Layered surrounding rock three-dimensional crustal stress field inversion intelligent analysis system and method

Abstract

The invention relates to the technical field of layered surrounding rock inversion analysis, and discloses a layered surrounding rock three-dimensional crustal stress field inversion intelligent analysis system and method, and the system comprises a dimensional geological value module, an intelligent inversion analysis module and an inversion result verification module. The method comprises the steps of collecting geological data in target layered surrounding rock; carrying out normalization processing on the geological data, and carrying out sensitivity analysis; constructing a three-dimensional geologic model, and performing numerical simulation; a crustal stress field inversion intelligent analysis model is generated based on the three-dimensional geologic model; training the crustal stress field inversion intelligent analysis model by using the training sample; inputting the real-time geological data into the crustal stress field inversion intelligent analysis model to obtain a crustal stress inversion analysis result; carrying out visual output on the ground stress analysis result; and adjusting the crustal stress field inversion intelligent analysis model according to the actually measured data. Scientific ground stress field analysis basis can be provided for complex projects such as tunnel engineering, underground engineering and hydropower stations.

Description

Technical Field

[0001] The present invention relates to the technical field of layered surrounding rock inversion analysis, and particularly to a three-dimensional in-situ stress field inversion intelligent analysis system and method for layered surrounding rock. Background Art

[0002] In-situ stress field inversion is to establish a mathematical model through on-site measured data (such as displacement, stress, strain, etc.) to inversely deduce the initial in-situ stress state of underground rock masses. Under complex geological conditions, such as deep buried large fault tectonic areas or high in-situ stress areas, traditional linear superposition methods are difficult to accurately reflect the true distribution of the in-situ stress field. With the development of artificial intelligence technology, the application of intelligent algorithms in in-situ stress field inversion has gradually increased. For example, the in-situ stress field inversion method based on support vector machine (SVR) minimizes the structural risk of the feature vector, reducing the influence of random errors in measured data on parameter inversion. The background art of the three-dimensional in-situ stress field inversion intelligent analysis system and method for layered surrounding rock covers the application from basic principles to complex geological conditions, as well as the introduction of intelligent algorithms and the verification of engineering practices. The development of these technologies provides important theoretical and practical support for solving in-situ stress problems in geotechnical engineering.

[0003] Prior art one, a Chinese patent with the patent number: 202310538323.5 discloses an intelligent real-time rock layer inversion and identification method based on drilling parameters characteristics, which specifically includes conducting simulated drilling tests on different types of rock samples on an indoor drilling rig test platform, and collecting drilling parameters during rock block drilling by installing sensors on the drilling rig, performing time-frequency analysis and processing on the collected drill pipe axial vibration signals and training them to be used for an artificial neural network model for identifying rock layers; inputting the collected drilling data into the trained artificial neural network model at the drilling operation site, and the artificial neural network model outputs real-time identification results such as the interface and lithology type of the drilled rock layer to classify the formation geology and geotechnical body, and during the drilling process, the staff can adjust the drilling parameters and path in a timely manner according to the drilling rock layer identification results. Although it is beneficial to ensure the drilling quality, improve the drilling efficiency, and save a large amount of manpower and material resources; however, the inversion method has certain limitations, ignoring the non-linear characteristics of the in-situ stress field, being sensitive to human factors, and lacking uniqueness.

[0004] Prior Art II, a Chinese patent with the patent number 202411773937.2 discloses a method for detecting and analyzing the floor of a coal seam working face, which relates to the field of detection and analysis technology. By observing the potential changes at different positions and elevations with a parallel electrical method instrument and through three-dimensional electrical method inversion, the resistivity distribution at different depths within the working face and its floor can be obtained. These resistivity distribution conditions provide technical support for giving objective and accurate geological interpretations, helping to better understand the hydrogeological conditions of the floor rock strata of the working face, and thus providing an important reference basis for safe mining. Although the deviation of the measured conductivity value can be accurately corrected by evaluating and correcting the thermal effect and electromagnetic effect existing in the environment, improving the accuracy and reliability of the conductivity measurement data; then inputting the corrected conductivity value SR(xi, yi, zi) of each grid point into the three-dimensional model and annotating it to construct a three-dimensional conductivity model for visualizing the distribution information of the electrical properties of the underground medium; however, the inversion method has certain limitations, ignoring the non-linear characteristics of the in-situ stress field, being sensitive to human factors, and lacking uniqueness.

[0005] Prior Art III, a Chinese patent with the patent number 202411699733.9 discloses a lithofacies analysis system and method based on a shale geology-mechanics coupling lithofacies classification system, which relates to the field of shale lithofacies analysis. The system construction module determines the geology-mechanics coupling lithofacies classification system according to the information data of the data acquisition module; the prediction module determines the single-well lithofacies prediction information; the feedback adjustment module adjusts the single-well lithofacies prediction information according to the set range of rock mechanics and in-situ stress parameters to obtain the adjustment information; the three-dimensional model construction module uses the genetic algorithm to perform seismic attribute chimerism and integrated inversion based on the seismic attribute data and the adjustment information, and uses the co-kriging method to construct a three-dimensional model for geology-mechanics coupling lithofacies well-seismic collaborative modeling. Although the parameter model determination module uses a layer-controlled and facies-controlled double-control method to control and constrain, determining the heterogeneity of the distribution of shale layer geology-mechanics characteristics in three-dimensional space to provide a basis for the construction area for the exploration and development of artificial oil and gas reservoirs; however, the inversion method has certain limitations, ignoring the non-linear characteristics of the in-situ stress field, being sensitive to human factors, and lacking uniqueness.

[0006] Currently, Prior Art I, Prior Art II, and Prior Art III have the problem that the inversion method has certain limitations, ignoring the non-linear characteristics of the in-situ stress field, being sensitive to human factors, and lacking uniqueness. To solve the above problems, the present invention provides a three-dimensional in-situ stress field inversion intelligent analysis system and method for stratified surrounding rock. Summary of the Invention

[0007] The main object of the present invention is to provide a three-dimensional in-situ stress field inversion intelligent analysis system and method for layered surrounding rock, so as to solve the problems existing in the prior art that the inversion method has certain limitations, ignores the non-linear characteristics of the in-situ stress field, is sensitive to human factors, and lacks uniqueness.

[0008] To achieve the above object, the present invention provides the following technical solutions: A three-dimensional in-situ stress field inversion intelligent analysis system for layered surrounding rock, the three-dimensional in-situ stress field inversion intelligent analysis system for layered surrounding rock includes: A three-dimensional geological numerical module, which is used to collect geological data such as fault distribution, rock formation attitude, physical and mechanical parameters, and in-situ stress measured data in the target layered surrounding rock; normalize the geological data and perform sensitivity analysis; construct a three-dimensional geological model and perform numerical simulation; An intelligent inversion analysis module, which is used to generate training samples by orthogonal experimental design parameter combinations and combine with the numerical simulation results; generate a three-dimensional in-situ stress field inversion intelligent analysis model based on the three-dimensional geological model; use the training samples to train the three-dimensional in-situ stress field inversion intelligent analysis model; An inversion result verification module, which is used to input real-time geological data into the three-dimensional in-situ stress field inversion intelligent analysis model to obtain the in-situ stress inversion analysis result; visually output the in-situ stress analysis result; and adjust the three-dimensional in-situ stress field inversion intelligent analysis model according to the measured data.

[0009] As a further improvement of the present invention, the three-dimensional geological numerical module includes: A rock formation grading sub-module, which is used to import drilling data, geological exploration data, geophysical data, and measured in-situ stress data, and define the spatial range of the three-dimensional geological model; determine the range of rock mass mechanical parameters and in-situ stress parameters in different regions through geological zoning; A shear extrusion action sub-module, which is used to perform three-dimensional mesh division to generate regular tetrahedral elements; process complex fault regions, allocate heterogeneous rock mechanical parameters to grid cells; and apply normal stress on the top surface of the three-dimensional geological model to simulate the self-weight stress field, and obtain the shear extrusion action reflecting the fault structure; A three-dimensional geological model adjustment sub-module, which is used to comprehensively characterize the non-linear superposition effect of the in-situ stress field through the piezometric coefficient and the rock mass volume force; compare the measured data and the simulated values of the in-situ stress components at the measuring points, and dynamically adjust the three-dimensional geological model by combining the comparison results with the real-time geological data.

[0010] As a further improvement of the present invention, the rock formation grading sub-module includes: Classification and sorting unit, which is used to import drilling data, geological survey data and geophysical data, divide the strata through borehole data, uniformly number and sort them, classify them into the same rock stratum unit according to the lithology pixel rows and vertical positions, and process the influence of faults on the continuity of the underlying strata; Fault trace unit, which is used to determine the fault dip and extension depth based on the fault distribution and rock stratum occurrence information in the geological survey data, combined with the geophysical inversion results, and generate fault trace lines; Define spatial range unit, which is used to correct the rock mass morphology using borehole data, combine seismic interpretation horizons with velocity / density data, divide different lithology units, fuse the fault geometric relationship with the stratum boundary data, and define the spatial range of the 3D geological model.

[0011] As a further improvement of the present invention, the shear extrusion action sub-module includes: Mesh division unit, which is used to perform mesh dissection on the 3D geological model; process the fault intersection relationship in the complex fault area; locally encrypt the mesh of the fault zone; and assign different rock mass mechanical parameters to the corresponding mesh units according to the fault type and measured data; Normal constraint unit, which is used to apply normal stress boundary conditions to the top surface of the 3D geological model, simulate the self-weight stress field of the rock mass by calculating the self-weight generated blowing response stress through the rock mass density; and apply normal constraints to the ground and lateral boundaries of the 3D geological model; Stress direction unit applies horizontal tectonic stress to the lateral boundary of the 3D geological model, simulates the horizontal tectonic stress component by applying a normal unit load to the lateral boundary, combines the piezometric coefficient to define the proportional relationship between the horizontal stress and the vertical stress; determines the tectonic stress direction by inverting the measured in-situ stress data.

[0012] As a further improvement of the present invention, the intelligent inversion analysis module includes: Form sample set sub-module, which is used to select parameter combinations that affect the in-situ stress field and perform numerical simulations using the 3D geological model; bring the parameter combinations into the 3D geological model, calculate the distribution results of the in-situ stress field under different working conditions; and form a training sample set based on the simulation results and parameter combinations; Generate inversion analysis model sub-module, which is used to generate an intelligent analysis model for in-situ stress field inversion based on the 3D geological model; determine the center position of the radial basis function through self-organizing learning, and establish the mapping relationship between the input parameters and the output in-situ stress field; screen key parameters and perform non-linear fitting; Optimize inversion analysis model sub-module, which is used to input the generated training samples into the intelligent analysis model for in-situ stress field inversion, adjust the fitting coefficient until it meets the preset value, then compare the measured value with the inversion value to optimize the intelligent analysis model for in-situ stress field inversion, and obtain the final inversion intelligent analysis model.

[0013] As a further improvement of the present invention, a sample set forming sub-module is provided, including: A stress field distribution unit, configured to select key parameters affecting the in-situ stress field in a three-dimensional geological model, determine a parameter combination scheme; input the selected parameter combination into the three-dimensional geological model for numerical simulation calculation, and obtain the in-situ stress field distribution results under different working conditions; A parameter combination unit, configured to extract the in-situ stress components and distribution laws of key measurement points from the three-dimensional geological model after each numerical simulation, and record the corresponding parameter combination; use the parameter combination as an input variable and the corresponding simulated in-situ stress field result as an output variable; A storage training sample set unit, configured to combine the input variable and the output variable into parameter-in-situ stress field paired data; eliminate abnormal data due to unqualified parameters, and optimize the sample through iterative optimization; store the verified parameter combination and the simulated in-situ stress field data in a preset format as a training sample set.

[0014] As a further improvement of the present invention, an inversion analysis model generating sub-module is provided, including: A weight setting unit, configured to generate an in-situ stress field inversion intelligent analysis model based on the three-dimensional geological model; calculate the stress contribution values of each parameter combination under the simulated working conditions, and determine the influence weight of the sensitivity of the in-situ stress field distribution on the output result by quantifying the sensitivity of each parameter to the in-situ stress field distribution; An equilibrium relationship unit, configured to sort the calculated stress contribution values according to the magnitude, and compare them with a pre-set error threshold; adjust the parameter combination through repeated iteration, perform non-linear fitting on the retained key parameters; evaluate the overall inversion error for each iteration; introduce a weight parameter to balance the relationship between data-driven and physical constraints; An iterative optimization unit, configured to establish a mapping relationship between the input parameters and the in-situ stress field output, determine the function center position through self-organizing learning, and adjust the fitting coefficient based on the selected key parameter set until the inversion intelligent analysis model meets the three-dimensional distribution feature standard of the preset in-situ stress field.

[0015] As a further improvement of the present invention, the equilibrium relationship unit includes: A weight assignment sub-unit, configured to obtain a data-driven term based on the sum of squares of the errors between the numerical simulation calculation value and the measured value, and use the historical strata surrounding rock data as a physical constraint term; assign different weights to the data-driven term and the physical constraint term; A weight optimization sub-unit, configured to dynamically adjust the weight parameter according to the comparison between the current inversion error and the preset threshold during the iteration process; screen out key parameters and assign higher weights by quantifying the sensitivity of the parameters to the in-situ stress field; eliminate non-key parameters; A parameter combination and weight configuration subunit, which is used to combine the prior distribution of physical constraints with the data-driven likelihood function to obtain the posterior probability; optimize the weight parameters based on the posterior probability; iterate sequentially until a preset tolerance is reached or the weight change reaches a preset trend, terminate the iteration, and output the optimized parameter combination and weight configuration.

[0016] As a further improvement of the present invention, an inversion result verification module includes: A forward verification sub-module, which is used to communicate the stress field inversion intelligent analysis model with the three-dimensional geological model, input the real-time geological data into the in-situ stress field inversion intelligent analysis model for simulation, and substitute the output parameters of the in-situ stress field inversion intelligent analysis model into the three-dimensional geological model for forward analysis and calculation; A visualization sub-module, which is used to generate the in-situ stress field distribution based on the characteristics of the terrain, faults, and rock layer structures of the three-dimensional geological model, as well as the applied self-weight pressure and tectonic stress boundary conditions; visually output the calculated in-situ stress components through a contour map; An adjustment and optimization sub-module, which is used to compare the calculated in-situ stress components with the real-time measured data. If it exceeds the preset range, it triggers an instruction to adjust the model; if a deviation is found during verification, it optimizes the weights of the stress field intelligent analysis model or the parameters of the three-dimensional geological model, iteratively updates the inversion result until the preset accuracy value is met.

[0017] To achieve the above object, the present invention also provides the following technical solutions: A three-dimensional in-situ stress field inversion intelligent analysis method for layered surrounding rock, which is applied to the three-dimensional in-situ stress field inversion intelligent analysis system for layered surrounding rock. The three-dimensional in-situ stress field inversion intelligent analysis method for layered surrounding rock includes: Collect geological data such as fault distribution, rock layer occurrence, physical and mechanical parameters, and in-situ stress measured data in the target layered surrounding rock; normalize the geological data and perform sensitivity analysis; construct a three-dimensional geological model and perform numerical simulation; The intelligent inversion institute designs parameter combinations through orthogonal experiments, generates training samples in combination with the numerical simulation results; generates an in-situ stress field inversion intelligent analysis model based on the three-dimensional geological model; uses the training samples to train the in-situ stress field inversion intelligent analysis model; Input the real-time geological data into the in-situ stress field inversion intelligent analysis model to obtain the in-situ stress inversion analysis result; visually output the in-situ stress analysis result; and adjust the in-situ stress field inversion intelligent analysis model according to the measured data.

[0018] The present invention collects geological information such as fault distribution, rock stratum occurrence, physical and mechanical parameters, and measured in-situ stress data in the target layered surrounding rock through a three-dimensional geological numerical module, normalizes these data, and constructs an accurate three-dimensional geological model in combination with sensitivity analysis. Through the intelligent inversion module, combined with the orthogonal experimental design parameter combination, training samples are generated by numerical simulation, and an intelligent analysis model for in-situ stress field inversion is constructed based on the three-dimensional geological model. Through the inversion result verification module, real-time geological data is input into the intelligent analysis model to obtain the in-situ stress inversion analysis result, and the result is displayed through visual output. By combining the three-dimensional numerical model and the intelligent inversion method, the accuracy of in-situ stress field inversion can be effectively improved. Especially in complex geological conditions such as deep structural areas or tunnel engineering, the distribution characteristics of the in-situ stress field can be more accurately reflected. Through visualization technology, the inversion result is intuitively displayed, facilitating engineers and researchers to understand the distribution law of the in-situ stress field. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the functional modules of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 2 It is a schematic diagram of the functional modules of the three-dimensional geological numerical module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 3 It is a schematic diagram of the functional modules of the rock stratum grading sub-module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 4 It is a schematic diagram of the functional modules of the shear extrusion action sub-module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 5 It is a schematic diagram of the functional modules of the intelligent inversion analysis module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 6 It is a schematic diagram of the functional modules of the sample set formation sub-module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 7 It is a schematic diagram of the functional modules of the inversion analysis model generation sub-module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 8 It is a schematic diagram of the functional modules of the balance relationship unit of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 9 It is a schematic diagram of the functional modules of the inversion result verification module of an embodiment of the intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock of the present invention; Figure 10Schematic diagram of the steps of an embodiment of the intelligent analysis method for inverting the three-dimensional in-situ stress field of stratified surrounding rock according to the present invention; Figure 11 Rock mass slope map of an embodiment of the intelligent analysis method for inverting the three-dimensional in-situ stress field of stratified surrounding rock according to the present invention; Figure 12 Schematic diagram of the structure of an embodiment of the electronic device according to the present invention; Figure 13 Schematic diagram of the structure of an embodiment of the storage medium according to the present invention. Detailed implementation manners

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0021] The terms "first", "second", and "third" in the present invention are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first", "second", and "third" may explicitly or implicitly include at least one of such features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship and movement conditions between components in a specific posture (as shown in the drawings). If the specific posture changes, the directional indications will also change accordingly. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0022] Referring to "embodiment" herein means that a specific feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0023] Such as Figure 1As shown in the figure, this embodiment provides an embodiment of an intelligent analysis system for inverse analysis of the three-dimensional in-situ stress field of stratified surrounding rock. In this embodiment, the intelligent analysis system for inverse analysis of the three-dimensional in-situ stress field of stratified surrounding rock specifically includes: A three-dimensional geological numerical module 1, configured to collect geological data such as fault distribution, rock stratum attitude, physical and mechanical parameters, and measured in-situ stress data in the target stratified surrounding rock; perform normalization processing on the geological data and conduct sensitivity analysis; construct a three-dimensional geological model and perform numerical simulation; An intelligent inverse analysis module 2, configured to generate training samples by combining the parameter combinations designed by orthogonal experiments through intelligent inversion and combining with the numerical simulation results; generate an intelligent analysis model for inverse analysis of the in-situ stress field based on the three-dimensional geological model; use the training samples to train the intelligent analysis model for inverse analysis of the in-situ stress field; An inverse result verification module 3, configured to input real-time geological data into the intelligent analysis model for inverse analysis of the in-situ stress field to obtain an in-situ stress inverse analysis result; visually output the in-situ stress analysis result; and adjust the intelligent analysis model for inverse analysis of the in-situ stress field according to the measured data.

[0024] Preferably, in this embodiment, through the three-dimensional geological numerical module, geological information such as fault distribution, rock stratum attitude, physical and mechanical parameters, and measured in-situ stress data in the target stratified surrounding rock is collected, these data are subjected to normalization processing, and combined with sensitivity analysis, an accurate three-dimensional geological model is constructed. Through the intelligent inversion module, by combining the parameter combinations designed by orthogonal experiments, training samples are generated using numerical simulation, and an intelligent analysis model for inverse analysis of the in-situ stress field is constructed based on the three-dimensional geological model. Through the inverse result verification module, real-time geological data is input into the intelligent analysis model to obtain an in-situ stress inverse analysis result, and the result is displayed through visual output. By combining the three-dimensional numerical model and the intelligent inversion method, the accuracy of the in-situ stress field inversion can be effectively improved. Especially in complex geological conditions, such as deep structural areas or tunnel projects, the distribution characteristics of the in-situ stress field can be more accurately reflected. Through visualization technology, the inversion results are intuitively displayed, which is convenient for engineers and researchers to understand the distribution law of the in-situ stress field. At the same time, through real-time data input and model adjustment, the model can be dynamically optimized to make it more in line with the actual engineering requirements, thereby improving the safety of engineering design and construction. It can provide a scientific basis for in-situ stress field analysis for complex projects such as tunnel projects, underground projects, and hydropower stations, help optimize the design and construction plans, reduce engineering risks, and improve economic benefits.

[0025] Furthermore, as Figure 2 shown, the three-dimensional geological numerical module 1 specifically includes: The rock stratum classification sub-module 11 is used to import drilling data, geological exploration data, geophysical data, and measured in-situ stress data, and define the spatial range of the three-dimensional geological model; determine the range of rock mass mechanical parameters and in-situ stress parameters in different regions through geological zoning; Among them, the geological exploration data includes fault distribution and rock stratum occurrence; the geophysical data includes seismic interpretation horizons and velocity / density information; the spatial range includes formation boundaries and fault geometric relationships; The shear extrusion action sub-module 12 is used to perform three-dimensional mesh division to generate regular tetrahedral elements; process complex fault regions and assign heterogeneous rock mechanical parameters to the mesh elements; and apply a normal stress on the top surface of the three-dimensional geological model to simulate the self-weight stress field, and obtain the shear extrusion action reflecting the fault structure; The three-dimensional geological model adjustment sub-module 13 is used to comprehensively characterize the non-linear superposition effect of the in-situ stress field through the pressure coefficient and the rock mass volume force; compare the measured data and the simulated values of the in-situ stress components at the measurement points, and dynamically adjust the three-dimensional geological model by combining the comparison results with real-time geological data.

[0026] Preferably, in this embodiment, by importing drilling data, geological exploration data, geophysical data, and measured in-situ stress data, the integration and fusion of multi-source heterogeneous data are realized. The range of rock mass mechanical parameters and in-situ stress parameters in different regions is determined through geological zoning, combined with the three-dimensional mesh division technology to generate regular tetrahedral elements, and the complex fault regions are processed, and the heterogeneous rock mechanical parameters are assigned to the mesh elements. The three-dimensional geological model adjustment module 13 comprehensively characterizes the non-linear superposition effect of the in-situ stress field through the pressure coefficient and the rock mass volume force, and dynamically adjusts the three-dimensional geological model by combining the comparison results of the measured data and the simulated values. A normal stress is applied on the top surface of the three-dimensional geological model to simulate the self-weight stress field and reflect the shear extrusion action of the fault structure. Through the integration and dynamic adjustment of multi-source data, the three-dimensional geological model can more realistically reflect the spatial distribution and mechanical properties of geological bodies, thereby improving the modeling accuracy. The three-dimensional geological model can meet the engineering requirements under complex geological conditions, such as subway site selection, engineering construction, etc. Using the automated mesh division technology and the heterogeneous rock mechanical parameter assignment method reduces manual intervention and improves the modeling efficiency.

[0027] Furthermore, as Figure 3 shown, the rock stratum classification sub-module 11 specifically includes: The classification and sorting unit 111 is used to import drilling data, geological survey data, and geophysical data, divide the strata through borehole data, uniformly number and sort them, classify them into the same rock stratum unit according to the lithology pixel row and vertical position, and process the influence of faults on the bottom layer continuity; Fault trace unit 112, which is used to determine the fault dip and extension depth based on the fault distribution and rock formation attitude information in geological survey data, and combine with the results of geophysical inversion to generate fault trace lines; Spatial range definition unit 113, which is used to correct the rock mass morphology using borehole data, combine seismic interpretation horizons with velocity / density data, divide different lithologic units, and fuse the fault geometric relationship with stratigraphic boundary data to define the spatial range of the three-dimensional geological model.

[0028] Preferably, in this embodiment, the strata are divided, uniformly numbered and sorted through borehole data, and the strata are classified into the same rock formation unit according to the lithologic pixels and vertical positions. Based on the fault distribution and rock formation attitude information in geological survey data, and combining with the results of geophysical inversion, the fault dip and extension depth are determined to generate fault trace lines. The rock mass morphology is corrected using borehole data, different lithologic units are divided by combining seismic interpretation horizons with velocity / density data, and the fault geometric relationship is fused with stratigraphic boundary data to define the spatial range of the three-dimensional geological model. Through the preprocessing and standardization operations of the classification and sorting unit, the original data is converted into structured information, which is convenient for subsequent analysis and modeling. The fault trace unit generates accurate fault trace lines by combining geological survey data and geophysical inversion results, thereby improving the accuracy and reliability of the three-dimensional geological model. The spatial range definition unit constructs a three-dimensional geological model containing complex geological structures by fusing seismic interpretation horizons with velocity / density data and the fault geometric relationship. The overall technology provides a more intuitive and comprehensive perspective for geological analysis through three-dimensional modeling and data fusion.

[0029] Furthermore, as Figure 4 shown, the shear extrusion sub-module 12 specifically includes: Mesh division unit 121, which is used to perform mesh dissection on the three-dimensional geological model; process the fault intersection relationship in the complex fault area; locally encrypt the mesh in the fault zone; and assign different rock mass mechanical parameters to the corresponding mesh elements according to the fault type and measured data; Normal constraint unit 122, which is used to apply normal stress boundary conditions to the top surface of the three-dimensional geological model, simulate the self-weight stress field of the rock mass by calculating the self-weight generated blowing response stress through the rock mass density; and apply normal constraints to the ground and lateral boundaries of the three-dimensional geological model; Stress direction unit 123 applies horizontal tectonic stress to the lateral boundary of the three-dimensional geological model, simulates the horizontal tectonic stress component by applying a normal unit load to the lateral boundary, combines the piezometric coefficient to define the proportional relationship between the horizontal stress and the vertical stress; determines the tectonic stress direction by inverting the measured in-situ stress data.

[0030] Preferably, in this embodiment, through the grid division of the three-dimensional geological model, the complex geological body is divided into multiple grid units to achieve accurate modeling of the geological structure. In the complex fault area, the local grid encryption technology is adopted to improve the resolution of the fault zone, so as to more accurately simulate the influence of the fault on the geological stress field; according to the fault type and measured data, different rock mass mechanical parameters (such as elastic modulus, Poisson's ratio, etc.) are assigned to the corresponding grid units to reflect the mechanical properties of different geological bodies; by optimizing the grid division method (such as tetrahedron optimization division), the accuracy and calculation efficiency of the model are improved; the normal stress boundary condition is applied to the top surface of the three-dimensional geological model, and the self-weight-induced stress is calculated through the rock mass density to simulate the self-weight stress field of the rock mass; the horizontal tectonic stress is applied to the lateral boundary, and the proportional relationship between the horizontal stress and the vertical stress is defined in combination with the coefficient of lateral pressure, and the direction of the tectonic stress is determined by inverting the measured in-situ stress data. Through the fine grid division and local encryption technology, the structure and stress distribution of the complex geological body can be more accurately simulated, and the calculation error can be reduced; the optimized grid division method and the automated modeling process significantly improve the modeling efficiency, reduce manual intervention, and reduce the calculation cost. Through the three-dimensional visualization technology, the geological model and the stress field distribution can be intuitively displayed, which is convenient for engineering design and risk assessment; it can effectively deal with engineering problems under complex geological conditions, such as fault influence, self-weight stress field simulation, etc., and provide a scientific basis for the design and construction of projects such as tunnels and mines.

[0031] Further, as Figure 5 shown, the intelligent inversion analysis module 2 specifically includes: A sample set formation sub-module 21, which is used to select a parameter combination that affects the in-situ stress field and perform numerical simulation using the three-dimensional geological model; bring the parameter combination into the three-dimensional geological model, calculate the distribution results of the in-situ stress field under different working conditions; and form a training sample set based on the simulation results and the parameter combination; An inversion analysis model generation sub-module 22, which is used to generate an intelligent analysis model for in-situ stress field inversion based on the three-dimensional geological model; determine the center position of the radial basis function through self-organizing learning, and establish the mapping relationship between the input parameters and the output in-situ stress field; screen the key parameters and perform non-linear fitting; An inversion analysis model optimization sub-module 23, which is used to input the generated training samples into the intelligent analysis model for in-situ stress field inversion, adjust the fitting coefficient until it meets the preset value, then compare the measured value with the inverted value to optimize the intelligent analysis model for in-situ stress field inversion, and obtain the final inversion intelligent analysis model.

[0032] Preferably, in this embodiment, a parameter combination affecting the in-situ stress field is selected, and numerical simulation is carried out using a three-dimensional geological model; the parameter combination is brought into the three-dimensional geological model to calculate the distribution results of the in-situ stress field under different working conditions, and a training sample set is formed based on the simulation results and the parameter combination. The center position of the radial basis function is determined through self-organizing learning, the mapping relationship between the input parameters and the output in-situ stress field is established, and the key parameters are screened for non-linear fitting. The generated training samples are input into the in-situ stress field inversion intelligent analysis model, and the fitting coefficient is adjusted until the preset value is met. At the same time, the model is optimized by comparing the measured values with the inversion values, and finally an accurate in-situ stress field inversion intelligent analysis model is obtained. By combining the three-dimensional geological model and the intelligent algorithm, the spatial distribution law of the in-situ stress field can be more accurately reflected, and the inversion accuracy can be significantly improved. Through the accurate in-situ stress field inversion results, a reliable basis can be provided for the design and construction of underground engineering, and the safety hazards caused by inaccurate stress field prediction can be reduced; it is particularly suitable for the in-situ stress field inversion under complex geological conditions, such as scenarios like deep-buried tunnels and hydropower station projects, and can effectively cope with the challenges brought by factors such as geological structures and rock mass properties; by introducing the intelligent algorithm, the complex mechanical models and heavy computational workload in traditional methods are reduced, and the computational efficiency is improved.

[0033] Further, as Figure 6 shown, the sample set forming sub-module 21 specifically includes: The stress field distribution unit 211 is used to select the key parameters affecting the in-situ stress field in the three-dimensional geological model, determine the parameter combination scheme; input the selected parameter combination into the three-dimensional geological model for numerical simulation calculation, and obtain the distribution results of the in-situ stress field under different working conditions; The parameter combination unit 212 is used to extract the in-situ stress components and distribution laws of the key measurement points from the three-dimensional geological model after each numerical simulation, and record the corresponding parameter combination; use the parameter combination as the input variable and the corresponding simulated in-situ stress field result as the output variable; The training sample set storage unit 213 is used to combine the input variable and the output variable into parameter in-situ stress field paired data; eliminate the abnormal data due to unqualified parameters, and optimize the sample through iterative optimization; store the verified parameter combination and the simulated in-situ stress field data in a preset format as the training sample set.

[0034] Preferably, this embodiment obtains the distribution results of the geostress field under different working conditions by screening key parameters in the three-dimensional geological model, determining the parameter combination scheme, and inputting these parameters into the model for numerical simulation calculation. After each numerical simulation is completed, the geostress components and their distribution laws of key measuring points are extracted from the three-dimensional geological model, and the corresponding parameter combinations are recorded. By combining the input variables and output variables into paired data, removing unqualified abnormal data, optimizing the samples, and finally storing them as training sample sets in a preset format. By optimizing the parameter combination and iterative optimization, the model can more accurately reflect the distribution law of the geostress field, solving the problem of insufficient fitting accuracy of the local stress abnormal area in the traditional model; by storing and optimizing samples, the model can better adapt to the distribution of the geostress field under different working conditions, improve the generalization ability of the model, and make it have higher practical value in actual engineering applications; through the automated process, manual intervention is reduced, and the efficiency and reliability of the model are improved. At the same time, through iterative optimization, the accuracy and stability of the model are further improved; by combining three-dimensional geological modeling with numerical simulation technology, the efficient coupling of the geological model and the stress field is realized, providing a new solution for the stress field analysis under complex geological conditions.

[0035] Furthermore, if Figure 7 As shown, the generating inversion analysis model submodule 22 specifically includes: The weight setting unit 221 is used to generate an intelligent analysis model for geostress field inversion based on the three-dimensional geological model; calculate the stress contribution value of each parameter combination under the simulated working condition, and determine the influence weight of the sensitivity of the geostress field distribution on the output result by quantifying the sensitivity of each parameter to the geostress field distribution; Among them, the weight adaptation kernel function of the weight is:

[0036] In the formula, For the k Combination of parameters in rock mass Measuring point j The stress contribution density tensor of ; Characterization m Curvature correction factor of the geological structure (calculated by the three-dimensional geological model); For measuring point j The principal value vector of the triaxial stress; Corresponding rock layer k Anisotropic damping factor of For the n Rock mass heterogeneity coefficient ν The gamma distribution probability correction term; Characterize the parallel / orthogonal axial strain coupling strength caused by tectonic movement; Represents the parameter vector Covariant gradient interaction with the low-dimensional latent space ; Is the weighted coefficient of the directional penetration measure; a weighted tensor integral equation that integrates the cross-theory of geomechanics, probability statistics, and machine learning. Its theoretical basis and functions can be analyzed layer by layer as follows: Geological structure curvature correction factor Originates from the curvature-stress coupling theory in continuum mechanics, corrects the secondary stress field generated by the bending of rock strata through the m-order curvature tensor, quantifies the topological influence of geological folds / faults and other structural forms on the in-situ stress field, and when m = 2, it corresponds to the stress concentration effect of double-curvature structures. Gamma distribution correction term Adopts the Bayesian rock mass parameter inversion framework, describes the prior probability of rock mass inhomogeneous parameters through the gamma distribution; introduces probability weights into the integral kernel, and automatically reduces its contribution when a certain type of rock mass parameter deviates from the typical value. Strain coupling strength function Based on the anisotropic constitutive relationship of porous media, where correspond to the coaxial / non-coaxial strains generated by tectonic movements respectively; depicts the stress principal axis rotation effect caused by tectonic movements such as thrust faults.

[0037] Gradient interaction term Realizes the non-linear dimensionality reduction from the parameter space to the latent space, and ensures the maintenance of geomechanical constraints during parameter optimization through the covariant gradient; when the adjustment direction of a certain parameter is orthogonal to the mechanical response trend of the latent space , the penalty term is automatically increased. Directional penetration measure Originates from the tensorization of conductivity in seepage theory, and is used here to control the directionality of stress propagation; takes =cos²φ (φ is the normal angle of the bedding plane) in sedimentary rock strata, reflecting the preferential conduction of stress along the bedding plane.

[0038] Weight dynamic allocation mechanism, realizes the "localized" calculation of parameter contributions through the exponential decay characteristics of the integral kernel, and avoids the boundary effect caused by traditional global weighting. Implicit in the multi-physical field coupling analysis formula - - Coupling relationship, completely describes the closed-loop feedback of "tectonic stress (σ) - lithology damping (τ) - strain history ( )". The model achieves a measured coincidence degree in the inversion of in-situ stress in deep coal mines. Its innovation lies in the organic integration of geological cognition ( , term) and data-driven (Γ, term) through tensor operations, and has better cross-scale adaptability compared with traditional inversion methods; A balance relationship unit 222 is used to sort the calculated stress contribution values according to their magnitudes and compare them with a pre-set error threshold; by repeatedly iterating and adjusting the parameter combinations, the key parameters are retained for non-linear fitting; the overall inversion error is evaluated in each iteration; a weight parameter is introduced to balance the relationship between data-driven and physical constraints; An iterative optimization unit 223 is used to establish a mapping relationship between the input parameters and the output of the in-situ stress field, determine the function center position through self-organizing learning, and adjust the fitting coefficients based on the selected key parameter set until the inversion intelligent analysis model meets the preset three-dimensional distribution characteristic standard of the in-situ stress field.

[0039] Preferably, in this embodiment, an in-situ stress field inversion intelligent analysis model is generated based on a three-dimensional geological model, the stress contribution values of each parameter combination under the simulation conditions are calculated, the sensitivity of each parameter to the distribution of the in-situ stress field is quantified to determine its influence weight on the output result; the calculated stress contribution values are sorted according to their magnitudes and compared with a pre-set error threshold; by repeatedly iterating and adjusting the parameter combinations, the key parameters are retained for non-linear fitting; a weight parameter is introduced to balance the relationship between data-driven and physical constraints; a mapping relationship between the input parameters and the output of the in-situ stress field is established; the function center position is determined through self-organizing learning; the fitting coefficients are adjusted based on the selected key parameter set until the inversion intelligent analysis model meets the preset three-dimensional distribution characteristic standard of the in-situ stress field; through sensitivity analysis and weight adjustment, the influence of each parameter on the distribution of the in-situ stress field can be more accurately reflected, thereby improving the accuracy of the inversion result; combined with iterative optimization and non-linear fitting, the adaptability of the model to complex geological conditions is further enhanced; by introducing a weight parameter and balancing the relationship between data-driven and physical constraints, the robustness of the model in the face of sparse data or uncertainties is enhanced; the iterative optimization unit ensures that the model can maintain a high fitting accuracy under different working conditions by screening key parameters and adjusting fitting coefficients; it can provide an efficient and reliable solution for the inversion of the in-situ stress field under complex geological conditions, and has important theoretical significance and practical engineering application value; through the combination of the three-dimensional geological model and the intelligent analysis model, it can provide a scientific basis for engineering design and construction and reduce engineering risks. Through self-organizing learning and non-linear fitting, the amount of calculation is reduced and the inversion efficiency is improved; the iterative optimization unit avoids unnecessary waste of computing resources by screening key parameters, further improving the computing efficiency.

[0040] Furthermore, as Figure 8 shown, the balance relationship unit 222 specifically includes: A weight assignment subunit 2221 is used to obtain a data-driven term based on the sum of the squares of the errors between the numerical simulation calculated values and the measured values, and use the historical strata surrounding rock data as a physical constraint term; different weights are assigned to the data-driven term and the physical constraint term; Among them, the constraint state phase space projection of the physical constraint term :

[0041] In the formula, Λ = { , , , , } is a hyperparameter composite set; represents the probability density inverse diffusion mapping in the η -dimensional non-Euclidean space; characterizes the p -th type of constraint kernel energy of the historical lithology database; The catalog operation is the attribute quotient on the anisotropic stress manifold; is the physical constraint elevation angle; is the data-driven depression angle; represents the parameter vector to be optimized; represents corresponding predicted strain tensor residual; represents the matrix trace operation; represents the error tolerance exponential inflation factor; represents the p -th type of constraint weight coefficient of the historical lithology database; represents the attribute quotient operation; represents the projection operator; represents the forward covariant gradient operator; represents the backward covariant gradient operator; represents the cotangent of the physical constraint elevation angle; represents the tangent of the data-driven depression angle; The weight optimization subunit 2222 is used to dynamically adjust the weight parameters according to the comparison between the current inversion error and the preset threshold during the iteration process; by quantifying the sensitivity of the in-situ stress field to the parameters, the key parameters are screened out and given higher weights; the non-key parameters are eliminated; The parameter combination and weight configuration subunit 2223 is used to combine the prior distribution of the physical constraint with the likelihood function of the data-driven to obtain the posterior probability; optimize the weight parameters based on the posterior probability; iterate sequentially until the preset tolerance is reached or the weight change reaches the preset trend, terminate the iteration, and output the optimized parameter combination and weight configuration.

[0042] Among them, the mixed variable value configuration of the convergence criterion of the parameter combination and weight configuration subunit 2223 :

[0043] In the formula, ∈ Θ is the feasible parameter semi-space; Define the temperature parameter of the inverse diffusion process; [·] is the characteristic indicator functional of the event measure q ; Corresponding impact over-relaxed Péclet number; Characterization pattern q Hyperbolic mass deficit differential of; Is the tensorial fusion of non-connection addition and shaping rings; Is the asymptotic critical value of the steady-state phase transition; Denotes the feasible parameter vector; Denotes the weight coefficient of event q; Denotes Adjoint operator of; Denotes the regularization mapping function; Denotes Inverse diffusion operator controlled by; Denotes the optimal weight configuration tensor; Denotes the sign function; Denotes the phase transition index of path s; Denotes the inverse event measure geometry; The mathematical framework of the balance relation unit 222 demonstrates the cutting-edge methodology of data-physics fusion inversion, the core of which lies in establishing a variational assimilation framework for dynamic weight optimization.

[0044] Constrained state phase space projection Deep mechanism of, non-Euclidean space inverse diffusion mapping Originates from the heat kernel diffusion inversion in Riemannian geometry, the reverse evolution of the probability density on the η-dimensional manifold; when η = 2.3 (empirical value of shale reservoir), it can effectively suppress the anisotropic noise in borehole data; nuclear energy constraint term Based on the rock mass memory effect theory, where Characterizes the current parameter And historical data Similarity of.

[0045] Angle tensor product , construct the symplectic dual space of physical constraint ( ∈[0, π / 2]) and data-driven ( ∈[π / 2, π]); dynamic balance: when ‖ - ‖ < π / 6, the constraint relaxation mechanism is automatically triggered.

[0046] Innovative characteristics of the mixed variational value configuration Impact over-relaxed , fusing the fluid mechanics Péclet number and the relaxation time of non-equilibrium statistical physics; When the parameter update is greater than 1, it enters the shock iteration mode. Hyperbolic mass deficit , reflecting the parameter space and the geodesic deviation of the physical manifold ▽ . When > 0.1, it will automatically trigger parameter reorganization. The shaping ring fusion (⊕F), a cyclic cohomology operation in non-commutative geometry, completes the tensor chain derivative under GPU acceleration.

[0047] The weight dynamic game mechanism, physical constraint weights; the topological criterion for parameter screening, using persistent homology to analyze the importance of parameters; the phase transition detection of the convergence criterion, capturing the topological mutation of the parameter space through sgn[ζ(s)-ψ(∞)]. The breakthrough lies in transforming geological prior knowledge into a computable topological constraint manifold while maintaining the data-driven adaptive characteristics.

[0048] Preferably, in this embodiment, the sum of the squares of the errors between the numerically simulated values and the measured values is calculated to obtain the data-driven term, and the historical strata surrounding rock data is used as the physical constraint term. Different weights are assigned to the data-driven term and the physical constraint term, reflecting the dynamic balance of information from different sources and avoiding the limitations of information from a single source; during the iteration process, the weight parameters are dynamically adjusted according to the comparison between the current inversion error and the preset threshold. By quantifying the sensitivity of the parameters to the in-situ stress field, the key parameters are screened out and given higher weights, while the non-key parameters are eliminated. The prior distribution of the physical constraint is combined with the data-driven likelihood function to obtain the posterior probability. The weight parameters are optimized based on the posterior probability, and the optimal solution is gradually approximated through iteration. The iteration is terminated by a preset tolerance or the trend of weight change to ensure that the model converges to a stable state and avoid overfitting or underfitting problems. The overall process embodies the idea of multi-objective optimization, that is, considering the balance between data-driven and physical constraints simultaneously. By iteratively adjusting the weight parameters, the model performance is gradually optimized. The method of dynamically adjusting the weights can automatically optimize the weights according to the error distribution, improving the adaptability of the model to complex problems. By dynamically adjusting the weight parameters, the model can better fit the actual observed data, reduce errors, and improve the prediction accuracy; combining data-driven and physical constraints can make full use of information from different sources, avoid the bias of a single source, and thus improve the overall reliability of the model; dynamic weight optimization and sensitivity analysis can identify and eliminate non-key parameters, reduce the computational complexity, and at the same time improve the robustness of the model to outliers; through iterative optimization and posterior probability analysis, the model can maintain stable performance under different conditions and avoid overfitting or underfitting problems. Through sensitivity analysis and dynamic weight adjustment, the model can quickly identify key parameters, reduce unnecessary computational volume, and improve computational efficiency. During the iterative optimization process, the iteration is terminated by a preset tolerance or the trend of weight change to avoid wasting unnecessary computational resources; the method of dynamically adjusting the weights can flexibly adjust the model parameters according to the actual observed data and prior knowledge to adapt to different application scenarios.

[0049] Furthermore, as Figure 9 shown, the inversion result verification module 3 specifically includes: A forward verification sub-module 31, which is used to communicate the stress field inversion intelligent analysis model with the three-dimensional geological model, input real-time geological data into the in-situ stress field inversion intelligent analysis model for simulation, and substitute the output parameters of the in-situ stress field inversion intelligent analysis model into the three-dimensional geological model for forward analysis and calculation; A visualization sub-module 32, which is used to generate the in-situ stress field distribution based on the characteristics of the terrain, faults, and rock layer structure of the three-dimensional geological model, as well as the applied self-weight pressure and tectonic stress boundary conditions; visually output the calculated in-situ stress components through a contour map; An adjustment and optimization sub-module 33, which is used to compare the calculated in-situ stress components with real-time measured data. If it exceeds the preset range, a model adjustment instruction is triggered; if a deviation is found during verification, the weights of the stress field intelligent analysis model or the parameters of the three-dimensional geological model are optimized, and the inversion result is iteratively updated until the preset accuracy value is met.

[0050] Preferably, this embodiment can input real-time geological data into the in-situ stress field inversion intelligent analysis model for simulation, and feedback the results to the three-dimensional geological model for forward analysis and calculation. It can handle the in-situ stress field inversion problems under complex geological conditions, such as self-weight pressure and tectonic stress boundary conditions, etc.; visually display the in-situ stress field distribution in the form of a contour map, which is convenient for engineering personnel to quickly understand the analysis results and provide decision support; by comparing with the measured data, continuously optimize the model weights or geological model parameters until the preset accuracy value is met. It can quickly respond to real-time data input, combined with forward verification and optimization adjustment, greatly shortening the problems of long time consumption and high cost of traditional methods. Through modular design and iterative optimization mechanism, the system can adapt to the complex engineering requirements under different geological conditions, such as engineering scenarios like deep-buried tunnels and underground power plant houses of hydropower stations; the visualization output in the form of a contour map makes the in-situ stress field distribution clear at a glance, providing an intuitive basis for engineering design and disaster prevention and control; through accurate in-situ stress field inversion and optimization adjustment, the accuracy of surrounding rock stability analysis is improved, thus effectively reducing the disaster risks such as rock bursts and large deformations.

[0051] As Figure 10 shown, this embodiment also provides an embodiment of the three-dimensional in-situ stress field inversion intelligent analysis method for layered surrounding rock. In this embodiment, the three-dimensional in-situ stress field inversion intelligent analysis method for layered surrounding rock is applied to the three-dimensional in-situ stress field inversion intelligent analysis system for layered surrounding rock as described in the above embodiment. The three-dimensional in-situ stress field inversion intelligent analysis method for layered surrounding rock specifically includes the following steps: Step S1: Collect geological data such as fault distribution, rock stratum attitude, physical and mechanical parameters, and measured in-situ stress data in the target layered surrounding rock; normalize the geological data and conduct sensitivity analysis; construct a three-dimensional geological model and perform numerical simulation; Step S2: The intelligent inversion institute designs parameter combinations through orthogonal experiments, generates training samples in combination with the numerical simulation results; generates an intelligent analysis model for in-situ stress field inversion based on the three-dimensional geological model; uses the training samples to train the intelligent analysis model for in-situ stress field inversion; Step S3: Input the real-time geological data into the intelligent analysis model for in-situ stress field inversion to obtain the in-situ stress inversion analysis result; visually output the in-situ stress analysis result; and adjust the intelligent analysis model for in-situ stress field inversion according to the measured data.

[0052] Table 1 Geological description

[0053] Preferably, in this embodiment, by collecting geological data (such as fault distribution, rock stratum attitude, physical and mechanical parameters, and measured in-situ stress data) of the target layered surrounding rock, and conducting normalization processing and sensitivity analysis, standardized basic data is provided for subsequent modeling and inversion. By constructing an accurate three-dimensional geological model, the spatial distribution characteristics of the geological body can be more comprehensively reflected, and combined with numerical simulation technology, accurate inversion of the in-situ stress field can be achieved. The orthogonal experiment design parameter combination is adopted, training samples are generated in combination with the numerical simulation results, and an intelligent analysis model for in-situ stress field inversion is generated based on the three-dimensional geological model. The input of real-time geological data is emphasized. By inputting the latest measured data into the intelligent analysis model, the model parameters are dynamically adjusted to improve the inversion accuracy. This dynamic adjustment mechanism can adapt to changes under complex geological conditions and ensure the real-time performance and accuracy of the model results. The in-situ stress analysis result is visually presented through visualization technology, and the model is further optimized according to the measured data. By combining numerical simulation and intelligent algorithms, this method significantly improves the accuracy of in-situ stress field inversion. For complex geological conditions such as deep-buried tunnels and large fault tectonic areas, this method can effectively handle multi-variable and non-linear problems and provide reliable data support for engineering design. The input of real-time geological data and the model dynamic adjustment mechanism enable this method to adapt to changes in geological conditions and ensure the timeliness and accuracy of the inversion results. The visual output not only improves the intuitiveness of the results but also provides a direct reference basis for engineering design and construction, helping to reduce engineering risks and improve economic benefits (for the specific principle, refer to the appendix Figure 11 )

[0054] As Figure 12 shown, this embodiment provides an embodiment of an electronic device. In this embodiment, the electronic device 4 includes a processor 41 and a memory 42 coupled to the processor 41.

[0055] The memory 42 stores program instructions for implementing the layout method of the intelligent analysis system for the three-dimensional in-situ stress field of layered surrounding rock in any of the above embodiments.

[0056] The processor 41 is configured to execute the program instructions stored in the memory 42 to perform the layout of the intelligent analysis system for the three-dimensional in-situ stress field of layered surrounding rock.

[0057] Among them, the processor 41 can also be referred to as a CPU (Central Processing Unit). The processor 41 may be an integrated circuit chip with the ability to process signals. The processor 41 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0058] Furthermore, Figure 13 FIG. is a schematic structural diagram of a storage medium according to an embodiment of the present application. The storage medium 5 of the embodiment of the present application stores program instructions 51 that can implement all the above methods. Among them, the program instructions 51 can be stored in the above storage medium in the form of a software product, and specifically include several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes, or a terminal device such as a computer, a server, a mobile phone, or a tablet.

[0059] In several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection of the device or unit may be in an electrical, mechanical, or other form.

[0060] In addition, in each embodiment of the present invention, each functional unit may be integrated into a processing unit, may exist separately as individual physical units, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit. The above is only the implementation mode of the present invention, and does not limit the patent scope of the present invention accordingly. Any equivalent structural or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied to other related technical fields, shall be included in the patent protection scope of the present invention by the same token.

[0061] The specific embodiments of the invention have been described in detail above, but they are only examples, and the present invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present invention. Therefore, all equivalent transformations, modifications, improvements, etc. made without departing from the spirit and principles of the present invention should be covered by the scope of the present invention.

Claims

1. A three-dimensional in-situ stress field inversion intelligent analysis system for stratified surrounding rock, characterized in that The three-dimensional in-situ stress field inversion intelligent analysis system for layered surrounding rock includes: A three-dimensional geological numerical module, which is used to collect geological data such as fault distribution, rock stratum occurrence, physical and mechanical parameters, and in-situ stress measured data in the target layered surrounding rock; normalize the geological data and conduct sensitivity analysis; construct a three-dimensional geological model and conduct numerical simulation; An intelligent inversion analysis module, which is used to generate training samples by combining the parameter combinations designed through orthogonal experiments and the numerical simulation results; generate an in-situ stress field inversion intelligent analysis model based on the three-dimensional geological model; and use the training samples to train the in-situ stress field inversion intelligent analysis model; An inversion result verification module, which is used to input real-time geological data into the in-situ stress field inversion intelligent analysis model to obtain the in-situ stress inversion analysis result; visually output the in-situ stress analysis result; and adjust the in-situ stress field inversion intelligent analysis model according to the measured data.

2. The intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock according to claim 1, characterized in that, The three-dimensional geological numerical module includes: A rock stratum grading sub-module, which is used to import drilling data, geological exploration data, geophysical data, and in-situ stress measured data, and define the spatial range of the three-dimensional geological model; determine the range of rock mass mechanical parameters and in-situ stress parameters in different regions through geological zoning; A shear extrusion action sub-module, which is used to conduct three-dimensional mesh division to generate regular tetrahedral elements; process complex fault regions and assign heterogeneous rock mechanical parameters to grid cells; and apply normal stress on the top surface of the three-dimensional geological model to simulate the self-weight stress field to obtain the shear extrusion action reflecting the fault structure; A three-dimensional geological model adjustment sub-module, which is used to comprehensively characterize the non-linear superposition effect of the in-situ stress field through the piezometric coefficient and the rock mass volume force; compare the measured data and the simulated values of the in-situ stress components at the measuring points, and dynamically adjust the three-dimensional geological model by combining the comparison results with the real-time geological data.

3. The intelligent analysis system for three-dimensional in-situ stress field inversion of stratified surrounding rock according to claim 2, wherein The rock stratum grading sub-module includes: A classification and sorting unit, which is used to import drilling data, geological exploration data, and geophysical data, divide the strata through borehole data, uniformly number and sort them, classify them into the same rock stratum unit according to the lithology pixel row and vertical position, and process the influence of faults on the continuity of the bottom layer; A fault trace unit, which is used to determine the fault dip and extension depth based on the fault distribution and rock stratum occurrence information in the geological exploration data, combined with the geophysical inversion results, and generate a fault trace line; A spatial range definition unit, which is used to correct the rock mass morphology using borehole data, combine seismic interpretation horizons with velocity / density data, divide different lithological units, fuse the fault geometric relationship and the stratigraphic boundary data, and define the spatial range of the three-dimensional geological model.

4. The intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock according to claim 2, characterized in that, The shear extrusion action sub-module includes: A mesh division unit, which is used to conduct mesh dissection on the three-dimensional geological model; process the fault intersection relationship in complex fault regions; locally encrypt the mesh in the fault zone; and assign different rock mass mechanical parameters to the corresponding grid cells according to the fault type and measured data; A normal constraint unit, which is used to apply normal stress boundary conditions on the top surface of the three-dimensional geological model, simulate the self-weight stress field of the rock mass by calculating the self-weight generated blowing response stress through the rock mass density; and apply normal constraints on the ground and lateral boundaries of the three-dimensional geological model; The stress direction unit applies horizontal tectonic stress to the lateral boundary of the 3D geological model. The normal unit load is applied through the lateral boundary to simulate the horizontal tectonic stress component, and the proportional relationship between the horizontal stress and the vertical stress is defined in combination with the coefficient of lateral pressure. The direction of tectonic stress is determined by inverting the measured in-situ stress data.

5. The intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock according to claim 1, characterized in that, The intelligent inversion analysis module includes: The sample set formation sub-module is used to select the parameter combinations that affect the in-situ stress field and perform numerical simulations using the 3D geological model; substitute the parameter combinations into the 3D geological model to calculate the distribution results of the in-situ stress field under different working conditions; and form a training sample set based on the simulation results and the parameter combinations. The inversion analysis model generation sub-module is used to generate an intelligent analysis model for in-situ stress field inversion based on the 3D geological model; determine the center position of the radial basis function through self-organizing learning, and establish the mapping relationship between the input parameters and the output in-situ stress field; screen the key parameters and perform non-linear fitting. The inversion analysis model optimization sub-module is used to input the generated training samples into the intelligent analysis model for in-situ stress field inversion, adjust the fitting coefficient until it meets the preset value, and then compare the measured value with the inverted value to optimize the intelligent analysis model for in-situ stress field inversion to obtain the final inversion intelligent analysis model.

6. The intelligent analysis system for inverse analysis of three-dimensional in-situ stress field of layered surrounding rock according to claim 5, characterized in that The sample set formation sub-module includes: The stress field distribution unit is used to screen the key parameters that affect the in-situ stress field in the 3D geological model and determine the parameter combination scheme; input the selected parameter combination into the 3D geological model for numerical simulation calculation to obtain the distribution results of the in-situ stress field under different working conditions. The parameter combination unit is used to extract the in-situ stress components and distribution laws of the key measurement points from the 3D geological model after each numerical simulation and record the corresponding parameter combination; use the parameter combination as the input variable and the corresponding simulated in-situ stress field result as the output variable. The training sample set storage unit is used to combine the input variable and the output variable into parameter-in-situ stress field paired data; eliminate the abnormal data due to unqualified parameters and optimize the sample through iterative optimization; store the verified parameter combination and the simulated in-situ stress field data in a preset format as the training sample set.

7. The intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock according to claim 6, characterized in that, The inversion analysis model generation sub-module includes: The weight setting unit is used to generate an intelligent analysis model for in-situ stress field inversion based on the 3D geological model; calculate the stress contribution values of each parameter combination under the simulation working conditions, and determine the influence weight of the sensitivity of the in-situ stress field distribution on the output result by quantifying the sensitivity of each parameter to the in-situ stress field distribution. The balance relationship unit is used to sort the calculated stress contribution values according to the magnitude and compare them with the pre-set error threshold; adjust the parameter combination through repeated iteration, retain the key parameters for non-linear fitting; evaluate the overall inversion error for each iteration; introduce weight parameters to balance the relationship between data-driven and physical constraints. The iterative optimization unit is used to establish the mapping relationship between the input parameters and the in-situ stress field output, determine the function center position through self-organizing learning, and adjust the fitting coefficient based on the screened key parameter set until the inversion intelligent analysis model meets the preset 3D distribution characteristic standard of the in-situ stress field.

8. The intelligent analysis system for inverse analysis of three-dimensional in-situ stress field of layered surrounding rock according to claim 7, characterized in that, The balance relationship unit includes: A weight - assigning subunit, which is used to obtain a data - driven term based on the sum of squares of the errors between the numerical simulation calculation values and the measured values, and use the historical data of layered rock surrounding rock as a physical constraint term; assign different weights to the data - driven term and the physical constraint term; A weight - optimizing subunit, which is used to dynamically adjust the weight parameters according to the comparison between the current inversion error and a preset threshold during the iteration process; screen out key parameters by quantifying the sensitivity of the parameters to the in - situ stress field and assign higher weights to them; eliminate non - key parameters; A parameter combination and weight configuration subunit, which is used to combine the prior distribution of physical constraints with the likelihood function of data - driven to obtain a posterior probability; optimize the weight parameters based on the posterior probability; iterate sequentially until a preset tolerance is reached or the weight change reaches a preset trend, terminate the iteration, and output the optimized parameter combination and weight configuration.

9. The intelligent analysis system for three-dimensional in-situ stress field inversion of layered surrounding rock according to claim 1, characterized in that, An inversion result verification module, including: A forward - modeling verification sub - module, which is used to communicate the in - situ stress field inversion intelligent analysis model with a three - dimensional geological model, input real - time geological data into the in - situ stress field inversion intelligent analysis model for simulation, and substitute the output parameters of the in - situ stress field inversion intelligent analysis model into the three - dimensional geological model for forward - modeling analysis and calculation; A visualization sub - module, which is used to generate an in - situ stress field distribution based on the terrain, faults, and rock - layer structure characteristics of the three - dimensional geological model, as well as applying self - weight pressure and tectonic stress boundary conditions; visually output the calculated in - situ stress components through a contour map; An adjustment and optimization sub - module, which is used to compare the calculated in - situ stress components with real - time measured data. If it exceeds the preset range, trigger an instruction to adjust the model; if a deviation is found during verification, optimize the weights of the in - situ stress field intelligent analysis model or the parameters of the three - dimensional geological model, and iteratively update the inversion result until the preset accuracy value is met.

10. A three-dimensional in-situ stress field inversion intelligent analysis method for layered surrounding rock, which is applied to the three-dimensional in-situ stress field inversion intelligent analysis system for layered surrounding rock as described in any one of claims 1 to 9, and is characterized in that, The intelligent analysis method for in - situ stress field inversion of layered surrounding rock includes: Collect geological data such as fault distribution, rock - layer occurrence, physical and mechanical parameters, and in - situ stress measured data in the target layered surrounding rock; normalize the geological data and conduct sensitivity analysis; construct a three - dimensional geological model and conduct numerical simulation; The intelligent inversion generates training samples through orthogonal experimental design parameter combinations and combines them with the numerical simulation results; generate an in - situ stress field inversion intelligent analysis model based on the three - dimensional geological model; use the training samples to train the in - situ stress field inversion intelligent analysis model; Input real - time geological data into the in - situ stress field inversion intelligent analysis model to obtain an in - situ stress inversion analysis result; visually output the in - situ stress analysis result; and adjust the in - situ stress field inversion intelligent analysis model according to the measured data.

Images