Nonlinear crustal stress field inversion method, device, equipment and medium

Abstract

The invention discloses a nonlinear crustal stress field inversion method, device, equipment and medium, relates to the technical field of geotechnical engineering, and aims to at least solve the technical problem of large deviation between an inversion result and an actual result. The method comprises the following steps: constructing a three-dimensional numerical model, and determining a plurality of stress component measured values; establishing a plurality of rough fluctuating surfaces corresponding to each model boundary surface in the three-dimensional numerical model, and generating a plurality of groups of nonlinear variation coefficient matrixes corresponding to each model boundary surface; adjusting the plurality of linear boundary conditions based on the plurality of groups of nonlinear variation coefficient matrixes to generate a plurality of nonlinear boundary conditions; applying a plurality of nonlinear boundary conditions on the three-dimensional numerical model to obtain a stress component calculation value of each preset measurement point; and performing regression analysis on the stress component calculated value and the stress component measured value to obtain a regression coefficient of each nonlinear boundary condition, weighting by taking the regression coefficient as a weight, and substituting into the three-dimensional numerical model for calculation to obtain a crustal stress field of the three-dimensional numerical model.

Description

Technical Field

[0001] The present invention relates to the field of geotechnical engineering technology, and in particular to a nonlinear geostress field inversion method, device, equipment and medium. Background Art

[0002] Due to the high cost of drilling holes for in-situ geostress measurement and the high site requirements, it is impossible to carry out large-scale initial geostress field measurement work for geotechnical engineering. Therefore, inverting the initial geostress field within the entire geotechnical engineering range through limited drilling measurement data is a common method for investigating the distribution law of geostress field in the engineering design stage.

[0003] The core assumption of the stress field inversion technology used in related technologies is usually to apply linear boundary conditions to the numerical model, and these boundary conditions are mostly unilaterally or bidirectionally symmetrical. However, this simplified treatment ignores the inherent heterogeneity of the geostress field and fails to fully capture and accurately reproduce the complex nonlinear characteristics of the geostress distribution at the engineering site. Therefore, in practical applications, the inversion results of these inversion methods often deviate greatly from the actual results, which leads to poor accuracy of geostress inversion of three-dimensional numerical models. Summary of the invention

[0004] The present invention provides a nonlinear geostress field inversion method, device, computer equipment and medium to at least solve the technical problem in the related art that when linear boundary conditions are applied to numerical models, the geostress field inversion results often have a large deviation from the actual situation. The technical solution of the present invention is as follows: According to a first aspect of an embodiment of the present invention, a nonlinear geostress field inversion method is provided, the method comprising: constructing a three-dimensional numerical model based on engineering geological data, geostress measured data and stratum lithology data of a region to be inverted, and determining measured values ​​of multiple stress components at multiple preset measurement points; establishing multiple rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model, and generating multiple groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the multiple rough undulating surfaces; determining multiple linear boundary conditions based on the geostress measured data of the region to be inverted, and adjusting the multiple linear boundary conditions based on the multiple groups of nonlinear variation coefficient matrices to generate multiple nonlinear boundary conditions; applying multiple nonlinear boundary conditions to the three-dimensional numerical model, and obtaining a calculated value of a stress component at each preset measurement point through numerical simulation; performing regression analysis on the calculated values ​​of multiple stress components and the measured values ​​of multiple stress components to obtain a regression coefficient of each nonlinear boundary condition, weighting them with the regression coefficient as a weight, and substituting them into the three-dimensional numerical model for calculation to obtain a geostress field of the three-dimensional numerical model.

[0005] According to a second aspect of an embodiment of the present invention, a nonlinear geostress field inversion device is provided, which is used to implement the nonlinear geostress field inversion method of the first aspect; the device includes: a first construction module, which is used to construct a three-dimensional numerical model based on engineering geological data, geostress measured data and stratum lithology data of the area to be inverted; a first determination module, which is used to determine the measured values ​​of multiple stress components at multiple preset measurement points; a second construction module, which is used to establish multiple rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model; a first generation module, which is used to generate multiple groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the multiple rough undulating surfaces; and a second determination module , which is used to determine multiple linear boundary conditions based on the measured geostress data of the area to be inverted; the second generation module is used to adjust the multiple linear boundary conditions based on multiple groups of nonlinear variation coefficient matrices to generate multiple nonlinear boundary conditions; the third generation module is used to apply multiple nonlinear boundary conditions on the three-dimensional numerical model, and obtain the calculated value of the stress component of each preset measuring point through numerical simulation; the fourth generation module is used to perform regression analysis on the calculated values ​​of multiple stress components and the measured values ​​of multiple stress components to obtain the regression coefficient of each nonlinear boundary condition, weight them with the regression coefficient as the weight, and substitute them into the three-dimensional numerical model for calculation to obtain the geostress field of the three-dimensional numerical model.

[0006] According to a third aspect of an embodiment of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the nonlinear geostress field inversion method of the first aspect can be implemented.

[0007] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the nonlinear geostress field inversion method of the first aspect can be implemented.

[0008] The technical solution provided by the embodiment of the present invention brings at least the following beneficial effects: a three-dimensional numerical model is established based on the area to be inverted, a nonlinear variation coefficient generation method is used to generate a nonlinear variation coefficient matrix that matches the boundary size of the three-dimensional numerical model, and the nonlinear variation coefficient matrix is ​​used as a key cornerstone to construct nonlinear boundary conditions in combination with multiple small grids divided by the model boundary. Thereafter, nonlinear boundary conditions are applied to the three-dimensional numerical model, and finally the stress field is inverted according to the regression coefficient. In the above manner, the limitations of the linear inversion technology used in the related art in expressing the initial non-uniform ground stress field are successfully overcome, and the accuracy of the ground stress inversion of the three-dimensional numerical model can be effectively improved.

[0009] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the spirit of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings: Figure 1 is a flow chart of a nonlinear geostress field inversion method according to an exemplary embodiment; Figure 2 is a schematic flow chart of a nonlinear geostress field inversion method according to an exemplary embodiment; Figure 3 is a schematic diagram of a process of creating a symmetrical displacement boundary in a two-dimensional case according to an exemplary embodiment; Figure 4 is a schematic diagram of a process of creating a symmetrical stress boundary in a two-dimensional case according to an exemplary embodiment; Figure 5 is a schematic diagram of a process of creating an asymmetric displacement boundary in a two-dimensional case according to an exemplary embodiment; Figure 6 is a schematic diagram of a process of creating an asymmetric stress boundary in a two-dimensional case according to an exemplary embodiment; Figure 7 is a schematic diagram of a process of creating a mixed boundary in a two-dimensional case according to an exemplary embodiment; Figure 8 is a schematic structural diagram of a three-dimensional numerical model according to an exemplary embodiment; Fig. 9 is one of the schematic diagrams of a process for generating a rough undulating surface by using a bubbling method according to an exemplary embodiment; Fig.10 is a second schematic diagram of a process for generating a rough undulating surface by using a bubbling method according to an exemplary embodiment; Fig.11 is a third schematic diagram of a process for generating a rough undulating surface by using a bubbling method according to an exemplary embodiment; Fig.12 is a maximum principal stress cloud diagram of a three-dimensional numerical model geostress field based on symmetric displacement boundary condition inversion according to an exemplary embodiment; Fig.13 is a minimum principal stress cloud diagram of a three-dimensional numerical model ground stress field based on symmetric displacement boundary condition inversion according to an exemplary embodiment; Fig.14 is a schematic diagram of a process of generating a rough undulating surface using the WM equation method according to an exemplary embodiment; Fig.15 is a maximum principal stress cloud diagram of a three-dimensional numerical model geostress field based on inversion of asymmetric stress boundary conditions according to an exemplary embodiment; Fig.16 is a minimum principal stress cloud diagram of a three-dimensional numerical model ground stress field based on inversion of asymmetric stress boundary conditions according to an exemplary embodiment; Fig.17 It is a schematic structural diagram of a nonlinear geostress field inversion device according to an exemplary embodiment. DETAILED DESCRIPTION

[0011] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0012] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0013] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0014] The embodiment of the present invention provides a nonlinear geostress field inversion method, such as Figure 1 As shown, the nonlinear geostress field inversion method of this embodiment includes: S10: constructing a three-dimensional numerical model based on the engineering geological data, measured ground stress data and stratum lithology data of the area to be inverted, and determining measured values ​​of multiple stress components at multiple preset measurement points; In this step, according to the geological report, the area including all the project impact areas is selected as the area to be inverted. Then, based on the engineering geological data and stratigraphic lithology data of the area to be inverted, the relative position relationship between the terrane, rock formation, fracture geological structure and the numerical model boundary is quantified, the distribution position is determined, and a three-dimensional numerical model reflecting the geological structure information of the area to be inverted is established using modeling software.

[0015] Afterwards, the measured values ​​of geostress are collected through the on-site geostress measuring points (i.e., the preset measuring points), and the measured values ​​of geostress are converted into stress components to obtain multiple geostress components in the geodetic coordinate system. The multiple geostress components are converted into multiple stress components in the calculation coordinates through the coordinate conversion formula, which are the measured values ​​of the stress components.

[0016] In one embodiment of the present application, a specific scheme for constructing a three-dimensional numerical model is provided. In S10, based on the engineering geological data, measured geostress data and stratum lithology data of the area to be inverted, a three-dimensional numerical model is constructed for geotechnical engineering, and a scheme for determining measured values ​​of multiple stress components at multiple preset measurement points is specifically provided, which includes the following steps S11-S13: S11: Obtain engineering geological data, stratum lithology data and measured ground stress data of multiple preset measurement points in the area to be inverted; S12: Establish a three-dimensional numerical model based on engineering geological data and stratum lithology data; S13: coordinate transformation is performed on the principal stress values ​​in the measured ground stress data to obtain measured values ​​of multiple stress components in the calculation coordinate system.

[0017] For steps S11-S13, according to the engineering geological data of the area to be inverted, the in-situ stress measuring points (i.e., preset measuring points) are pre-set to obtain the measured values ​​of in-situ stress. According to the engineering geological data and the stratum lithology data, a regional three-dimensional numerical model is constructed using modeling software. The measured in-situ stress values ​​obtained at each preset measuring point include three principal stress values. The elastic mechanics formula is used to convert the measured principal stress values ​​of the area to be inverted into the measured values ​​of the six stress components in the calculation coordinate system corresponding to the numerical calculation model.

[0018] S20: establishing a plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model, and generating a plurality of groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the plurality of rough undulating surfaces.

[0019] In this step, the geological interface is usually complex and irregular, and the terrain undulation has a significant impact on the distribution of the geostress field. In order to reduce the difference between the model and the actual geological conditions, a rough undulating surface is constructed for each model boundary surface (i.e., the side of the model) to simulate the geological interface more realistically, thereby improving the accuracy of the model. Subsequently, based on the three-dimensional coordinate data of each rough undulating surface, multiple sets of nonlinear variation coefficient matrices are generated for each model boundary surface.

[0020] S21: using a preset undulating surface generation algorithm to generate a plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model according to a preset number; In this step, for any model boundary surface of the three-dimensional numerical model, a preset undulating surface generation algorithm is used to generate multiple rough surfaces corresponding to the model boundary surface. Thereafter, the three-dimensional coordinate data of each rough undulating surface in the calculation coordinate system is transformed to obtain a set of nonlinear variation coefficient matrices corresponding to each rough undulating surface. Finally, the nonlinear variation coefficient matrices of multiple rough undulating surfaces are summarized to obtain multiple sets of nonlinear variation coefficient matrices corresponding to the model boundary surface.

[0021] Optionally, the preset undulating surface generation algorithm includes but is not limited to: bubbling method, WM equation method, waveform superposition, rock structure surface scanning data, terrain undulation data, etc.

[0022] Furthermore, in order to better simulate the nonlinear characteristics of the geostress field, the preset number can be set to a range of 5 to 15, such as 5, 10, 15, etc. For example, 10 rough undulating surfaces are generated for each model boundary surface, and then 10 sets of nonlinear variation coefficient matrices are generated for each model boundary surface.

[0023] S22: for any rough and undulating surface, dividing the rough and undulating surface into a plurality of grids based on a preset division size; S23: Obtain three-dimensional coordinate data of each grid; S24: performing coordinate transformation on the three-dimensional coordinate data to generate a set of nonlinear variation coefficient matrices of the model boundary surface.

[0024] For steps S22-S24, each rough undulating surface has X, Y, and Z three-dimensional coordinate data in the calculation coordinate system. The mean value of the rough undulating surface is approximately 0, and its original form cannot be directly used. It must be converted into a coefficient matrix that fluctuates around 1 and uniformly discretized. Therefore, based on the differential idea, each rough undulating surface is discretized into multiple grids, and the Z value of each grid in the three-dimensional coordinate data in the calculation coordinate system is adjusted to generate a set of nonlinear change coefficient matrices corresponding to each model boundary surface.

[0025] Optionally, the size of the nonlinear variation coefficient matrix should match the size of the model boundary, therefore. Based on the size of the model lateral boundary, determine the preset division size of the grid. For example, if the size of the model lateral boundary is 1000m×1000m, the preset division size of each grid can be 20m×20m, and the lateral boundary of the three-dimensional numerical model is divided into 50×50=2500 grids. Accordingly, in order to match the layout of these 2500 small grids and effectively adjust the stress (displacement) on these grids, the size of the nonlinear variation coefficient matrix to be generated should also be set to 50×50. This matrix will be able to reflect the nonlinear changes in the lateral stress (displacement) of each small grid in detail, thereby ensuring the accuracy and reliability of the three-dimensional numerical model.

[0026] In the above way, a rough undulating surface is constructed by presetting the undulating surface generation algorithm to reflect the actual nonlinear behavior of the rock, and then a nonlinear variation coefficient matrix is ​​generated as the basis for creating nonlinear boundary conditions in the numerical model.

[0027] In one embodiment of the present application, a specific scheme for generating a nonlinear change coefficient matrix is ​​provided. In S24, coordinate transformation is performed on the three-dimensional coordinate data to generate a set of nonlinear change coefficient matrix schemes for the model boundary surface, which specifically includes the following steps S241-S243: S241: Determine first elevation coordinate data of each grid based on the three-dimensional coordinate data; S242: using a preset elevation coordinate conversion formula to convert the first elevation coordinate data of each grid to obtain converted second elevation coordinate data; S243: Based on the multiple horizontal coordinate data and the multiple second elevation coordinate data of the multiple grids, a set of nonlinear variation coefficient matrices corresponding to the model boundary surface is generated.

[0028] For steps S241-S243, each rough undulating surface has X, Y, and Z three-dimensional coordinate data, wherein the X coordinate and the Y coordinate are horizontal coordinates, and the Z coordinate is an elevation coordinate. For any rough undulating surface, the first elevation coordinate data (i.e., the Z coordinate) of each grid in the rough undulating surface is obtained, the X and Y grid coordinates of the undulating surface are kept unchanged, and the first elevation coordinate data is adjusted by a preset elevation coordinate conversion formula to adjust each first elevation coordinate data to a coefficient that varies around 1. At this time, the X, Y, and Z point cloud data of the rough undulating surface where this series of grids are located are converted into a set of nonlinear variation coefficient matrices X, Y, and Z'.

[0029] Among them, the preset elevation coordinate conversion formula is: ; (1) In the formula, is the converted second elevation coordinate data corresponding to the i-th grid; is the adjustment factor; is the first elevation coordinate data of the i-th grid; max(Z) is the maximum elevation coordinate data of the rough undulating surface where the grid is located.

[0030] Optionally, since the nonlinear variation coefficient matrix will be further weighted in the subsequent regression calculation, the value of the adjustment coefficient is not strictly limited and can be between 0 and 1. In practical applications, when the adjustment coefficient is 0.2, the obtained nonlinear variation coefficient matrix fluctuates in the range of 0.8-1.2.

[0031] S30: determining a plurality of linear boundary conditions based on measured geostress data of the area to be inverted, and adjusting the plurality of linear boundary conditions based on a plurality of groups of nonlinear variation coefficient matrices to generate a plurality of nonlinear boundary conditions; S40: applying nonlinear boundary conditions to the three-dimensional numerical model, and obtaining calculated values ​​of stress components at each preset measuring point through numerical simulation.

[0032] For steps S30-S40, the stress component with a greater impact on the inversion area is determined according to the measured value of the ground stress as the linear boundary condition applied to the three-dimensional numerical model. Thereafter, the linear boundary condition is adjusted by a nonlinear variation coefficient to obtain a nonlinear boundary adjustment. Thereafter, the nonlinear boundary condition is applied to the three-dimensional numerical model, and the calculated value of the stress component of each preset measurement point is extracted through numerical simulation.

[0033] Optionally, the geostress field in the area to be inverted is regarded as a linear superposition of the self-weight stress field and the tectonic stress field. The self-weight stress field and the tectonic stress field are simulated by decomposition, and finally the geostress field is formed based on the superposition principle. The geostress field is composed of the self-weight stress field and the tectonic stress field. Therefore, in addition to the self-weight stress, six boundary conditions are used to define the stress state of the rock mass. Specifically, the tectonic stress boundary conditions include at least one of the following: two horizontally uniformly distributed loads, two horizontally triangularly distributed loads, a uniformly distributed shear load in a horizontal plane, and a triangularly distributed shear load in a horizontal plane.

[0034] In actual application scenarios, select any set of nonlinear variation coefficient matrices to adjust a linear boundary condition, that is, multiply the nonlinear variation coefficient of the grid at the same position and the linear stress (displacement) variable value to obtain the nonlinear stress (displacement) component value at each grid. Calculate in sequence according to the above method to obtain the nonlinear boundary conditions of each model boundary surface in the three-dimensional numerical model. For example, by adjusting six linear structural boundary conditions respectively through 10 sets of nonlinear variation coefficients, 60 nonlinear boundary conditions can be obtained.

[0035] S50: performing regression analysis on the calculated values ​​of multiple stress components and the measured values ​​of multiple stress components to obtain a regression coefficient of each nonlinear boundary condition, weighting them using the regression coefficient as a weight, and substituting them into the three-dimensional numerical model for calculation to obtain the geostress field of the three-dimensional numerical model.

[0036] In this step, all calculated values ​​of stress components are extracted as independent variables, and the measured values ​​of transformed stress components are used as dependent variables. The regression coefficient of each nonlinear boundary condition is determined by the least squares method and the linear superposition principle. Afterwards, the stress data under the nonlinear boundary conditions is multiplied by the regression coefficient and substituted into the three-dimensional numerical model for calculation through linear superposition to obtain a non-uniform and high-precision geostress field.

[0037] In one embodiment of the present application, a specific inversion stress field scheme for generating a three-dimensional numerical model is provided. In S50, a regression analysis is performed on multiple calculated values ​​of stress components and multiple measured values ​​of stress components to obtain a regression coefficient of each nonlinear boundary condition, which is weighted using the regression coefficient as a weight and brought into the three-dimensional numerical model for calculation to obtain a ground stress field scheme of the three-dimensional numerical model, which specifically includes the following steps S51-S53: S51: Based on the measured values ​​of multiple stress components and the calculated values ​​of multiple stress components, a residual sum of squares formula of the least squares method is established; S52: Based on the principle of least squares method, the residual sum of squares is minimized and the regression coefficient of each nonlinear boundary condition is calculated; S53: Determine an expression for calculating the regression stress value based on the regression coefficient; S54: Based on the expression for calculating the regression stress value, invert the regression stress value of any point in the three-dimensional numerical model to obtain the geostress field of the three-dimensional numerical model after inversion.

[0038] For steps S51-S54, when performing multivariate linear regression calculations, different three-dimensional numerical models are independently calculated under the influence of each nonlinear boundary condition. Then, the regression coefficient of each nonlinear boundary condition is derived through regression using the measured values ​​of the stress components of each preset measurement point. Finally, a regression equation reflecting the relationship between the initial geostress field and each influencing factor is established.

[0039] The regression equation is: ; (2) In the formula, is the regression stress value of the kth preset measurement point; is the regression coefficient of the i-th nonlinear boundary condition; is the calculated value of the stress component under the i-th nonlinear boundary condition at the k-th preset measurement point.

[0040] It can be understood that i varies from 1 to n, and n is different for different nonlinear boundary conditions. In this embodiment, n is equal to 7. Assume that the measured value of the jth stress component at the kth preset measurement point is recorded as , where m represents the total number of preset measurement points and b represents the total number of stress components. ) and the calculated value of the stress component ( ) can be expressed as In order to minimize the cumulative error, the residual sum of squares between the measured stress component and the calculated stress component must be calculated. The residual sum of squares formula Q is: ; (3) In the formula, Q is the residual sum of squares; is the measured value of the stress component of the jth stress component at the kth preset measurement point; is the calculated stress component value of the jth stress component at the kth preset measurement point under the i-th nonlinear boundary condition, j=1, 2, 3, ..., b; m is the total number of preset measurement points; is the regression coefficient of the i-th nonlinear boundary condition, i=1, 2, 3, ..., n; n is the total number of regression coefficients.

[0041] Then, according to the least squares method, the residual sum of squares is minimized, that is, Q Taking the partial derivative and setting it to zero, we get The normal equations of (i=1, 2, ..., n) are solved to obtain the regression coefficients under each nonlinear boundary condition, and its solution is unique. The optimal nonlinear boundary is obtained by adjusting the linear boundary under the guidance of a set of nonlinear variation coefficient matrices that generate rough undulating surfaces.

[0042] The regression coefficient is: ; (4) In the formula, is the regression coefficient of the i-th nonlinear boundary condition; is the regression coefficient of the rth nonlinear variation coefficient matrix under the ith nonlinear boundary condition, r=1, 2, ..., λ; λ is the number of nonlinear variation coefficient matrices under each nonlinear boundary condition.

[0043] Afterwards, the expression for calculating the regression stress value is determined based on the regression coefficient, indicating that the initial geostress field is the result of the combined effects of gravity and tectonic movement. Based on the expression for calculating the regression stress value, the regression stress value of any point in the three-dimensional numerical model is inverted, thereby obtaining the spatial distribution of the overall stress field in the three-dimensional numerical model after inversion.

[0044] The expression for calculating the regression stress value is: ; (5) In the formula, is the regression stress value of the kth preset measurement point; is the regression coefficient of the rth nonlinear variation coefficient matrix for the i-th nonlinear boundary condition, i=1,2,...,n, r=1,2,...,λ; is a single-column matrix of the calculated values ​​of the stress components at the kth preset measurement point under the i-th nonlinear boundary condition; It is the calculated value of stress at the kth preset measuring point under the action of deadweight.

[0045] In one embodiment of the present application, a specific regression coefficient rationality test scheme is provided. After S52, that is, based on the principle of least squares method, the residual sum of squares is minimized, and the regression coefficient scheme for each nonlinear boundary condition is calculated, the following steps S521-S523 are also included: S521: Calculate the regression stress value of any point in the three-dimensional numerical model based on the regression equation; S522: Based on the regression stress value, the rationality of the regression coefficient is tested; S523: If the test is qualified, the regression coefficient is adopted; if the test is unqualified, the nonlinear variation coefficient is adjusted and recalculated until a reasonable regression coefficient is obtained.

[0046] For steps S521-S523, based on the established regression equation, the regression stress value of any point in the three-dimensional numerical model is calculated, which is the regression result. Afterwards, the rationality test of the regression result is performed. If the regression result test is qualified, the regression coefficient is adopted; if the regression result test is unqualified, the nonlinear change coefficient is adjusted and recalculated until a reasonable regression coefficient is obtained.

[0047] Through the above method, the rationality of the regression coefficient is checked to ensure that the final regression equation conforms to the actual situation and ensure the accuracy of the inverted stress field.

[0048] In one embodiment of the present application, a specific regression equation rationality test scheme is provided. In S522, a rationality test scheme is performed on the regression coefficient based on the regression stress value, which specifically includes the following steps S5221-S5223: S5221: Calculate an average value of the measured stress components based on the measured values ​​of the multiple stress components; S5222: Calculate the regression sum of squares and the residual sum of squares based on the average value of the regression stress value and the measured stress component; S5223: Calculate the complex correlation coefficient based on the regression sum of squares and the residual sum of squares. If the complex correlation coefficient is within a preset qualified interval, it is determined that the regression coefficient test is qualified. Among them, the expression of the average value of the measured stress component is: ; (6) In the formula, is the average value of the measured stress component, is the measured value of the stress component of the jth stress component at the kth preset measurement point; m is the total number of preset measurement points; n is the total number of regression coefficients; The expression for the regression sum of squares is: ; (7) In the formula, is the regression stress value of the jth stress component at the kth preset measurement point, is the average value of the measured stress component; The expression of the multiple correlation coefficient is: ; (8) In the formula, R is the multiple correlation coefficient; Q is the residual sum of squares; U is the regression sum of squares.

[0049] For steps S5221-S5223, after calculating the average value of the measured stress component, the regression sum of squares and the residual sum of squares are calculated based on the regression stress value and the average value of the measured stress component. Thereafter, the complex correlation coefficient is calculated based on the regression sum of squares and the residual sum of squares. If the complex correlation coefficient is within the preset qualified interval, it is determined that the regression coefficient test is qualified.

[0050] Optionally, the qualified interval range of R is set to [0, 1]. Generally, the closer the R value is to 1, the better the regression effect is.

[0051] It can be seen that in the above scheme, a three-dimensional numerical model is established based on the area to be inverted, and a nonlinear variation coefficient generation method is used to generate a nonlinear variation coefficient matrix that matches the boundary size of the three-dimensional numerical model. The nonlinear variation coefficient matrix is ​​used as the key cornerstone, and the nonlinear boundary conditions are constructed in combination with multiple small grids divided by the model boundary. Subsequently, nonlinear boundary conditions are applied to the three-dimensional numerical model, and finally the stress field is inverted according to the regression coefficient. Through the above method, the limitations of the linear inversion technology used in the relevant technology in expressing the initial non-uniform ground stress field are successfully overcome, and the accuracy of the ground stress inversion of the three-dimensional numerical model can be effectively improved.

[0052] In a second embodiment of the present application, a specific nonlinear geostress field inversion method is provided for the first embodiment. Figure 2 As shown, the following steps are included: Step 1: Establish a three-dimensional numerical model of the specific project area to be inverted, determine the rock mass parameters and collect field stress measured data.

[0053] Step 2: Generate a rough undulating surface according to the boundary size of the three-dimensional numerical model, and the rough undulating surface has X, Y, and Z three-dimensional coordinate data. Keep the X and Y grid coordinates of the rough undulating surface unchanged, adjust the Z coordinate of the rough undulating surface, and adjust the Z value to a series of coefficients that vary around 1. At this time, the X, Y, and Z point cloud data of the undulating surface are converted into a set of nonlinear variation coefficient matrices X, Y, and Z'. The preset undulating surface generation algorithm of the rough undulating surface includes but is not limited to: Bubbling method, WM equation method, waveform superposition, rock structure surface scanning data, terrain undulation data, etc.

[0054] Step 3: The nonlinear variation coefficient matrix generated by the preset rough surface generation algorithm forms the basis for creating nonlinear boundary conditions in the three-dimensional numerical model. Optionally, nonlinear boundary conditions can be divided into the following five types according to the different boundary application methods: symmetric displacement boundary, symmetric stress boundary, asymmetric displacement boundary, asymmetric stress boundary and stress-displacement mixed boundary. Figures 3 to 7 As shown in the figure, the two-dimensional creation process of various nonlinear boundaries in the XZ plane is demonstrated, where: Figure 3 Schematic diagram of the creation process of symmetric displacement boundaries. Figure 4 Schematic diagram of the creation process of symmetric displacement boundaries. Figure 5 Schematic diagram of the creation process of asymmetric displacement boundary. Figure 6 Schematic diagram of the creation process of asymmetric stress boundary. Figure 7 is a schematic diagram of the creation process of the stress-displacement mixed boundary. Specifically, σ x It means that the stress boundary is applied. Indicates that a displacement boundary is applied. Create n sets of nonlinear coefficient matrices. Under the guidance of n sets of nonlinear coefficient matrices, n sets of nonlinear boundaries are obtained, and the weights of these nonlinear coefficient matrices (δ1 to δ n ), and finally the optimal nonlinear boundary is obtained by fitting. It should be noted that Figure 3 and Figure 4 As shown in , when the same nonlinear coefficient matrix is ​​used at both ends of the numerical model, the resulting nonlinear boundary will show symmetry, and the optimal nonlinear boundary finally fitted also maintains the symmetric characteristics; as shown in Figure 5 and Figure 6 As shown in , when different nonlinear coefficient matrices are applied on both sides of the numerical model, the resulting nonlinear boundary will be asymmetric, and the optimal nonlinear boundary obtained by the final fitting will also be asymmetric; Figure 7As shown in the figure, when the right boundary of the numerical model remains fixed, the linear stress boundary on the left is adjusted using n groups of different nonlinear coefficient matrices, thereby generating n groups of different nonlinear stress boundaries. The left boundary is then fitted to determine the optimal boundary under this specific condition.

[0055] Step 4: Under different nonlinear boundary conditions, the stress components at different preset measurement points are calculated, and the regression coefficients of each pair of nonlinear boundaries are obtained through the basic principle of least squares method. Finally, the ground stress field is derived by linearly superimposing nonlinear boundary conditions.

[0056] In the above manner, a nonlinear variation coefficient generation method is used to generate a nonlinear variation coefficient matrix that matches the model boundary size. This matrix constitutes the key cornerstone for constructing a nonlinear boundary in a numerical model. The boundary of the model is divided into multiple small grids for more accurate analysis. Through this nonlinear variation coefficient matrix, the lateral boundary of the numerical model can be comprehensively adjusted and optimized. Compared with the linear boundary conditions used in related technologies, this application overcomes the limitations of linear inversion technology in expressing the initial non-uniform stress field and significantly improves the accuracy of the inversion.

[0057] In a third embodiment of the present application, experimental data of a specific nonlinear geostress field inversion method based on symmetric displacement boundaries is provided for the first embodiment.

[0058] Taking a mining project as an example, the applicability of this application method is discussed.

[0059] Step 1: On-site field data collection.

[0060] Specifically, according to the engineering geological report, the entire project impact area is selected, the regional structural development is fully considered, and a three-dimensional numerical model is established. Among them, the orientations of the X-axis, Y-axis and Z-axis of the numerical model are: the X-axis direction is along the ore body strike, the Y-axis direction is along the ore body dip, and the Z-axis direction is vertically upward. Figure 8 The figure shows the schematic diagram of the structure of the three-dimensional numerical model. In the process of establishing the three-dimensional numerical model, in order to reduce the influence of the boundary effect, the calculation range should be appropriately expanded according to the size of the area to be inverted. Taking into account the influence of topography and structural factors, it is determined that 1000m is taken as the long side of the model along the axis of the ore body, 1000m is taken as the short side of the model along the dip direction of the ore body, and 1000m is taken as the height of the model along the vertical direction to establish a three-dimensional numerical model. The measured ground stress data of multiple preset measurement points of the mining project are collected. The measured ground stress data of each measurement point contains three principal stress values: σ1, σ2 and σ3. Through the coordinate conversion formula between the principal stress and the total stress component, the measured principal stress values ​​of the area to be inverted are converted into 6 stress components corresponding to the numerical calculation model: , , , , , , and the measured values ​​of stress components are obtained. As shown in Table 1, the coordinates of the preset measurement points and their stress components corresponding to the three-dimensional numerical model are shown.

[0061] Table 1

[0062] Optionally, the elastic constitutive model is selected as the constitutive model used for the inversion of the engineering ground stress field, and the physical and mechanical parameters of the formation are selected: bulk modulus, shear modulus and density. As shown in Table 2, the physical and mechanical parameters of the rock mass corresponding to the rock mass type are shown.

[0063] Table 2

[0064] Step 2: Create nonlinear displacement boundary conditions.

[0065] Step (1): Preselect linear structural displacement boundary conditions: Select linear structural displacement boundary conditions that have a greater impact on the in-situ stress in the inversion area to create nonlinear displacement boundary conditions. In this embodiment, self-weight stress, unit gradient displacement in the X direction, and unit gradient displacement in the Y direction are selected as key stress influencing factors. In addition, the unit displacement is set to 5 cm.

[0066] Step (2): Select the bubbling method to generate a nonlinear variation coefficient matrix.

[0067] Specifically, the bubbling method simulates the rough undulating surface by simulating the growth and fusion process of bubbles to generate random geometric shapes. Specifically, by defining a series of "bubbles" (i.e., random points or areas of Gaussian distribution or other distributions), and then gradually "growing" these bubbles until they contact each other or reach the set boundary conditions, the growth process of the bubbles can be controlled by random functions or regular functions to generate complex rough undulating surfaces, and then through the elevation coordinate transformation, a nonlinear variation coefficient matrix is ​​obtained. Figures 9 to 11 The figure shows a schematic diagram of the creation process of generating a rough undulating surface using the bubbling method. The radius of each bubble, denoted as Rb, conforms to the normal distribution described by formula (9). These hemispherical spheres bubble up (bubbling up) or down (bubbling down) randomly, and their Z coordinates are summed. As the number of bubbles increases, the synthesized surface gradually transforms into a rough surface.

[0068] ; (9) Where μ and σ represent the mean and standard deviation of a specific statistical parameter respectively; X~(μ, σ2) means that the parameter follows a normal distribution; in the bubbling method, Rb~(μb, ), μb and σb represent the mean and standard deviation of the bubble radius generated, respectively. An increase in μb indicates a larger average radius of the bubble, while a higher σb indicates a greater difference in bubble radius, and vice versa. In practical applications, μb is usually set to 0, and the radius size of the bubble generated is determined only by σb. If a negative radius is generated, its absolute value is taken as the bubble radius. The number of bubbles N is also the key to using the bubbling method to create a rough surface. The parameter combination set in this embodiment is: σb=10, μb=10, N=15000. In addition, 10 sets of non-uniform coefficient matrices are used for each different nonlinear displacement boundary in the regression process.

[0069] Step (3): Creation of nonlinear displacement boundary.

[0070] Specifically, the nonlinear variation coefficient matrix generated by the bubbling method constitutes the basis for creating nonlinear displacement boundary conditions in the three-dimensional numerical model. The size of the nonlinear variation coefficient should match the size of the three-dimensional numerical model boundary, and the model boundary is subdivided into multiple small grids. Therefore, the nonlinear variation coefficient matrix can comprehensively adjust the lateral displacement boundary of the three-dimensional numerical model. In this embodiment, the lateral boundary size of the three-dimensional numerical model is 1000m×1000m, and the size of each small grid is 20m×20m. The lateral boundary of the three-dimensional numerical model is divided into 50×50=2500 grids, and the size of the nonlinear variation coefficient matrix to be generated is 50×50.

[0071] Step 3: Construct the geostress field of the three-dimensional numerical model.

[0072] Specifically, the stress component values ​​at different preset measurement points are calculated, and the regression coefficients corresponding to each nonlinear displacement boundary condition are obtained through the basic principle of the least squares method. Finally, the nonlinear displacement boundary condition is obtained through linear superposition, and is substituted into the three-dimensional numerical model for calculation to obtain a non-uniform high-precision ground stress field. Fig.12 and Fig.13 As shown in Figure 3, the maximum principal stress cloud map and the minimum principal stress cloud map of the three-dimensional numerical model geostress field are shown. As shown in Table 3, the measured values, regression values ​​and absolute errors of the stress components of the geostress measurement points in the calculation coordinate system are shown. Among them, the average absolute error is 1.50MPa, the complex correlation coefficient R of the regression value is 0.9411, and the complex correlation coefficient is close to 1, which shows that the calculation results of this embodiment have a high correlation with the measured values, further verifying the accuracy and reliability of the model.

[0073] Table 3

[0074] In a fourth embodiment of the present application, experimental data of a specific nonlinear geostress field inversion method based on an asymmetric stress boundary is provided for the first embodiment.

[0075] Step 1: On-site field data collection.

[0076] This step is the same as step 1 in Example 2.

[0077] Step 2: Create nonlinear stress boundary conditions.

[0078] Step (1): Select linear tectonic stress boundary conditions: Select factors that have a greater impact on the ground stress in the inversion area to create nonlinear stress boundary conditions. In this embodiment, self-weight stress, X-axis unit gradient stress, and Y-axis unit gradient stress are selected as key stress influencing factors. In addition, the stress value corresponding to the unit stress is set to 1MPa.

[0079] Step (2): Select the WM equation method as the method for generating the nonlinear variation coefficient matrix.

[0080] Specifically, Fig.14 The figure shows a flow chart of generating a rough undulating surface using the WM equation method. The WM fractal function is used to expand the one-dimensional function to two dimensions, and the calculated function value is used as the height value to construct a three-dimensional surface. By adjusting the parameters, the roughness, frequency, directionality and detail level of the undulating surface can be controlled to generate a variety of three-dimensional surfaces. Finally, the generated three-dimensional surfaces are superimposed to obtain the required three-dimensional rough undulating surface.

[0081] Among them, the elevation value z (x, y) of the rough undulating surface can be expressed as: ; (10) Where An, Bn and Cn are independent random numbers, An and Bn follow uniform distribution, so that An, Bn∈[0, 2π], Cn obeys standard normal distribution; D is the fractal dimension; set 1<λ<1.5.

[0082] Step (3): Creation of nonlinear stress boundary.

[0083] Specifically, the nonlinear variation coefficient matrix generated by the WM equation method forms the basis for creating nonlinear stress boundary conditions in the three-dimensional numerical model. The size of the nonlinear variation coefficient should match the size of the three-dimensional numerical model boundary, which is subdivided into multiple small grids. Therefore, the nonlinear variation coefficient matrix can comprehensively adjust the lateral stress boundary of the three-dimensional numerical model.

[0084] The linear stress boundaries on the four sides of the model are adjusted in turn using the nonlinear variation coefficient matrix generated by the WM equation method to obtain the asymmetric nonlinear model stress boundaries.

[0085] In this embodiment, 10 pairs of nonlinear variation coefficient matrices are applied to the boundary of the three-dimensional numerical model, and each pair of nonlinear variation coefficients acts independently on the opposite sides of the model boundary. The lateral boundary size of the three-dimensional numerical model is 1000m×1000m, and the size of each small grid is 20m×20m. The lateral boundary of the three-dimensional numerical model is divided into 50×50=2500 grids, and the size of the nonlinear variation coefficient matrix to be generated is 50×50.

[0086] Step 3: Construct the geostress field of the three-dimensional numerical model.

[0087] Specifically, the stress component values ​​at different preset measurement points are calculated, and the regression coefficients corresponding to each nonlinear stress boundary condition are obtained through the basic principle of the least squares method. Finally, the nonlinear asymmetric stress boundary condition is obtained through linear superposition, and then substituted into the three-dimensional numerical model for calculation to obtain a non-uniform high-precision ground stress field. Fig.15 and Fig.16 As shown in Table 4, the maximum principal stress cloud map and the minimum principal stress cloud map of the three-dimensional numerical model geostress field are shown. As shown in Table 4, the measured values, regression values ​​and absolute errors of the stress components of the geostress measurement points in the calculation coordinate system are shown. The average absolute error is 1.49 MPa, the complex correlation coefficient R of the regression value is 0.9225, and the complex correlation coefficient is close to 1, which shows that the calculation results of this embodiment have a high correlation with the measured values, further verifying the accuracy and reliability of the model.

[0088] Table 4

[0089] In a fifth embodiment of the present application, experimental data of a specific nonlinear geostress field inversion method based on a stress-displacement mixed boundary is provided for the first embodiment.

[0090] Step 1: On-site field data collection.

[0091] This step is the same as step 1 in Example 2.

[0092] Step 2: Create nonlinear stress boundary conditions.

[0093] Step (1): Select linear tectonic stress boundary conditions: Select factors that have a greater impact on the ground stress in the inversion area to create nonlinear stress boundary conditions. In this embodiment, self-weight stress, X-axis unit gradient stress, and Y-axis unit gradient stress are selected as key stress influencing factors. In addition, the stress value corresponding to the unit stress is set to 1MPa.

[0094] Step (2): Generate a nonlinear variation coefficient matrix using rock structural surface data.

[0095] Specifically, the real rock structure of the area to be inverted is selected, and the real rock structure image is collected. By scanning the real rock structure structural surface data, a rough undulating surface is obtained, and then the elevation coordinates of the structural surface of the real rock structure are converted into a nonlinear variation coefficient matrix.

[0096] Step (3): Creation of nonlinear stress boundary.

[0097] Specifically, the nonlinear variation coefficient matrix generated by scanning the real rock structure surface data constitutes the basis for creating nonlinear stress boundary conditions. The dimension of this coefficient matrix needs to accurately match the size of the model boundary, so the three-dimensional numerical model boundary is carefully divided into many small grids. Based on this, the nonlinear variation coefficient matrix can comprehensively adjust the lateral stress boundary in the three-dimensional numerical model. In this embodiment, the lateral boundary size of the three-dimensional numerical model is set to 1000m×1000m, and in order to simulate and analyze more accurately, the size of each small grid is set to 20m×20m. According to this division standard, the lateral boundary of the entire three-dimensional numerical model is divided into 50 horizontal grids and 50 vertical grids, totaling 2500 small grids. Accordingly, in order to match the layout of these 2500 small grids and effectively adjust the stress magnitude on these grids, the size of the nonlinear variation coefficient matrix to be generated should also be set to 50×50. This nonlinear variation coefficient matrix will be able to reflect the nonlinear changes in the lateral stress of each small grid in detail, thereby ensuring the accuracy and reliability of the numerical model.

[0098] In this embodiment, in view of the asymmetric characteristics of the model, a stress-displacement mixed boundary condition is selected for numerical calculation, which combines the displacement boundary and the stress boundary. Specifically, the displacement boundary is achieved by fixing the velocity (or displacement) of the boundary on one side of the model, while the stress boundary applies stress on the other side of the model. Therefore, when setting the boundary conditions of the model, we only need to make nonlinear adjustments to the stress boundary on one side, while the displacement boundary remains relatively fixed.

[0099] Step 3: Construct the geostress field of the three-dimensional numerical model.

[0100] Specifically, under different nonlinear mixed boundaries, the stress components at different measuring points were calculated, and the weight coefficients of each nonlinear mixed boundary were obtained through the basic principle of least squares method. Finally, the geostress field was derived by linearly superimposing the nonlinear mixed boundaries.

[0101] In a sixth embodiment of the present application, experimental data of a specific nonlinear geostress field inversion method for a nonlinear asymmetric displacement boundary is provided for the first embodiment.

[0102] Step 1: On-site field data collection.

[0103] Specifically, according to the engineering geological report, the area including all engineering impact areas was selected, and the regional structural development was fully considered to establish a three-dimensional numerical model. Among them, the orientations of the X-axis, Y-axis and Z-axis of the numerical model are: the X-axis direction is along the strike of the ore body, the Y-axis direction is along the dip of the ore body, and the Z-axis direction is vertically upward. In the process of establishing a three-dimensional numerical model, in order to reduce the influence of the boundary effect, the calculation range should be appropriately expanded according to the size of the area to be inverted. Taking into account the influence of topography and structural factors, it is determined that 1000m is taken as the long side of the model along the axis of the ore body, 1000m is taken as the short side of the model along the dip of the ore body, and 1000m-800m is taken as the model height in the vertical direction to establish a three-dimensional numerical model. The measured ground stress data of multiple preset measurement points of the mining project are collected. The measured ground stress data of each measurement point include three principal stress values: σ1, σ2 and σ3. Through the coordinate conversion formula between the principal stress and the total stress component, the measured principal stress value of the area to be inverted is converted into 6 stress components corresponding to the numerical calculation model to obtain the measured value of the stress component. As shown in Table 5, the coordinates of the preset measurement points and their stress components corresponding to the three-dimensional numerical model.

[0104] Table 5

[0105] Optionally, the elastic constitutive model is selected as the constitutive model used for the inversion of the engineering ground stress field, and the physical and mechanical parameters of the formation are selected: bulk modulus, shear modulus and density. As shown in Table 6, the physical and mechanical parameters of the rock mass corresponding to the three-dimensional numerical model are shown.

[0106] Table 6

[0107] Step 2: Create nonlinear displacement boundary conditions.

[0108] Step (1): Preselect linear structural displacement boundary conditions: Select linear structural displacement boundary conditions that have a greater impact on the in-situ stress in the inversion area to create nonlinear displacement boundary conditions. In this embodiment, self-weight stress, unit gradient displacement in the X direction, and unit gradient displacement in the Y direction are selected as key stress influencing factors. In addition, the unit displacement is set to 5 cm.

[0109] Step (2): Select the bubbling method to generate a nonlinear variation coefficient matrix.

[0110] The parameter combination set in this embodiment is: σb=10, μb=0, N=15000. In addition, 10 groups of non-uniform coefficient matrices are used for each different nonlinear displacement boundary in the regression process.

[0111] Step (3): Creation of nonlinear displacement boundary.

[0112] Specifically, the nonlinear variation coefficient matrix generated by the bubbling method constitutes the basis for creating nonlinear displacement boundary conditions in the three-dimensional numerical model. The size of the nonlinear variation coefficient should match the size of the three-dimensional numerical model boundary, and the model boundary is subdivided into multiple small grids. Therefore, the nonlinear variation coefficient matrix can comprehensively adjust the lateral displacement boundary of the three-dimensional numerical model. In the present embodiment, 10 pairs of nonlinear variation coefficient matrices are applied to the boundary of the three-dimensional numerical model, and each pair of nonlinear variation coefficients acts independently on the opposite sides of the model boundary. The lateral boundary size of the three-dimensional numerical model is 1000m×1000m, and the size of each small grid is 20m×20m. The lateral boundary of the three-dimensional numerical model is divided into 50×50=2500 grids, and the size of the nonlinear variation coefficient matrix to be generated is 50×50.

[0113] Step 3: Construct the geostress field of the three-dimensional numerical model.

[0114] Specifically, the calculated values ​​of stress components at different preset measurement points are calculated, and the regression coefficients corresponding to each nonlinear displacement boundary condition are obtained through the basic principle of the least squares method. Finally, the nonlinear asymmetric displacement boundary is obtained by linear superposition, and it is substituted into the three-dimensional numerical model for calculation to obtain a non-uniform and high-precision ground stress field. As shown in Table 7, the measured values, regression values ​​and absolute errors of the stress components of the ground stress measurement points under the calculation coordinate system are shown. Among them, the average absolute error is 1.01MPa, the complex correlation coefficient R of the regression value is 0.9063, and the complex correlation coefficient is close to 1, which shows that the calculation results of this embodiment have a high correlation with the measured values, which further verifies the accuracy and reliability of the model.

[0115] Table 7

[0116] The calculation results show that the inversion results of the geostress field obtained by the embodiment of the present application are highly consistent with the distribution law of the actual geostress field, thereby verifying the effectiveness and reliability of the method.

[0117] In one embodiment, a nonlinear geostress field inversion device is provided, and the nonlinear geostress field inversion device corresponds one-to-one to the nonlinear geostress field inversion method in the above embodiment. Fig.17As shown, the nonlinear geostress field inversion device includes: a first construction module 101, a first determination module 102, a second construction module 103, a first generation module 104, a second determination module 105, a second generation module 106, a third generation module 107 and a fourth generation module 108. The functional modules are described in detail as follows: The first construction module 101 is used to construct a three-dimensional numerical model based on engineering geological data, measured geostress data and stratum lithology data of the area to be inverted; the first determination module 102 is used to determine measured values ​​of multiple stress components at multiple preset measurement points; The second construction module 103 is used to establish a plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model; the first generation module 104 is used to generate a plurality of groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the plurality of rough undulating surfaces; The second determination module 105 is used to determine a plurality of linear boundary conditions based on the measured ground stress data of the area to be inverted; the second generation module 106 is used to adjust the plurality of linear boundary conditions based on a plurality of groups of nonlinear variation coefficient matrices to generate a plurality of nonlinear boundary conditions; The third generation module 107 is used to apply multiple nonlinear boundary conditions on the three-dimensional numerical model, and obtain the calculated value of the stress component of each preset measurement point through numerical simulation; The fourth generation module 108 is used to perform regression analysis on the calculated values ​​of multiple stress components and the measured values ​​of multiple stress components to obtain the regression coefficient of each nonlinear boundary condition, weight them with the regression coefficient as the weight, and bring them into the three-dimensional numerical model for calculation to obtain the geostress field of the three-dimensional numerical model.

[0118] The present invention provides a nonlinear geostress field inversion device, which establishes a three-dimensional numerical model based on the area to be inverted, uses a nonlinear variation coefficient generation method to generate a nonlinear variation coefficient matrix that matches the boundary size of the three-dimensional numerical model, uses the nonlinear variation coefficient matrix as a key cornerstone, and constructs nonlinear boundary conditions in combination with multiple small grids divided by the model boundary. Thereafter, nonlinear boundary conditions are applied to the three-dimensional numerical model, and finally the stress field is inverted according to the regression coefficient. Through the above method, the limitations of the linear inversion technology used in the related art in expressing the initial non-uniform geostress field are successfully overcome, and the accuracy of the three-dimensional numerical model geostress inversion can be effectively improved.

[0119] Regarding the device in the above embodiment, the specific manner in which each unit module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here. Each module in the above nonlinear geostress field inversion device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0120] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned nonlinear geostress field inversion method are implemented.

[0121] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned nonlinear geostress field inversion method are implemented.

[0122] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0123] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile readable storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0124] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0125] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A nonlinear geostress field inversion method, characterized in that: include: Based on the engineering geological data, measured geostress data and stratum lithology data of the area to be inverted, a three-dimensional numerical model is constructed, and the measured values ​​of multiple stress components at multiple preset measurement points are determined; Establishing a plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model, and generating a plurality of groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the plurality of rough undulating surfaces; Based on the measured geostress data of the area to be inverted, a plurality of linear boundary conditions are determined, and based on the plurality of groups of nonlinear variation coefficient matrices, the plurality of linear boundary conditions are adjusted to generate a plurality of nonlinear boundary conditions; Applying the multiple nonlinear boundary conditions to the three-dimensional numerical model, and obtaining the calculated value of the stress component of each preset measuring point through numerical simulation; Regression analysis is performed on multiple calculated values ​​of stress components and multiple measured values ​​of stress components to obtain a regression coefficient of each nonlinear boundary condition, which is weighted using the regression coefficient as a weight and substituted into the three-dimensional numerical model for calculation to obtain the geostress field of the three-dimensional numerical model.

2. The method according to claim 1, characterized in that The step of constructing a three-dimensional numerical model based on the engineering geological data, measured ground stress data and stratum lithology data of the area to be inverted, and determining measured values ​​of multiple stress components at multiple preset measurement points specifically includes: Acquire engineering geological data of the area to be inverted, the stratum lithology data, and the measured ground stress data of the plurality of preset measurement points; Establishing the three-dimensional numerical model based on the engineering geological data and the stratum lithology data; Coordinate transformation is performed on multiple principal stress values ​​in the measured geostress data to obtain measured values ​​of multiple stress components in a calculation coordinate system.

3. The method according to claim 1, characterized in that The step of establishing a plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model, and generating a plurality of groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the plurality of rough undulating surfaces, specifically comprises: Using a preset undulating surface generation algorithm, generating the plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model according to a preset number; For any rough and undulating surface, the rough and undulating surface is divided into a plurality of grids based on a preset division size; Get the three-dimensional coordinate data of each grid; Coordinate transformation is performed on the three-dimensional coordinate data to generate a set of nonlinear change coefficient matrices of the model boundary surface.

4. The method according to claim 3, characterized in that The step of performing coordinate transformation on the three-dimensional coordinate data to generate a set of nonlinear variation coefficient matrices of the model boundary surface specifically includes: Based on the three-dimensional coordinate data, determining first elevation coordinate data of each grid; Using a preset elevation coordinate conversion formula, converting the first elevation coordinate data of each grid to obtain converted second elevation coordinate data; Based on the multiple horizontal coordinate data and the multiple second elevation coordinate data of the multiple grids, a set of nonlinear variation coefficient matrices corresponding to the model boundary surface are generated; Among them, the preset elevation coordinate conversion formula is: ； In the formula, is the converted second elevation coordinate data corresponding to the i-th grid; is the adjustment factor; is the first elevation coordinate data of the ith grid; max(Z) is the maximum elevation coordinate data of the rough undulating surface where the grid is located.

5. The method according to claim 1, characterized in that The step of performing regression analysis on the calculated values ​​of multiple stress components and the measured values ​​of multiple stress components to obtain the regression coefficient of each nonlinear boundary condition, weighting them with the regression coefficient as weight, and substituting them into the three-dimensional numerical model for calculation to obtain the geostress field of the three-dimensional numerical model specifically includes: Based on the measured values ​​of the multiple stress components and the calculated values ​​of the multiple stress components, a residual sum of squares formula of the least squares method is established; The residual sum of squares formula is: ； In the formula, Q is the residual sum of squares; is the measured value of the stress component of the jth stress component at the kth preset measurement point; is the calculated stress component value of the jth stress component at the kth preset measurement point under the i-th nonlinear boundary condition, j=1, 2, 3, ..., b; m is the total number of preset measurement points; is the regression coefficient of the i-th nonlinear boundary condition, i=1, 2, 3, ..., n; n is the total number of regression coefficients; Based on the principle of least squares method, the residual sum of squares is minimized, and the regression coefficient of each nonlinear boundary condition is calculated; Among them, the regression coefficient is: ； In the formula, is the regression coefficient of the i-th nonlinear boundary condition; is the regression coefficient of the rth nonlinear variation coefficient matrix under the ith nonlinear boundary condition, r=1, 2, ..., λ; λ is the number of nonlinear variation coefficient matrices under each nonlinear boundary condition; Determine an expression for calculating the regression stress value based on the regression coefficient; The expression for calculating the regression stress value is: ； In the formula, is the regression stress value of the kth preset measurement point; is the regression coefficient of the rth nonlinear variation coefficient matrix for the i-th nonlinear boundary condition, i=1,2,...,n, r=1,2,...,λ; is a single-column matrix of the calculated values ​​of the stress components at the kth preset measurement point under the i-th nonlinear boundary condition; is the calculated stress value of the kth preset measurement point under the action of deadweight; Based on the regression equation, the regression stress value of any point in the three-dimensional numerical model is inverted to obtain the in-situ stress field of the three-dimensional numerical model after inversion.

6. The method according to claim 5, characterized in that After determining the expression for calculating the regression stress value based on the regression coefficient, the method further includes: Based on the regression equation, calculating the regression stress value of any point in the three-dimensional numerical model; Based on the regression stress value, the rationality of the regression coefficient is tested; If the test is qualified, the regression coefficient is adopted; if the test is unqualified, the nonlinear variation coefficient is adjusted and recalculated until a reasonable regression coefficient is obtained.

7. The method according to claim 6, characterized in that The step of performing a rationality test on the regression coefficient based on the regression stress value specifically includes: Based on the measured values ​​of multiple stress components, an average value of the measured stress components is calculated; Among them, the expression of the average value of the measured stress component is: ； In the formula, is the average value of the measured stress component, is the measured value of the stress component of the jth stress component at the kth preset measurement point; m is the total number of preset measurement points; n is the total number of regression coefficients; Calculating a regression sum of squares and a residual sum of squares based on the regression stress value and the average value of the measured stress component; The expression of regression square sum is: ； In the formula, is the regression stress value of the jth stress component at the kth preset measurement point, is the average value of the measured stress component; Based on the regression sum of squares and the residual sum of squares, a complex correlation coefficient is calculated, and if the complex correlation coefficient is within a preset qualified interval, it is determined that the regression coefficient test is qualified; The expression of the multiple correlation coefficient is: ； In the formula, R is the multiple correlation coefficient; Q is the residual sum of squares; U is the regression sum of squares.

8. A device for realizing the nonlinear geostress field inversion of claim 1, characterized in that: include: The first construction module is used to construct a three-dimensional numerical model based on the engineering geological data, ground stress measured data and stratum lithology data of the area to be inverted; A first determination module is used to determine the actual measured values ​​of multiple stress components at multiple preset measurement points; A second construction module is used to establish a plurality of rough undulating surfaces corresponding to each model boundary surface in the three-dimensional numerical model; A first generating module, configured to generate a plurality of groups of nonlinear variation coefficient matrices corresponding to each model boundary surface based on the three-dimensional coordinate data of the plurality of rough undulating surfaces; A second determination module is used to determine a plurality of linear boundary conditions based on the measured geostress data of the area to be inverted; A second generating module is used to adjust the plurality of linear boundary conditions based on the plurality of groups of nonlinear variation coefficient matrices to generate a plurality of nonlinear boundary conditions; A third generating module is used to apply the multiple nonlinear boundary conditions on the three-dimensional numerical model, and obtain the calculated value of the stress component of each preset measuring point through numerical simulation; The fourth generation module is used to perform regression analysis on multiple calculated values ​​of stress components and multiple measured values ​​of stress components to obtain the regression coefficient of each nonlinear boundary condition, weight them with the regression coefficient as the weight, and substitute them into the three-dimensional numerical model for calculation to obtain the geostress field of the three-dimensional numerical model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the nonlinear geostress field inversion method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the nonlinear geostress field inversion method according to any one of claims 1 to 7 are implemented.

Images