversion system according to claim 7, characterized in that: The functional relationship between the InSAR observed deformation and the three-dimensional surface deformation induced by the underground space is expressed as: ； Said is the line-of-sight deformation value, is the radar incident angle, is the azimuth of the radar line of sight, is the vertical deformation component of the three-dimensional deformation of the surface. is the northward deformation component of the three-dimensional surface deformation, is the eastward deformation component of the three-dimensional surface deformation; Said , and The calculation formula is: ； Said represents the angle between the direction of the underground excavation space and the clockwise direction to the east. represents the angle between the direction of the underground excavation space and the clockwise direction of the north. is the ground deformation value of the underground excavation space direction, is the ground deformation value in the inclination direction of the underground excavation space, is the excavation thickness, is the inclination angle, is the sedimentation coefficient.

9. The InSAR-based underground excavation spatial geometric parameter inversion system according to claim 7, characterized in that: The expression of the likelihood function is: ； Said is the surface deformation data, is the InSAR observation data, are independent Gaussian distributed residuals with constant variance.

10. The InSAR-based underground excavation spatial geometric parameter inversion system according to claim 9, characterized in that: The posterior distribution calculation module is specifically used to invert the formula of the geometric parameters of the underground space based on the Bayesian theorem and the likelihood function: ； Said is the posterior distribution of underground space geometric parameters, is the likelihood distribution based on InSAR observation data, is the parameter prior distribution, is the marginal likelihood function; the posterior distribution of underground space geometric parameters is obtained according to the inversion formula .

Images