Three-dimensional disturbance stress field inversion method and device for surrounding rock under multi-point cooperative monitoring

Abstract

This application provides a method and apparatus for inverting the three-dimensional disturbed stress field of surrounding rock under multi-point collaborative monitoring, belonging to the field of stress field inversion technology. The method includes: acquiring multi-source monitoring data; preprocessing and spatiotemporally registering the multi-source monitoring data; establishing a three-dimensional mesh model including geological structural features and engineering excavation process; performing forward modeling based on the three-dimensional mesh model to establish a response surface proxy model; constructing a multi-objective inversion function including displacement error and stress error; iteratively solving the multi-objective inversion function to obtain the optimal parameter set; substituting the optimal parameter set into the three-dimensional mesh model for full-field calculation to reconstruct the three-dimensional disturbed stress field; and evaluating and providing early warning of surrounding rock stability based on the three-dimensional disturbed stress field, and providing feedback to guide engineering construction. This application can solve the technical problem of large deviations between the calculated results of the three-dimensional disturbed stress field of surrounding rock and the actual stress state in the prior art.

Description

Technical Field

[0001] This application relates to the field of stress field inversion technology, and in particular to a method and apparatus for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring. Background Technology

[0002] With the continuous expansion of the scale of underground transportation tunnels, underground energy storage projects and large-scale hydropower cavern projects, underground engineering is gradually developing towards areas with deep burial, high ground stress and complex geological structures. During the step-by-step excavation process, the surrounding rock exhibits significant stress redistribution and nonlinear mechanical response characteristics. The stability of the surrounding rock is increasingly becoming an important factor restricting engineering safety and construction efficiency.

[0003] The evolution of the three-dimensional stress field generated by excavation disturbance in surrounding rock exhibits significant spatial non-uniformity and time dependence. Failure to accurately grasp the distribution law of the disturbed stress field can easily lead to risks such as localized stress concentration, expansion of the plastic zone, and enhanced fracture activity. Therefore, accurate acquisition and dynamic analysis of the disturbed stress field of surrounding rock has become a key technical issue in the design and construction control of underground engineering. Currently, existing methods for analyzing surrounding rock stress mainly include numerical simulation analysis based on finite element or finite difference methods, and empirical correction methods based on single-point or limited monitoring data. Traditional numerical simulation methods typically rely on geological exploration data and laboratory rock mechanics test parameters to establish a three-dimensional model and predict the stress distribution of surrounding rock through step-by-step excavation simulation. However, model parameters are mostly derived from experimental statistical results or empirical values, making it difficult to reflect the true mechanical properties of the surrounding rock on-site, resulting in discrepancies between the calculated results and the actual stress state. Furthermore, while some projects use on-site stress gauges or displacement gauges for monitoring, the number of monitoring points is limited and there is a lack of multi-source data fusion mechanisms. The obtained monitoring data is mostly used for post-analysis and fails to effectively participate in the numerical model parameter correction process.

[0004] In summary, the existing technology has a technical problem: the numerical model parameters cannot be inverted and corrected based on multi-point measured data, resulting in a large deviation between the calculated results of the three-dimensional disturbance stress field of the surrounding rock and the actual stress state, which further affects the accuracy of the surrounding rock stability assessment and the scientific nature of construction safety decisions. Summary of the Invention

[0005] The purpose of this application is to provide a method and device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, in order to solve the technical problem in the prior art that the numerical model parameters cannot be corrected based on multi-point measured data, resulting in a large deviation between the calculated result of the three-dimensional disturbance stress field of surrounding rock and the actual stress state, which further affects the accuracy of surrounding rock stability assessment and the scientific nature of construction safety decision-making.

[0006] In view of the above problems, this application provides a method and device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring.

[0007] Firstly, this application provides a method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring. This method is implemented using a device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, and includes: deploying a multi-point collaborative monitoring system in the surrounding rock area of ​​an underground engineering project to acquire multi-source monitoring data of the surrounding rock during excavation disturbance; preprocessing and spatiotemporally registering the multi-source monitoring data to construct a monitoring database; establishing a three-dimensional mesh model containing geological structural features and the engineering excavation process based on the monitoring database; performing forward modeling based on the three-dimensional mesh model to establish a response surface proxy model; constructing a multi-objective inversion function containing displacement error and stress error; iteratively solving the multi-objective inversion function based on the response surface proxy model to obtain the optimal parameter set; substituting the optimal parameter set into the three-dimensional mesh model for full-field calculation to reconstruct the three-dimensional disturbance stress field; and evaluating and issuing early warnings for the stability of the surrounding rock based on the three-dimensional disturbance stress field, and providing feedback to guide engineering construction.

[0008] Preferably, the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring further includes: determining key monitoring areas based on the geometric shape and geological conditions of the underground engineering cavern group; arranging multiple monitoring boreholes at the top arch, sidewalls, and endwalls of the key monitoring areas; installing fiber optic grating three-dimensional stress gauges at different depths of the monitoring boreholes to form a spatial three-dimensional monitoring network for acquiring three-dimensional stress increment data; installing multiple displacement gauges in the monitoring boreholes to acquire displacement deformation data at different depths inside the surrounding rock; arranging a microseismic sensor array on the surface of the surrounding rock to monitor the spatial distribution characteristics of the surrounding rock; and obtaining the multi-source monitoring data based on excavation events, combined with the three-dimensional stress increment data, displacement deformation data, and spatial distribution characteristics.

[0009] Preferably, the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring further includes: demodulating the wavelength data in the three-dimensional stress increment data and converting it into three-dimensional stress tensor data using a temperature compensation algorithm; performing outlier removal and filtering denoising on the displacement deformation data to obtain effective displacement time series data; performing rupture source localization and source mechanism inversion on the waveform signal of the spatial distribution characteristics to obtain the spatial location and energy information of the rupture event as microseismic data; uniformly mapping the three-dimensional stress tensor data, effective displacement time series data, and microseismic data to the same spatial coordinate system for spatiotemporal registration of multi-source monitoring data; and archiving the registered multi-source monitoring data according to the excavation step sequence and timestamp to construct the monitoring database.

[0010] Preferably, the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring further includes: establishing a geological geometric model including stratigraphic interfaces, faults, joints, and weak interlayers based on geological exploration data and the monitoring database; performing grid discretization processing on the geological geometric model to generate a three-dimensional grid model; assigning initial elastic modulus, Poisson's ratio, and strength parameters to the three-dimensional grid model based on indoor and outdoor rock mechanics test results; determining the initial geostress parameters of the three-dimensional grid model based on measured geostress test results; and setting up a step-by-step excavation and unloading simulation step in the three-dimensional grid model according to the construction excavation sequence and time schedule.

[0011] Preferably, the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring further includes: selecting initial elastic modulus, Poisson's ratio, strength parameters, and initial geostress parameters as variables to be inverted, and determining the value range of the variables to be inverted; generating parameter samples to be inverted within the value range using an orthogonal design method; inputting the parameter samples to be inverted into the three-dimensional mesh model for forward modeling to obtain calculated values; establishing a nonlinear mapping sample set between the parameter samples to be inverted and the monitoring response based on the calculated values; and constructing the response surface proxy model based on the nonlinear mapping sample set.

[0012] Preferably, the method for inverting the three-dimensional disturbance stress field of the surrounding rock under multi-point collaborative monitoring further includes: calculating the relative error between the multi-source monitoring data and the calculated value, constructing displacement objective function components and stress objective function components; setting weight coefficients for displacement and stress terms, and weighting and summing the displacement objective function components and stress objective function components to construct a multi-objective inversion function.

[0013] Preferably, the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring further includes: using a global optimization search strategy to randomly generate an initial population within the value range of the parameter samples to be inverted, and calling the response surface surrogate model to calculate the fitness value of individuals; performing selection, crossover, and mutation operations on the initial population according to the fitness value to iteratively generate a new generation population; determining whether the current iteration result meets the convergence criterion, and if not, continuing iterative optimization to output the optimal parameter set that meets the convergence criterion.

[0014] Preferably, the method for inverting the three-dimensional perturbation stress field of surrounding rock under multi-point collaborative monitoring further includes: substituting the optimal parameter set into the three-dimensional mesh model to simulate the stress redistribution during the excavation process of the surrounding rock, extracting the magnitude and direction of the principal stress, and constructing a tensor dataset of the three-dimensional perturbation stress field; using an interpolation algorithm to perform spatial smoothing processing on the tensor dataset to generate a continuously distributed three-dimensional perturbation stress cloud map; and combining the distribution of rupture events monitored by microseismic monitoring to locally correct and verify the three-dimensional perturbation stress field of the three-dimensional perturbation stress cloud map.

[0015] Preferably, the method for inverting the three-dimensional disturbance stress field of the surrounding rock under multi-point collaborative monitoring further includes: calculating the stress concentration factor and safety factor of the surrounding rock based on the three-dimensional disturbance stress field, and plotting the distribution range of the plastic zone; comparing the fracture concentration area of ​​the microseismic monitoring with the distribution range of the plastic zone, and performing stability evaluation; setting an early warning threshold according to the stability evaluation result, and outputting an early warning signal for the stable state of the surrounding rock; and dynamically optimizing the excavation sequence according to the early warning signal and the inversion result.

[0016] Secondly, this application also provides a device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, used to execute the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in the first aspect, including: a multi-source monitoring data acquisition module, used to deploy a multi-point collaborative monitoring system in the surrounding rock area of ​​underground engineering to acquire multi-source monitoring data of the surrounding rock during the excavation disturbance process; a monitoring database construction module, used to preprocess and spatiotemporally register the multi-source monitoring data to construct a monitoring database; a three-dimensional mesh model establishment module, used to establish a three-dimensional mesh model containing geological structural features and engineering excavation process based on the monitoring database; and a response surface. The system includes the following modules: a proxy model building module for performing forward modeling based on a 3D mesh model to establish a response surface proxy model; a multi-objective inversion function construction module for constructing a multi-objective inversion function that includes displacement and stress errors; an optimal parameter set acquisition module for iteratively solving the multi-objective inversion function based on the response surface proxy model to obtain the optimal parameter set; a 3D disturbance stress field reconstruction module for substituting the optimal parameter set into the 3D mesh model for full-field calculation to reconstruct the 3D disturbance stress field; and a stability evaluation module for evaluating and providing early warning of surrounding rock stability based on the 3D disturbance stress field, and providing feedback to guide engineering construction.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by achieving the technical goal of inverting surrounding rock parameters and reconstructing stress field based on multi-point collaborative monitoring data and three-dimensional geomechanical model, it achieves the technical effects of accurately characterizing the three-dimensional disturbance stress field of surrounding rock, synchronously mapping plastic zone and fracture concentration zone, timely and reliable construction early warning, and dynamic optimization of excavation sequence.

[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring in this application.

[0021] Figure 2 This is a schematic diagram of the structure of the three-dimensional disturbance stress field inversion device for surrounding rock under multi-point collaborative monitoring in this application.

[0022] Figure labeling: 1. Multi-source monitoring data acquisition module; 2. Monitoring database construction module; 3. Three-dimensional mesh model establishment module; 4. Response surface proxy model establishment module; 5. Multi-objective inversion function construction module; 6. Optimal parameter set acquisition module; 7. Three-dimensional disturbance stress field reconstruction module; 8. Stability evaluation module. Detailed Implementation

[0023] This application provides a method and apparatus for inverting the three-dimensional disturbed stress field of surrounding rock under multi-point collaborative monitoring. This solves the technical problem in existing technologies where the numerical model parameters cannot be corrected based on multi-point measured data, leading to significant deviations between the calculated three-dimensional disturbed stress field and the actual stress state, further affecting the accuracy of surrounding rock stability assessment and the scientific basis of construction safety decisions. The application achieves the technical goals of inverting surrounding rock parameters and reconstructing the stress field with precision based on multi-point collaborative monitoring data and a three-dimensional geomechanical model. This results in accurate characterization of the three-dimensional disturbed stress field of surrounding rock, synchronous mapping of plastic zones and fracture concentration zones, timely and reliable construction early warning, and dynamic optimization of excavation sequence.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, which is applied to a three-dimensional disturbance stress field inversion device for surrounding rock under multi-point collaborative monitoring. The method specifically includes the following steps: A multi-point collaborative monitoring system is deployed in the surrounding rock area of ​​underground engineering to obtain multi-source monitoring data of the surrounding rock during the excavation disturbance process.

[0026] Furthermore, this application also includes: determining key monitoring areas based on the geometric shape and geological conditions of the underground engineering cavern group; arranging multiple monitoring boreholes at the top arch, sidewalls, and endwalls of the key monitoring areas; installing fiber optic grating three-dimensional stress gauges at different depths of the monitoring boreholes to form a spatial three-dimensional monitoring network for acquiring three-dimensional stress increment data; installing multi-point displacement gauges inside the monitoring boreholes to acquire displacement deformation data at different depths within the surrounding rock; arranging a microseismic sensor array on the surface of the surrounding rock to monitor the spatial distribution characteristics of the surrounding rock; and obtaining the multi-source monitoring data based on excavation events, combined with the three-dimensional stress increment data, displacement deformation data, and spatial distribution characteristics.

[0027] Specifically, determining key monitoring areas based on the geometric morphology and geological conditions of underground engineering cavern groups involves comprehensively analyzing the spatial distribution, cross-sectional dimensions, spacing, and burial depth of multiple cavern structures in the underground engineering before deploying monitoring. This analysis is combined with geological conditions such as stratum lithology, structural plane development, fault and joint distribution, original geostress field characteristics, and groundwater availability to identify sensitive sections that may experience significant stress redistribution, deformation concentration, or fracture evolution under excavation disturbance. These sensitive sections are then designated as key monitoring areas to improve the targeting and effectiveness of monitoring deployment.

[0028] Furthermore, multiple monitoring boreholes are arranged at the top arch, side walls, and end walls of the key monitoring area. Fiber Bragg grating three-dimensional stress gauges are installed at different depths of the monitoring boreholes to form a spatial three-dimensional monitoring network for acquiring three-dimensional stress increment data. This involves drilling multiple monitoring channels along the key stress-bearing parts of the surrounding rock structure within the identified key monitoring area. Each monitoring borehole extends radially or axially to different depths within the surrounding rock, and three-dimensional stress gauges based on the fiber Bragg grating sensing principle are deployed at different burial depths. The fiber Bragg grating three-dimensional stress gauges achieve high-precision measurement of stress component changes by monitoring changes in grating wavelength, thereby constructing a three-dimensional monitoring network structure covering different orientations and depths in space to continuously collect three-dimensional stress increment data generated by the surrounding rock during excavation disturbance.

[0029] Furthermore, installing multi-point displacement gauges inside the monitoring borehole to obtain displacement deformation data at different depths within the surrounding rock involves simultaneously arranging multi-point displacement gauge devices inside the monitoring borehole where stress sensors have already been installed. The multi-point displacement gauges are fixed to different depths of the surrounding rock via anchoring ends and measuring rod structures, enabling real-time measurement of the relative displacement changes at each measuring point within the surrounding rock. This allows for the acquisition of deformation data such as radial displacement, axial displacement, or shear displacement of the surrounding rock at different depths, reflecting the evolution of deformation within the surrounding rock over time and space.

[0030] Subsequently, a microseismic sensor array is deployed on the surface of the surrounding rock to monitor the spatial distribution characteristics of the surrounding rock. This involves rationally deploying multiple microseismic sensor units on the surface of the surrounding rock and in the adjacent area of ​​the cavern. Each microseismic sensor records the elastic wave signals generated by the micro-fracture activities inside the surrounding rock through a high-sensitivity seismic detector. The three-dimensional spatial coordinates and released energy of the fracture source are determined through a multi-sensor joint positioning algorithm, thereby obtaining the spatial distribution characteristics and activity intensity information of the micro-fracture events inside the surrounding rock, which is used to characterize the damage evolution state of the surrounding rock.

[0031] Finally, based on the excavation event, combined with three-dimensional stress increment data, displacement deformation data, and spatial distribution characteristics, multi-source monitoring data is obtained. This refers to using the construction excavation process and time nodes as disturbance triggering conditions, as well as the three-dimensional stress increment data, displacement deformation data, and microseismic fracture spatial distribution information collected in different time periods, to integrate and correlate them. The data is then fused according to a unified time reference and spatial coordinate system to form a comprehensive monitoring data set reflecting the stress state, deformation state, and fracture state of the surrounding rock. This data serves as the basic data source for subsequent stress field inversion analysis and stability assessment.

[0032] The multi-source monitoring data is preprocessed and spatiotemporally registered to construct a monitoring database.

[0033] Furthermore, this application also includes: demodulating the wavelength data in the three-dimensional stress increment data and converting it into three-dimensional stress tensor data using a temperature compensation algorithm; performing outlier removal and filtering denoising on the displacement deformation data to obtain effective displacement time series data; locating the rupture source and inverting the source mechanism of the waveform signal with the spatial distribution characteristics to obtain the spatial location and energy information of the rupture event as microseismic data; mapping the three-dimensional stress tensor data, effective displacement time series data, and microseismic data to the same spatial coordinate system for spatiotemporal registration of multi-source monitoring data; and archiving the registered multi-source monitoring data according to the excavation step sequence and timestamp to construct the monitoring database.

[0034] Specifically, demodulating the wavelength data in the three-dimensional stress increment data and converting it into three-dimensional stress tensor data using a temperature compensation algorithm involves spectral analysis of the center wavelength drift output by the fiber optic grating three-dimensional stress meter, extracting the wavelength change values ​​corresponding to different gratings using a wavelength demodulation device, establishing a functional relationship between wavelength change and strain based on the strain sensitivity coefficient of the fiber optic grating, and introducing a temperature compensation algorithm to eliminate the coupling effect of ambient temperature changes on wavelength drift. Then, based on the constitutive relationship of the rock mass, the strain components are converted into corresponding stress components, forming three-dimensional stress tensor data containing three principal stress components and shear stress components, which is used to characterize the changes in the actual stress state inside the surrounding rock.

[0035] Furthermore, outlier removal and filtering are performed on the displacement deformation data to obtain effective displacement time series data. This refers to the use of statistical discrimination methods to identify abnormal data points caused by sensor disturbances, data transmission errors, or instantaneous interference in continuous displacement monitoring records collected by multi-point displacement gauges. Outlier removal is then performed by setting threshold ranges or using box plot discrimination criteria. At the same time, digital filtering algorithms are combined to smooth and denoise the original displacement sequence, eliminating high-frequency random fluctuation components, thereby obtaining continuous displacement time series data that can truly reflect the deformation evolution trend of the surrounding rock.

[0036] Subsequently, the spatial distribution characteristics of the waveform signal are used to locate the rupture source and invert the focal mechanism to obtain the spatial location and energy information of the rupture event as microseismic data. This involves identifying the first arrival and picking up the arrival time of the elastic wave waveform signal collected by the microseismic sensor array, calculating the three-dimensional spatial coordinates of the rupture source through the time difference inversion algorithm to locate the rupture source, and simultaneously performing focal mechanism inversion based on amplitude information and phase polarity distribution to determine the rupture type and energy level. This allows for the extraction of spatial location parameters and energy parameters of the rupture event, and the relevant parameters are organized into microseismic data to characterize the damage evolution process inside the surrounding rock.

[0037] Furthermore, mapping three-dimensional stress tensor data, effective displacement time series data, and microseismic data to the same spatial coordinate system for spatiotemporal registration of multi-source monitoring data involves establishing a unified spatial reference coordinate system and time reference system. This involves converting and mapping data acquired by different monitoring systems according to the spatial coordinates corresponding to the sensor installation locations, and synchronously calibrating various types of data based on a unified timestamp. This ensures that stress data, displacement data, and microseismic data correspond one-to-one in both spatial location and time dimension, thereby completing the spatiotemporal registration process for multi-source monitoring data.

[0038] Finally, the registered multi-source monitoring data are archived according to the excavation steps and timestamps to construct a monitoring database. This involves classifying, organizing, and structuring the stress, displacement, and microseismic data that have completed spatiotemporal registration based on the excavation process number and corresponding time node during the construction process. A data storage structure containing fields such as spatial coordinates, time information, physical quantity type, and monitoring values ​​is established to form a monitoring database for subsequent forward and inverse calculations.

[0039] A three-dimensional mesh model containing geological structural features and engineering excavation process was established based on the monitoring database.

[0040] Furthermore, this application also includes: establishing a geological geometric model including stratigraphic interfaces, faults, joints, and weak interlayers based on geological exploration data and the monitoring database; performing grid discretization processing on the geological geometric model to generate a three-dimensional grid model; assigning initial elastic modulus, Poisson's ratio, and strength parameters to the three-dimensional grid model based on indoor and outdoor rock mechanics test results; determining the initial geostress parameters of the three-dimensional grid model based on measured geostress test results; and setting simulated steps of step-by-step excavation and unloading in the three-dimensional grid model according to the construction excavation sequence and time schedule.

[0041] Specifically, based on geological exploration data and monitoring databases, establishing a geological geometric model that includes stratigraphic interfaces, faults, joints, and weak interlayers involves comprehensively utilizing drilling columnar sections, geological logging data, geophysical exploration results, and field exposure information, while also incorporating stress response characteristics and deformation distribution information recorded in the monitoring database. This process spatially reconstructs the underground rock mass structure, clarifies the positional relationships of stratigraphic interfaces between different rock layers, depicts the spatial distribution of fault structures, describes the development direction and density distribution of joints and fractures, and identifies the thickness range and spatial extension characteristics of weak interlayers. This establishes a three-dimensional geological geometric model that reflects the true structural morphology of the surrounding rock, providing a structural foundation for subsequent numerical analysis.

[0042] Furthermore, the geological geometric model is discretized into a three-dimensional mesh model. This involves using numerical methods such as finite element method, finite difference method, or discrete element method to divide the geological geometric entity in continuous space into multiple interconnected discrete elements. Each element has clear node coordinates and topological relationships. By dividing the complex geometric structure into elements and numbering the nodes, a three-dimensional mesh model that can be used for numerical calculation is formed, thus transforming the continuous rock mechanics problem into a solvable discrete mathematical model.

[0043] Subsequently, based on the results of indoor and outdoor rock mechanics tests, the three-dimensional mesh model was assigned initial elastic modulus, Poisson's ratio, and strength parameters. This means that based on the results of uniaxial compression tests, triaxial compression tests, Brazilian splitting tests, and in-situ loading tests, the elastic modulus parameters corresponding to different rock layers were extracted to characterize the stiffness characteristics of the rock mass, and the Poisson's ratio parameters were extracted to reflect the proportional relationship between lateral deformation and longitudinal deformation. Combined with compressive strength, tensile strength, and strength indices such as internal friction angle and cohesion, the relevant mechanical parameters were assigned to the corresponding mesh elements, so that the three-dimensional mesh model has physical properties that reflect the mechanical behavior of the rock mass.

[0044] Furthermore, based on the measured geostress test results, the initial geostress parameters of the three-dimensional mesh model are determined. This means obtaining the original geostress component values ​​of the underground rock mass through hydraulic fracturing tests, stress relief methods, or borehole stress gauge tests, and converting the magnitude and direction information of the principal stresses in different directions into model boundaries or initial field conditions, which are then input into the three-dimensional mesh model as the initial geostress parameters for numerical calculation, thereby ensuring that the numerical simulation process can reflect the actual geostress environment.

[0045] Finally, according to the construction excavation sequence and time schedule, the simulation steps of step-by-step excavation and unloading are set in the three-dimensional mesh model. This means that the tunnel excavation process is divided into multiple stages according to the actual construction organization plan of the project, and the corresponding regional units are gradually deleted or the constraints are reduced in the numerical model in chronological order to simulate the stress release and redistribution process of the surrounding rock during the excavation process. At the same time, the stress and displacement response results are recorded at each stage, thereby realizing the dynamic simulation of the step-by-step excavation and unloading process.

[0046] Forward modeling is performed based on a 3D mesh model to establish a response surface proxy model.

[0047] Furthermore, this application also includes: selecting initial elastic modulus, Poisson's ratio, strength parameters, and initial geostress parameters as variables to be inverted, and determining the value range of the variables to be inverted; generating parameter samples to be inverted within the value range using an orthogonal design method; inputting the parameter samples to be inverted into the three-dimensional mesh model for forward modeling to obtain calculated values; establishing a nonlinear mapping sample set between the parameter samples to be inverted and the monitoring response based on the calculated values; and constructing the response surface surrogate model based on the nonlinear mapping sample set.

[0048] Specifically, selecting initial elastic modulus, Poisson's ratio, strength parameters, and initial geostress parameters as variables to be inverted, and determining the value range of the variables to be inverted, means that based on the established three-dimensional mesh model, key parameters that have a significant impact on the mechanical response of the surrounding rock are used as adjustable variables. Among them, elastic modulus is used to characterize the stiffness characteristics of the rock mass, Poisson's ratio is used to reflect the proportional relationship between lateral and longitudinal deformation, strength parameters including cohesion and internal friction angle are used to describe the rock mass failure criteria, and initial geostress parameters are used to characterize the original geostress field distribution. Based on rock mechanics test results, field test results, and statistical intervals of engineering experience, the upper and lower limits of allowable variation of each parameter are determined, thus forming the parameter space of the variables to be inverted.

[0049] Furthermore, by using orthogonal design methods to generate a sample of parameters to be inverted within a range of values, it is to adopt the principle of orthogonal experimental design, select a representative combination of parameters under multi-parameter and multi-level conditions, and construct an orthogonal table to achieve uniform distribution and mutual independence among different parameter levels, thereby reducing the number of samples while ensuring sample coverage, forming a set of structured sample of parameters to be inverted for subsequent numerical calculation and response analysis.

[0050] Subsequently, the parameter samples to be inverted are input into the three-dimensional mesh model for forward modeling to obtain the calculated values. This means that each set of parameter samples generated by the orthogonal design is assigned to the material properties and initial stress field conditions of the corresponding element in the three-dimensional mesh model. The numerical solution process is executed according to the predetermined excavation simulation steps to calculate the displacement response, stress distribution, and plastic zone range of the surrounding rock under various parameter combinations. The obtained simulation output results are recorded as the calculated values.

[0051] Furthermore, based on the calculated values, a nonlinear mapping sample set between the parameter samples to be inverted and the monitoring response is established. This involves pairing and organizing each set of parameter samples to be inverted with the corresponding numerical simulation output results, while extracting actual monitoring response data from the monitoring database. Through comparative analysis, a correspondence between the parameter space and the response space is constructed. The parameter set is used as the input variable, and physical quantities such as displacement response and stress response are used as the output variables to form a data sample set containing the input-output mapping relationship, which is used to characterize the nonlinear correlation between parameters and response.

[0052] Finally, based on the nonlinear mapping sample set, a response surface surrogate model is constructed. This involves using multinomial regression, radial basis function, or other function approximation techniques to fit and analyze the above-mentioned input-output sample set, establishing a mathematical expression model that can approximately represent the functional relationship between the variable to be inverted and the monitored response. This allows the complex numerical calculation process to be quickly predicted by simplifying the functional form, thus forming a response surface surrogate model, which can be used to replace a large number of repetitive numerical simulation calculations in the subsequent inversion optimization process.

[0053] Construct a multi-objective inversion function that includes displacement error and stress error.

[0054] Furthermore, this application also includes: calculating the relative error between the multi-source monitoring data and the calculated value, constructing displacement objective function components and stress objective function components; setting weight coefficients for the displacement and stress terms, and weighting and summing the displacement objective function components and stress objective function components to construct a multi-objective inversion function.

[0055] Specifically, calculating the relative error between multi-source monitoring data and calculated values, and constructing displacement and stress objective function components, involves comparing the actual displacement time-series data obtained from the monitoring database with the corresponding displacement results obtained from the forward modeling of the numerical model point by point. The degree of difference between the two is calculated by using a relative error expression. At the same time, the three-dimensional stress tensor data obtained from monitoring is matched with the stress component results output by the numerical simulation, and the relative deviation value between the stress components is calculated. Based on this, the displacement error set and the stress error set are statistically summarized in the form of sum of squares or root mean square, respectively, to construct displacement and stress objective function components for quantifying the degree of displacement matching and stress matching, so that the objective function can reflect the goodness or badness of the fit between different physical quantities.

[0056] Furthermore, weighting coefficients are set for displacement and stress terms, and the objective function components of displacement and stress are weighted and summed to construct a multi-objective inversion function. This means that, based on the differences between displacement and stress data in terms of monitoring accuracy, data quantity, and engineering importance, corresponding weighting coefficients are assigned to the objective function components of displacement and stress respectively. The two objective function components are combined through linear weighting to form a unified scalar evaluation function, enabling the inversion process to be comprehensively optimized while taking into account both displacement fitting accuracy and stress fitting accuracy. This establishes a multi-objective inversion function expression for parameter optimization.

[0057] The multi-objective inversion function is iteratively solved based on the response surface surrogate model to obtain the optimal parameter set.

[0058] Furthermore, this application also includes: using a global optimization search strategy to randomly generate an initial population within the value range of the parameter samples to be inverted, and calling the response surface surrogate model to calculate the fitness value of individuals; performing selection, crossover, and mutation operations on the initial population according to the fitness value to iteratively generate a new generation population; determining whether the current iteration result meets the convergence criterion, and if not, continuing iterative optimization to output the optimal parameter set that meets the convergence criterion.

[0059] Specifically, the global optimization search strategy involves randomly generating an initial population within the range of values ​​for the parameters to be inverted, and then calling the response surface surrogate model to calculate the fitness value of each individual. This means using a random search algorithm with global optimization capabilities as the parameter inversion tool. Within the pre-defined upper and lower limits of each variable to be inverted, multiple parameter combination individuals are constructed through a random number generation mechanism. Each individual corresponds to a complete set of mechanical parameters and initial geostress parameters. Each parameter combination is then input into the response surface surrogate model for rapid prediction calculation to obtain the corresponding objective function value. The objective function value is used as the fitness value to measure the quality of the parameter combination, thereby enabling the evaluation of solutions for different parameter combinations.

[0060] Furthermore, the initial population is selected, crossovered, and mutated based on its fitness value to iteratively generate a new generation population. This involves selecting individuals with better fitness from the initial population as parent individuals based on the fitness value ranking results, exchanging some parameter variables between different parent individuals through crossover to generate new parameter combinations, and introducing random perturbations in some parameter dimensions through mutation to enhance population diversity and search range, thereby forming a new generation population containing multiple new parameter combinations. On this basis, fitness evaluation and evolutionary operations are repeated to achieve gradual approximation optimization of the parameter space.

[0061] Subsequently, it is determined whether the current iteration result meets the convergence criterion. If it does not, iterative optimization continues, and the optimal parameter set that meets the convergence criterion is output. This means that after each generation of population evolution is completed, the optimization process is judged according to the pre-set convergence criteria. The convergence criteria include conditions such as the change of the objective function being lower than a set threshold, the change of the optimal fitness value tending to be stable for several consecutive generations, or reaching the maximum number of iterations. When the convergence requirement is not met, the selection, crossover, and mutation process is continued for a new round of iteration. When the convergence criterion is met, the optimization calculation is stopped, and the parameter combination corresponding to the optimal fitness value is output as the optimal parameter set obtained by inversion for subsequent full-field stress calculation.

[0062] The optimal parameter set is substituted into the three-dimensional mesh model for full-field calculation to reconstruct the three-dimensional perturbation stress field.

[0063] Furthermore, this application also includes: substituting the optimal parameter set into the three-dimensional mesh model to simulate stress redistribution during the surrounding rock excavation process, extracting the magnitude and direction of the principal stresses, and constructing a tensor dataset of a three-dimensional perturbation stress field; using an interpolation algorithm to perform spatial smoothing on the tensor dataset to generate a continuously distributed three-dimensional perturbation stress cloud map; and combining the distribution of rupture events monitored by microseismic monitoring to locally correct and verify the three-dimensional perturbation stress field of the three-dimensional perturbation stress cloud map.

[0064] Specifically, substituting the optimal parameter set into a three-dimensional mesh model to simulate stress redistribution during the excavation of surrounding rock, extracting the magnitude and direction of principal stresses, and constructing a tensor dataset of a three-dimensional disturbed stress field refers to reassigning the optimal parameter combination, such as the elastic modulus, Poisson's ratio, strength parameters, and initial ground stress parameters obtained through inversion optimization, to the corresponding element attributes of the three-dimensional mesh model. Numerical calculations are then performed according to a predetermined step-by-step excavation and unloading simulation process to obtain the stress redistribution results of the surrounding rock under excavation disturbance. Eigenvalue decomposition is then performed on the stress tensors of each element output by the calculation to extract the magnitude and corresponding spatial direction information of the maximum principal stress, intermediate principal stress, and minimum principal stress. The stress components and principal stress direction data of each element are then organized into a tensor dataset containing spatial coordinates and stress components to characterize the spatial distribution state of the three-dimensional disturbed stress field.

[0065] Furthermore, spatial smoothing of the tensor dataset is performed using interpolation algorithms to generate a continuously distributed three-dimensional perturbation stress cloud map. This refers to the process of approximating and making continuous the stress values ​​between different unit nodes using spatial interpolation methods on the stress tensor data output by the discrete units of the numerical model. This allows the discrete stress data to form a continuously changing field in the spatial domain, and generates a visual representation based on the principal stress magnitude or equivalent stress index. A three-dimensional perturbation stress cloud map is constructed using color grading or isosurfaces to intuitively reflect the stress concentration and distribution trend in different areas of the surrounding rock.

[0066] Subsequently, based on the distribution of rupture events from microseismic monitoring, the three-dimensional perturbation stress field of the three-dimensional perturbation stress cloud map is locally corrected and verified. This involves superimposing the spatial location and energy level information of rupture events obtained from microseismic monitoring onto the three-dimensional perturbation stress field results, comparing the spatial correspondence between rupture concentration areas and high stress concentration areas, and when a deviation is found between the stress prediction results and the rupture activity distribution, fine-tuning the parameters of local areas or correcting the local stress results is performed to improve the consistency between the stress field prediction results and the actual surrounding rock damage distribution. The rationality of the perturbation stress field results is then verified using the characteristics of microseismic activity.

[0067] Based on the aforementioned three-dimensional disturbance stress field, the stability of the surrounding rock is assessed and an early warning is issued, and feedback is provided to guide engineering construction.

[0068] Furthermore, this application also includes: calculating the stress concentration factor and safety factor of the surrounding rock based on the three-dimensional disturbance stress field, and plotting the distribution range of the plastic zone; comparing the fracture concentration area of ​​the microseismic monitoring with the distribution range of the plastic zone, and performing stability evaluation; setting an early warning threshold based on the stability evaluation results, and outputting an early warning signal for the stability state of the surrounding rock; and dynamically optimizing the excavation sequence based on the early warning signal and the inversion results.

[0069] Specifically, based on the three-dimensional perturbation stress field, the stress concentration factor and safety factor of the surrounding rock are calculated, and the distribution range of the plastic zone is plotted. This means that after the three-dimensional perturbation stress field is reconstructed, the principal stress components or equivalent stress values ​​of each spatial unit are extracted, and the ratio of the local maximum stress to the original ground stress or far-field stress is calculated to obtain the stress concentration factor, which reflects the degree of stress amplification. At the same time, according to the rock mass strength criterion, the calculated stress is compared and analyzed with the corresponding rock mass strength parameters to obtain the safety factor value, which is used to characterize the ability of the surrounding rock to resist failure. Based on the yield criterion, it is determined whether each unit has entered the plastic state, and the area that meets the yield condition is spatially marked to form a plastic zone distribution range map.

[0070] Furthermore, comparing the distribution range of the fracture concentration zone and the plastic zone from microseismic monitoring and conducting stability assessment involves superimposing the high-density areas of fracture events obtained from microseismic monitoring results with the spatial locations of the plastic zones obtained from numerical simulation. The degree of consistency between the two in spatial distribution is compared. When the fracture concentration zone and the plastic zone highly overlap, the surrounding rock is determined to be in a state of stress concentration and active damage evolution. When there is a significant deviation between the two distributions, a comprehensive analysis is conducted by combining the energy release level and development trend, thereby making a quantitative or graded assessment of the overall stability of the surrounding rock.

[0071] Subsequently, based on the stability assessment results, an early warning threshold is set, and an early warning signal for the stability of the surrounding rock is output. This means that different risk level ranges are divided according to indicators such as the range of safety factor values, the variation range of stress concentration factor, and the energy level of rupture events, and corresponding numerical thresholds are set for each level. When the real-time calculation results or monitoring indicators reach or exceed the preset threshold, an early warning signal of the corresponding level is generated. The early warning signal includes status indicators such as stable, attention, warning, or danger, which are used to prompt the construction site to take corresponding control measures.

[0072] Finally, based on the early warning signals and inversion results, the excavation sequence is dynamically optimized. This means that after obtaining the early warning information on the stability of the surrounding rock, the original construction excavation sequence is adjusted in combination with the current three-dimensional disturbance stress field inversion results and stress distribution trends. For example, the excavation length of each section is changed, the timing of support is adjusted, or the construction interval is optimized so that the excavation activities avoid high-risk areas or reduce the degree of stress concentration, thereby achieving dynamic regulation and risk control of the construction process.

[0073] In summary, the three-dimensional disturbance stress field inversion method for surrounding rock under multi-point collaborative monitoring provided in this application has the following technical effects: by achieving the technical goal of inverting surrounding rock parameters and reconstructing stress field based on multi-point collaborative monitoring data and three-dimensional geomechanical model, it achieves the technical effects of accurately characterizing the three-dimensional disturbance stress field of surrounding rock, synchronously mapping plastic zone and fracture concentration zone, timely and reliable construction early warning, and dynamic optimization of excavation sequence.

[0074] Example 2: Based on the same inventive concept as the three-dimensional disturbance stress field inversion method for surrounding rock under multi-point collaborative monitoring in the foregoing examples, this application also provides a device for inverting the three-dimensional disturbance stress field for surrounding rock under multi-point collaborative monitoring. Please refer to the appendix. Figure 2 The system includes: a multi-source monitoring data acquisition module 1, used to deploy a multi-point collaborative monitoring system in the surrounding rock area of ​​underground engineering to acquire multi-source monitoring data of the surrounding rock during the excavation disturbance process; a monitoring database construction module 2, used to preprocess and spatiotemporally register the multi-source monitoring data to construct a monitoring database; a three-dimensional mesh model establishment module 3, used to establish a three-dimensional mesh model containing geological structural features and engineering excavation process based on the monitoring database; a response surface proxy model establishment module 4, used to perform forward modeling calculations based on the three-dimensional mesh model to establish a response surface proxy model; a multi-objective inversion function construction module 5, used to construct a multi-objective inversion function containing displacement error and stress error; an optimal parameter set acquisition module 6, used to iteratively solve the multi-objective inversion function based on the response surface proxy model to obtain the optimal parameter set; a three-dimensional disturbance stress field reconstruction module 7, used to substitute the optimal parameter set into the three-dimensional mesh model for full-field calculation to reconstruct the three-dimensional disturbance stress field; and a stability evaluation module 8, used to evaluate and warn of the surrounding rock stability based on the three-dimensional disturbance stress field and provide feedback to guide engineering construction.

[0075] Furthermore, the multi-point collaborative monitoring device for inverting the three-dimensional disturbance stress field of the surrounding rock is also used for: determining key monitoring areas based on the geometry and geological conditions of the underground engineering cavern group; arranging multiple monitoring boreholes at the top arch, sidewalls, and endwalls of the key monitoring areas, and installing fiber optic grating three-dimensional stress gauges at different depths of the monitoring boreholes to form a spatial three-dimensional monitoring network for acquiring three-dimensional stress increment data; installing multiple displacement gauges in the monitoring boreholes to acquire displacement deformation data at different depths inside the surrounding rock; arranging a microseismic sensor array on the surface of the surrounding rock to monitor the spatial distribution characteristics of the surrounding rock; and obtaining the multi-source monitoring data based on excavation events, combined with the three-dimensional stress increment data, displacement deformation data, and spatial distribution characteristics.

[0076] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used for: demodulating the wavelength data in the three-dimensional stress increment data and converting it into three-dimensional stress tensor data in combination with a temperature compensation algorithm; performing outlier removal and filtering denoising on the displacement deformation data to obtain effective displacement time series data; performing rupture source location and source mechanism inversion on the waveform signal of the spatial distribution characteristics to obtain the spatial location and energy information of the rupture event as microseismic data; uniformly mapping the three-dimensional stress tensor data, effective displacement time series data and microseismic data to the same spatial coordinate system for spatiotemporal registration of multi-source monitoring data; and archiving the registered multi-source monitoring data according to the excavation step sequence and timestamp to construct the monitoring database.

[0077] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used for: establishing a geological geometric model including stratigraphic interfaces, faults, joints, and weak interlayers based on geological exploration data and the monitoring database; performing grid discretization processing on the geological geometric model to generate a three-dimensional grid model; assigning initial elastic modulus, Poisson's ratio, and strength parameters to the three-dimensional grid model based on indoor and outdoor rock mechanics test results; determining the initial geostress parameters of the three-dimensional grid model based on measured geostress test results; and setting up a step-by-step excavation and unloading simulation step in the three-dimensional grid model according to the construction excavation sequence and time schedule.

[0078] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used for: selecting initial elastic modulus, Poisson's ratio, strength parameters, and initial geostress parameters as variables to be inverted, and determining the value range of the variables to be inverted; generating parameter samples to be inverted within the value range using an orthogonal design method; inputting the parameter samples to be inverted into the three-dimensional mesh model for forward modeling to obtain calculated values; establishing a nonlinear mapping sample set between the parameter samples to be inverted and the monitoring response based on the calculated values; and constructing the response surface proxy model based on the nonlinear mapping sample set.

[0079] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used to: calculate the relative error between the multi-source monitoring data and the calculated value, construct displacement objective function components and stress objective function components; set the weight coefficients of displacement and stress terms, and weight the displacement objective function components and stress objective function components to construct a multi-objective inversion function.

[0080] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used to: apply a global optimization search strategy to randomly generate an initial population within the value range of the parameter samples to be inverted, and call the response surface surrogate model to calculate the fitness value of individuals; perform selection, crossover, and mutation operations on the initial population according to the fitness value to iteratively generate a new generation population; determine whether the current iteration result meets the convergence criterion, and if not, continue iterative optimization to output the optimal parameter set that meets the convergence criterion.

[0081] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used to: substitute the optimal parameter set into the three-dimensional mesh model to simulate the stress redistribution during the surrounding rock excavation process, extract the magnitude and direction of the principal stress, and construct a tensor dataset of the three-dimensional disturbance stress field; use an interpolation algorithm to perform spatial smoothing processing on the tensor dataset to generate a continuously distributed three-dimensional disturbance stress cloud map; and combine the distribution of rupture events monitored by microseismic monitoring to locally correct and verify the three-dimensional disturbance stress field of the three-dimensional disturbance stress cloud map.

[0082] Furthermore, the multi-point collaborative monitoring surrounding rock three-dimensional disturbance stress field inversion device is also used for: calculating the stress concentration factor and safety factor of the surrounding rock based on the three-dimensional disturbance stress field, and drawing the distribution range of the plastic zone; comparing the fracture concentration area of ​​the microseismic monitoring with the distribution range of the plastic zone, and performing stability evaluation; setting an early warning threshold according to the stability evaluation result, and outputting an early warning signal for the stability state of the surrounding rock; and dynamically optimizing the excavation sequence according to the early warning signal and the inversion result.

[0083] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The method and specific examples of inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring in the aforementioned embodiment 1 are also applicable to the device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring in this embodiment. Through the foregoing detailed description of the method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, those skilled in the art can clearly understand the device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0084] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0085] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, characterized in that, include: A multi-point collaborative monitoring system was deployed in the surrounding rock area of ​​the underground project to obtain multi-source monitoring data of the surrounding rock during the excavation disturbance process. The multi-source monitoring data are preprocessed and spatiotemporally registered to construct a monitoring database; A three-dimensional mesh model containing geological structural features and engineering excavation process was established based on the monitoring database; Forward modeling is performed based on a 3D mesh model to establish a response surface proxy model; Construct a multi-objective inversion function that includes displacement error and stress error; The multi-objective inversion function is iteratively solved based on the response surface surrogate model to obtain the optimal parameter set. Substitute the optimal parameter set into the three-dimensional mesh model to perform full-field calculations and reconstruct the three-dimensional perturbation stress field. Based on the aforementioned three-dimensional disturbance stress field, the stability of the surrounding rock is assessed and an early warning is issued, and feedback is provided to guide engineering construction.

2. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 1, characterized in that, A multi-point collaborative monitoring system is deployed in the surrounding rock area of ​​underground engineering to acquire multi-source monitoring data of the surrounding rock during the excavation disturbance process, including: Based on the geometric shape and geological conditions of the underground engineering cavern complex, key monitoring areas were identified; Multiple monitoring boreholes are arranged at the top arch, side walls and end walls of the key monitoring area. Fiber grating three-dimensional stress gauges are installed at different depths of the monitoring boreholes to form a spatial three-dimensional monitoring network for obtaining three-dimensional stress increment data. Multiple displacement gauges are installed inside the monitoring borehole to obtain displacement and deformation data at different depths within the surrounding rock. A microseismic sensor array is deployed on the surface of the surrounding rock to monitor the spatial distribution characteristics of the surrounding rock; Based on the excavation event, the multi-source monitoring data is obtained by combining the three-dimensional stress increment data, displacement deformation data, and spatial distribution characteristics.

3. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 2, characterized in that, The multi-source monitoring data is preprocessed and spatiotemporally registered to construct a monitoring database, including: The wavelength data in the three-dimensional stress increment data is demodulated and converted into three-dimensional stress tensor data using a temperature compensation algorithm. The displacement deformation data is subjected to outlier removal and filtering denoising to obtain effective displacement time series data; The waveform signals with the spatial distribution characteristics are used to locate the rupture source and invert the source mechanism to obtain the spatial location and energy information of the rupture event as microseismic data. The three-dimensional stress tensor data, effective displacement time series data and microseismic data are uniformly mapped to the same spatial coordinate system to perform spatiotemporal registration of multi-source monitoring data; The registered multi-source monitoring data are archived according to the excavation sequence and timestamps to construct the monitoring database.

4. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 1, characterized in that, Based on the monitoring database, a three-dimensional mesh model containing geological structural features and the engineering excavation process is established, including: Based on geological exploration data and the aforementioned monitoring database, a geological geometric model including stratigraphic interfaces, faults, joints, and weak interlayers was established. The geological geometric model is discretized into a mesh to generate a three-dimensional mesh model; Based on the results of indoor and outdoor rock mechanics tests, the three-dimensional mesh model was assigned initial elastic modulus, Poisson's ratio, and strength parameters. Based on the measured ground stress test results, the initial ground stress parameters of the three-dimensional mesh model are determined; According to the construction excavation sequence and time schedule, the simulation steps of step-by-step excavation and unloading are set in the three-dimensional mesh model.

5. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 4, characterized in that, Forward modeling is performed based on a 3D mesh model, and a response surface proxy model is established, including: The initial elastic modulus, Poisson's ratio, strength parameters, and initial geostress parameters are selected as variables to be inverted, and the range of values ​​for the variables to be inverted is determined. Using orthogonal design methods, generate a sample of parameters to be inverted within the range of the stated values; The parameter samples to be inverted are input into the three-dimensional mesh model for forward modeling to obtain calculated values. Based on the calculated values, a nonlinear mapping sample set between the parameter samples to be inverted and the monitoring response is established; Based on the nonlinear mapping sample set, the response surface proxy model is constructed.

6. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 5, characterized in that, Construct a multi-objective inversion function that includes displacement error and stress error, including: Calculate the relative error between the multi-source monitoring data and the calculated value, and construct the displacement metric function component and the stress metric function component; By setting weighting coefficients for the displacement and stress terms, the objective function components of the displacement and stress terms are weighted and summed to construct a multi-objective inversion function.

7. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 5, characterized in that, The multi-objective inversion function is iteratively solved based on the response surface surrogate model to obtain the optimal parameter set, including: Using a global optimization search strategy, an initial population is randomly generated within the range of values ​​of the parameter samples to be inverted, and the fitness value of each individual is calculated by calling the response surface surrogate model. The initial population is selected, crossovered, and mutated based on its fitness value to iteratively generate a new generation of population; Determine whether the current iteration result satisfies the convergence criterion. If it does not, continue iterative optimization and output the optimal parameter set that satisfies the convergence criterion.

8. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 1, characterized in that, Substituting the optimal parameter set into the three-dimensional mesh model for full-field calculation, the three-dimensional perturbation stress field is reconstructed, including: The optimal parameter set is substituted into the three-dimensional mesh model to simulate the stress redistribution during the surrounding rock excavation process, and the magnitude and direction of the principal stress are extracted to construct a tensor dataset of the three-dimensional perturbation stress field. The tensor dataset is spatially smoothed using an interpolation algorithm to generate a continuously distributed three-dimensional perturbation stress cloud map. Based on the distribution of rupture events from microseismic monitoring, the three-dimensional perturbation stress field of the three-dimensional perturbation stress cloud map is locally corrected and verified.

9. The method for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring as described in claim 1, characterized in that, Based on the aforementioned three-dimensional disturbed stress field, the stability of the surrounding rock is assessed and an early warning is issued, and feedback is provided to guide engineering construction, including: Based on the three-dimensional perturbation stress field, the stress concentration factor and safety factor of the surrounding rock are calculated, and the distribution range of the plastic zone is plotted. The distribution range of the fracture concentration area detected by microseismic monitoring is compared with that of the plastic zone, and a stability assessment is performed. Based on the stability assessment results, an early warning threshold is set, and an early warning signal for the stability of the surrounding rock is output. Based on the early warning signals and inversion results, the excavation sequence is dynamically optimized.

10. A device for inverting the three-dimensional disturbance stress field of surrounding rock under multi-point collaborative monitoring, characterized in that, The steps for implementing the three-dimensional disturbance stress field inversion method for surrounding rock under multi-point collaborative monitoring as described in any one of claims 1 to 9 include: The multi-source monitoring data acquisition module is used to deploy a multi-point collaborative monitoring system in the surrounding rock area of ​​underground engineering to acquire multi-source monitoring data of the surrounding rock during the excavation disturbance process. The monitoring database construction module is used to preprocess and spatiotemporally register the multi-source monitoring data to construct the monitoring database; A three-dimensional mesh model building module is used to build a three-dimensional mesh model containing geological structural features and engineering excavation process based on the monitoring database. The response surface surrogate model building module is used to perform forward modeling calculations based on a 3D mesh model and build a response surface surrogate model. The multi-objective inversion function construction module is used to construct multi-objective inversion functions that include displacement error and stress error; The optimal parameter set acquisition module is used to iteratively solve the multi-objective inversion function based on the response surface surrogate model to obtain the optimal parameter set. The three-dimensional perturbation stress field reconstruction module is used to substitute the optimal parameter set into the three-dimensional mesh model for full-field calculation and reconstruct the three-dimensional perturbation stress field. The stability assessment module is used to assess and warn of the surrounding rock stability based on the three-dimensional disturbance stress field, and to provide feedback to guide engineering construction.

Images