Integrated Sky-Ground Geological Environment Monitoring and Assessment Method and System

Abstract

This application relates to the field of data processing technology and discloses an integrated sky-ground geological environment monitoring and assessment method and system. The method includes: integrating sky remote sensing, surface monitoring, and underground exploration data into a monitoring matrix, processing the data to generate a raw database; constructing a spatiotemporal standard field based on the database and fusing it to generate a correlation spectrum; decomposing the correlation spectrum to obtain element feature chains; constructing a coupled dynamic field based on the feature chains and analytically obtaining an evolutionary parameter set; establishing a risk assessment matrix based on the parameter set and analyzing to generate an early warning indicator system; and designing a governance plan by combining the early warning system and historical cases. This application achieves deep integration of space-based remote sensing data, surface monitoring data, and underground exploration data by constructing a unified spatiotemporal standard field, a multi-level feature transfer network, and a dynamic coupling model, thereby improving the accuracy and reliability of geological environment monitoring and assessment.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a method and system for integrated air-ground geological environment monitoring and assessment. Background Technology

[0002] Geological environment monitoring and assessment is a crucial aspect of geological disaster prevention and control. Traditional geological environment monitoring methods primarily rely on single data sources, such as ground monitoring stations and satellite remote sensing. In recent years, with the development of multi-source remote sensing technology, Internet of Things (IoT) technology, and artificial intelligence (AI) technology, geological environment monitoring has gradually moved towards multi-source data fusion. Currently, various monitoring technologies have been developed, including InSAR deformation monitoring, GNSS continuous monitoring, and groundwater dynamic monitoring, and have been widely applied in geological disaster monitoring and early warning. Simultaneously, data analysis methods based on deep learning, numerical simulation methods based on physical mechanisms, and expert systems based on knowledge graphs are also continuously developing, providing new technical means for geological environment monitoring and assessment.

[0003] However, existing geological environment monitoring and assessment methods suffer from problems such as insufficient data fusion, incomplete monitoring levels, and simplistic assessment methods. Specifically, space-based remote sensing data, surface monitoring data, and underground exploration data are often processed independently, lacking an effective collaborative analysis mechanism; the spatiotemporal resolution of monitoring data is inconsistent, making accurate integration of multi-source data difficult; risk assessment methods rely too heavily on empirical judgment, lacking systematic and quantitative assessment standards; the early warning indicator system is inadequate, failing to accurately reflect the evolutionary characteristics and development trends of the geological environment; and the formulation of remediation plans lacks data support, hindering precise implementation. Summary of the Invention

[0004] This application provides a method and system for integrated space-ground geological environment monitoring and assessment, which achieves deep integration of space-based remote sensing data, surface monitoring data and underground exploration data by constructing a unified spatiotemporal standard field, a multi-level feature transfer network and a dynamic coupling model, thereby improving the accuracy and reliability of geological environment monitoring and assessment.

[0005] Firstly, this application provides an integrated sky-ground geological environment monitoring and assessment method, which includes: integrating pre-collected sky remote sensing parameters, surface monitoring parameters, and underground detection data into an integrated sky-ground monitoring matrix, and generating a raw geological environment database after hierarchical outlier processing; constructing an integrated sky-ground spatiotemporal standard field based on the raw geological environment database, and generating a geological environment correlation spectrum through spatiotemporal registration and data fusion; performing multi-dimensional feature decomposition on the geological environment correlation spectrum, and generating a geological element feature chain through sky-ground collaborative analysis; constructing a sky-ground coupled dynamic field based on the geological element feature chain, and obtaining a geological environment evolution parameter set through multi-field coupling analysis; establishing an integrated sky-ground risk assessment matrix based on the geological environment evolution parameter set, and generating a risk warning indicator system through quantitative hierarchical analysis; and designing a sky-ground collaborative scheme based on the risk warning indicator system and a historical case database to generate a geological environment governance scheme.

[0006] Secondly, this application provides an integrated air-ground geological environment monitoring and assessment system, which includes:

[0007] The data acquisition module is used to integrate pre-collected sky remote sensing parameters, surface monitoring parameters and underground detection data into the sky-ground integrated monitoring matrix, and generate a raw geological environment database after hierarchical outlier processing.

[0008] The generation module is used to construct an integrated space-air-ground spatiotemporal standard field based on the original geological environment database, and generate a geological environment correlation spectrum through spatiotemporal registration and data fusion.

[0009] The decomposition module is used to perform multi-dimensional feature decomposition on the geological environment correlation spectrum and generate geological element feature chains through sky-ground collaborative analysis.

[0010] The coupling module is used to construct a dynamic field of sky-ground coupling based on the feature chain of geological elements, and to obtain a set of geological environment evolution parameters through multi-field coupling analysis;

[0011] The hierarchical module is used to establish an integrated air-ground risk assessment matrix based on the set of geological environment evolution parameters, and to generate a risk warning indicator system through quantitative hierarchical analysis.

[0012] The design module is used to design a sky-ground collaborative solution based on the risk warning indicator system and combined with the historical case library, and generate a geological environment governance solution.

[0013] The technical solution provided in this application integrates pre-collected sky remote sensing parameters, surface monitoring parameters, and underground detection data into a sky-ground integrated monitoring matrix, achieving collaborative acquisition of multi-source data and ensuring the comprehensiveness and systematic nature of data acquisition. The resulting geological environment raw database, after hierarchical outlier processing, possesses high data quality. Based on the geological environment raw database, a sky-ground integrated spatiotemporal standard field is constructed. Through spatiotemporal registration and data fusion, a geological environment correlation spectrum is generated, achieving deep fusion and spatiotemporal alignment of multi-source heterogeneous data. The geological environment correlation spectrum undergoes multi-dimensional feature decomposition and is analyzed through sky-ground collaborative resolution. This method generates a geological element feature chain, extracts key features of geological elements, and establishes the correlation between elements. Based on the geological element feature chain, a space-air-ground coupled dynamic field is constructed. Through multi-field coupling analysis, a set of geological environment evolution parameters is obtained, achieving a precise characterization of the dynamic evolution characteristics of the geological environment. An integrated space-air-ground risk assessment matrix is ​​established based on the geological environment evolution parameter set. The risk warning indicator system generated through quantitative hierarchical analysis has strong scientific validity and operability. Using the risk warning indicator system as a benchmark and combining it with a historical case database, a space-air-ground collaborative solution is designed, generating a geological environment governance solution with strong pertinence and practicality. This method organically integrates monitoring data from the sky, surface, and subsurface levels, establishing a complete data processing, analysis, assessment, and governance process, significantly improving the accuracy and reliability of geological environment monitoring and assessment. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of one embodiment of the integrated air-ground geological environment monitoring and assessment method in this application.

[0016] Figure 2 This is a schematic diagram of rock mass stress in an embodiment of this application;

[0017] Figure 3 This is a schematic diagram of formation displacement in an embodiment of this application;

[0018] Figure 4 This is a schematic diagram of one embodiment of the integrated air-ground geological environment monitoring and assessment system in this application. Detailed Implementation

[0019] This application provides an integrated sky-ground geological environment monitoring and assessment method and system. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0020] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the integrated air-ground geological environment monitoring and assessment method in this application includes:

[0021] Step S101: Integrate the pre-collected sky remote sensing parameters, surface monitoring parameters and underground detection data into the sky-ground integrated monitoring matrix, and generate the original geological environment database after layered outlier processing;

[0022] Step S102: Construct an integrated space-air-ground spatiotemporal standard field based on the original geological environment database, and generate a geological environment correlation spectrum through spatiotemporal registration and data fusion;

[0023] Step S103: Perform multidimensional feature decomposition on the geological environment correlation spectrum and generate geological element feature chains through sky-ground collaborative analysis;

[0024] Step S104: Construct a dynamic field coupled between the sky and the ground based on the geological element characteristic chain, and obtain a set of geological environment evolution parameters through multi-field coupling analysis;

[0025] Step S105: Establish an integrated air-ground risk assessment matrix based on the geological environment evolution parameter set, and generate a risk warning indicator system through quantitative hierarchical analysis;

[0026] Step S106: Based on the risk warning indicator system and combined with the historical case database, design a sky-ground collaborative solution to generate a geological environment governance plan.

[0027] It is understood that the executing entity of this application can be an integrated air-ground geological environment monitoring and assessment system, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiment uses a server as an example for illustration.

[0028] Specifically, three types of basic data are collected: aerial remote sensing parameters, including surface deformation, geological structure information, and land cover data; surface monitoring parameters, including surface crack width, surface displacement, rainfall, and groundwater level; and underground detection data, including rock mass stress, stratum displacement, and groundwater pressure. For example... Figure 2 The diagram shown is a schematic representation of rock mass stress in an embodiment of this application. Figure 3 The diagram shown illustrates the stratigraphic displacement in this embodiment. These data, after standardization, form an initial integrated sky-ground monitoring matrix. Hierarchical outlier processing is performed on the data in the monitoring matrix. By calculating the mean and standard deviation of each layer, a threshold of 3 times the standard deviation is set to filter out outlier data points, and the processed data is stored in the original geological environment database. When constructing an integrated sky-ground spatiotemporal standard field based on the original geological environment database, the timestamps and spatial coordinates of the data are extracted to establish a unified spatiotemporal reference grid. Data within each time window is standardized, and the correlation coefficients between different monitoring points are calculated to form a spatiotemporal correlation matrix. Spatiotemporal registration eliminates data differences caused by different sampling frequencies and fills in data gaps. The registered data is grouped according to geological element attributes, and the correlation strength between elements within the group is calculated to construct a correlation strength matrix. Hierarchical clustering and spectral analysis are performed on the correlation strength matrix to generate a geological environment correlation spectrum.

[0029] In the process of multidimensional feature decomposition of the geological environment correlation spectrum, the correlation spectrum data is subspaced according to time series to form a time-varying feature subspace. Singular value decomposition is performed on the time-varying feature subspace to extract the main eigenvectors and generate a feature basis matrix. The feature basis matrix is ​​reorganized according to spatial location information to construct a spatial feature field. Tensor decomposition is performed on the spatial feature field and the time eigenvectors to extract spatiotemporal coupling features. The spatiotemporal coupling features are classified and aggregated according to geological element types to establish an element correlation matrix. Through hierarchical decomposition and feature reconstruction, a geological element feature chain is formed. When constructing a sky-ground coupled dynamic field based on the geological element feature chain, the feature chain is partitioned according to spatial distribution characteristics, and time series analysis is performed on the feature data of each region to form a regional dynamic feature matrix. Spatial interpolation and time series decomposition are performed on the regional dynamic feature matrix to extract periodic and trend components. The changing feature field is classified according to geological element types, and the coupling coefficients between different elements are calculated to form a sky-ground coupled dynamic field. Spatiotemporal decomposition is performed on each component of the coupled dynamic field to extract the inter-field interaction relationships and establish a coupling strength matrix. The dominant coupled modes are extracted by eigenvalue decomposition, an evolutionary feature space is constructed, and a set of geological environment evolution parameters is generated.

[0030] In establishing an integrated air-ground risk assessment matrix based on geological environment evolution parameter sets, the parameter sets are divided into multiple assessment units according to spatial distribution characteristics. The parameters of each assessment unit are normalized to form a standardized parameter matrix. Weight coefficients between parameters are calculated to construct parameter weight vectors. These weight vectors are then combined with the standardized parameter matrix to generate a risk assessment matrix. Eigenvalue decomposition is performed on the risk assessment matrix to extract key risk factors and establish a risk measurement space. This risk measurement space is then divided into multiple levels, and threshold ranges for each interval are calculated to form a risk warning indicator system. When designing air-ground collaborative solutions based on this risk warning indicator system, the indicator system is categorized according to risk levels, and the indicator characteristics for each level are extracted. The governance scheme data in the historical case database are organized by attribute, and a scheme feature matrix is ​​constructed. Risk feature sequences are matched with the scheme feature matrix through similarity calculation to generate an initial scheme set. Parameter optimization is performed on the initial scheme set to form an air-ground collaborative governance strategy. This collaborative governance strategy is then partitioned according to spatial distribution, and the governance parameters for each partition are adjusted to establish a partitioned governance matrix. Through multi-level analysis and parameter combination, geological environment governance schemes are generated.

[0031] For example, when conducting geological environment monitoring in a certain area, InSAR remote sensing data from the sky is collected to obtain surface deformation information. Displacement sensors and crack monitoring instruments are deployed on the surface, and stress-strain monitoring equipment is installed underground. These data are then registered in time and space and processed for outliers to form a unified monitoring dataset. Multidimensional feature decomposition is performed on the data to extract key geological element features and establish a coupled dynamic field. By analyzing the data correlation between different monitoring points, the coupling coefficients between various parameters are calculated to identify potential risk areas. Based on risk warning indicators and historical management experience, targeted prevention and control plans are developed, including measures such as surface reinforcement and underground support.

[0032] In this embodiment, by integrating pre-collected sky remote sensing parameters, surface monitoring parameters, and underground detection data into an integrated sky-ground monitoring matrix, collaborative acquisition of multi-source data is achieved, ensuring the comprehensiveness and systematic nature of data acquisition. The original geological environment database generated after hierarchical outlier processing has high data quality. Based on the original geological environment database, an integrated sky-ground spatiotemporal standard field is constructed. Through spatiotemporal registration and data fusion, a geological environment correlation spectrum is generated, achieving deep fusion and spatiotemporal alignment of multi-source heterogeneous data. The geological environment correlation spectrum is decomposed into multi-dimensional features, and a geological element feature chain is generated through sky-ground collaborative analysis. Key features of geological elements are extracted, and the correlation between elements is established. Based on the geological element feature chain, a sky-ground coupled dynamic field is constructed. Through multi-field coupling analysis, a geological environment evolution parameter set is obtained, achieving accurate characterization of the dynamic evolution characteristics of the geological environment. An integrated sky-ground risk assessment matrix is ​​established based on the geological environment evolution parameter set. The risk warning indicator system generated through quantitative hierarchical analysis has strong scientific validity and operability. Based on the risk warning indicator system, combined with a historical case database, a sky-ground collaborative scheme is designed, and the generated geological environment governance scheme has strong pertinence and practicality. This method organically integrates monitoring data from the sky, surface, and underground, establishing a complete data processing, analysis, evaluation, and governance process, which significantly improves the accuracy and reliability of geological environment monitoring and assessment.

[0033] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0034] (1) The sky remote sensing parameters are divided into three data subsets: surface deformation, surface cover and geological structure. Each data subset is standardized to form a sky layer data matrix.

[0035] (2) Classify the surface monitoring parameters according to point and surface elements, perform spatial interpolation on the point element data, and perform grid resampling on the surface element data to generate a surface layer data matrix.

[0036] (3) The underground exploration data is layered according to depth, and the parameters of the data at different depths are normalized and spatially registered to construct an underground layer data matrix.

[0037] (4) Time series alignment of the sky layer data matrix, the surface layer data matrix and the underground layer data matrix, and generation of the sky-ground integrated monitoring matrix through three-dimensional tensor operation;

[0038] (5) Perform hierarchical statistical analysis on each monitoring parameter in the integrated sky-ground monitoring matrix, set an outlier threshold, and remove outlier data points;

[0039] (6) After removing abnormal data points, the integrated sky-ground monitoring matrix is ​​processed by data interpolation and smoothing to generate the original geological environment database.

[0040] Specifically, the data is divided into three basic subsets: surface deformation variables, including quantitative indicators such as surface subsidence, uplift, and displacement; surface cover variables, including surface characteristic data such as vegetation cover, proportion of bare surface, and water area; and geological structural variables, including structural element information such as fault strike, dip angle, and joint density. These three types of data are standardized using a maximum-minimum standardization method to uniformly transform data of different dimensions into the [0,1] interval. The processed data is organized into a sky-layer data matrix. The processing of surface monitoring parameters distinguishes between point and surface feature characteristics. Point feature data mainly includes discrete monitoring data such as GPS monitoring points, tilt measurement points, and ground fissure monitoring points. Kriging spatial interpolation is used to process these point feature data. By calculating spatial autocorrelation and variability functions, the attribute values ​​of unknown points are estimated, realizing the transformation from point data to continuous surface data. The surface feature data includes continuously distributed data such as InSAR deformation data and surface deformation monitoring zones. These data are resampled into grids to unify data with different resolutions to the same spatial resolution, forming regular grid data, which are then integrated into a surface layer data matrix.

[0041] Subsurface detection data processing is layered according to depth, typically dividing the subsurface space into shallow (0-50m), intermediate (50-200m), and deep (>200m) layers. Monitoring data for each layer, including parameters such as borehole stress, rock displacement, and groundwater level, are normalized. Normalization uses the Z-score standardization method to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1. Simultaneously, spatial registration is performed to unify monitoring data from different depths and locations under the same spatial reference frame, constructing a subsurface layer data matrix. Time series alignment of the three-layer data matrices is then performed. The timestamps of all data are unified to the same time base; data with different sampling frequencies are resampled; high-frequency data is downsampled; or low-frequency data is interpolated to ensure all data have the same time interval. Finally, the time-aligned three-layer data matrices are combined through three-dimensional tensor operations to form an integrated sky-ground monitoring matrix. The tensor operation process includes tensor decomposition and reconstruction, preserving the spatiotemporal correlation characteristics between data.

[0042] A hierarchical statistical analysis was performed on the resulting integrated sky-ground monitoring matrix, calculating the mean, standard deviation, skewness, kurtosis, and other statistical characteristics of the data in each layer. An outlier threshold was set based on these characteristics, typically using the mean ± 3 times the standard deviation as the criterion. Data points exceeding this threshold were marked and removed. Finally, the data matrix after outlier removal underwent data imputation and smoothing. Data imputation employed a spatiotemporal interpolation method, comprehensively considering the temporal continuity and spatial correlation of the data to reasonably estimate missing values. Data smoothing used a moving average method to eliminate random fluctuations in the data, forming the original geological environment database.

[0043] For example, in areas prone to geological disasters, surface deformation data is acquired through satellite remote sensing, displacement data is acquired through ground monitoring points, and stress data is acquired through deep boreholes. The deformation data acquired through remote sensing is first classified, extracting information such as surface subsidence and surface crack distribution, and then standardized to form a matrix. Displacement data from ground monitoring points is spatially interpolated to generate a continuous distribution field. Stress data from deep boreholes is stratified by depth and then normalized. These three types of data are aligned according to time series, such as unifying monitoring data from different frequencies to a daily sampling frequency, forming a unified monitoring matrix. Outliers are identified through statistical analysis, such as a sudden, abnormally large displacement value at a monitoring point, and are then removed. Finally, the data is smoothed to obtain the monitoring dataset.

[0044] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0045] (1) Extract timestamp information and spatial location information from the original geological environment database, establish a spatiotemporal reference grid, and map discrete data points to a unified spatiotemporal coordinate system;

[0046] (2) The monitoring data in the spatiotemporal reference grid is segmented according to time windows, and the data in each time window is standardized to form a time series dataset;

[0047] (3) Construct a spatiotemporal correlation matrix based on the time series dataset, calculate the correlation coefficient between different monitoring points, and generate an integrated sky-ground spatiotemporal standard field;

[0048] (4) Perform spatiotemporal registration on the integrated sky-ground spatiotemporal standard field, align data items with different sampling frequencies, and fill in data gaps.

[0049] (5) Group the registered data according to the geological element attributes, calculate the correlation strength between elements within the group, and construct the correlation strength matrix;

[0050] (6) Divide the correlation strength matrix into multiple feature sub-matrices according to the hierarchical relationship, perform spectral decomposition and frequency domain analysis on each feature sub-matrix, merge the decomposition results after hierarchical clustering, construct the correlation spectrum based on the clustering results and extract the spectral features to form the geological environment correlation spectrum.

[0051] Specifically, the timestamp information (accurate to the second) and spatial location information (including longitude, latitude, and elevation) of each monitoring point are extracted to establish a three-dimensional spatiotemporal reference grid. The spatiotemporal reference grid adopts a latitude-longitude projected coordinate system with a horizontal resolution of 0.001 degrees, a vertical resolution of 1 meter, and a temporal resolution of 1 hour. Spatial interpolation methods (including inverse distance weighting and Kriging interpolation) are used to map the discrete monitoring point data onto unified grid nodes. The monitoring data in the spatiotemporal reference grid are segmented according to fixed time windows (e.g., 24 hours), and the data within each time window are standardized. The standardization process uses the Z-score method to transform monitoring data of different dimensions into a unified standard normal distribution interval. The standardized data forms a time-series dataset containing standardized time series data for multiple monitoring indicators such as surface deformation, groundwater level, and rainfall.

[0052] When constructing a spatiotemporal correlation matrix based on a time-series dataset, the Pearson correlation coefficient between different monitoring points is calculated. For time-series data, a sliding window method is used to calculate the dynamic correlation coefficient, with a window length of 30 days and a step size of 1 day. The calculated correlation coefficient matrix reflects the degree of correlation between different monitoring points, thereby generating an integrated sky-ground spatiotemporal standard field.

[0053] For spatiotemporal registration, the following spatiotemporal registration model is adopted:

[0054]

[0055] Where: C align (t) represents the registered spatiotemporal data; X i (t) represents the space-based monitoring data sequence; Y j (t) represents the surface monitoring data sequence; Z k (t) represents the underground monitoring data sequence; α i β j γ k These represent the weight coefficients for different data sources; n, m, and p represent the number of the three types of data sources, respectively.

[0056] After registration, the data is grouped according to geological element attributes (such as topography, geological structure, hydrogeology, etc.), and the correlation strength between elements within each group is calculated. The correlation strength calculation uses grey relational analysis, quantifying the correlation between elements by calculating the grey relational degree between different time series. Each element in the correlation strength matrix represents the degree of correlation between two geological elements, with values ​​ranging from [0,1]. The correlation strength matrix is ​​divided into multiple feature sub-matrices according to hierarchical relationships, with each sub-matrix corresponding to a geological element category. Spectral decomposition is performed on the feature sub-matrices to extract key eigenvalues ​​and eigenvectors, and analysis is conducted in the frequency domain. Frequency domain analysis uses wavelet transform, selecting Morlet wavelets as the basis function to perform multi-scale decomposition of the time series. The decomposition results are categorized using hierarchical clustering, employing Ward's minimum variance method as the clustering algorithm, classifying elements into different categories based on the similarity of spectral features. Finally, a correlation spectrum is constructed based on the clustering results, extracting spectral features to form a geological environment correlation spectrum.

[0057] In the data processing example, monitoring data was extracted from the original database, including satellite InSAR deformation data (sampling interval of 12 days), ground GPS deformation monitoring data (sampling interval of 1 hour), and underground microseismic monitoring data (sampling interval of 10 minutes). These data with different spatiotemporal resolutions were unified to the same coordinate system using a spatiotemporal reference grid. For InSAR data, spatial interpolation methods were used to resample the 100-meter resolution deformation data onto a standard grid; for GPS data, point observation data were extended to the entire monitoring area using Kriging interpolation; for microseismic data, spatial locations were determined based on the seismic source location results, and the temporal resolution was unified to 1 hour.

[0058] During the spatiotemporal registration process, a registration model is used to align the sampled data at different frequencies. The correlation matrix between monitoring points is obtained by calculating the Pearson correlation coefficient. The correlation coefficient is calculated based on a sliding time window to obtain the dynamic correlation variation characteristics. In constructing the correlation spectrum, the monitoring data are first grouped according to geological element attributes, such as classifying deformation data, hydrological data, and tectonic data separately. The correlation strength between elements within each group is calculated to form a correlation strength matrix. Spectral decomposition and frequency domain analysis are performed on the correlation strength matrix to extract the main characteristic modes, and the features are classified and integrated using hierarchical clustering. The resulting correlation spectrum reflects the correlation relationships and evolutionary characteristics between different geological elements.

[0059] In one specific embodiment, the process of executing step S103 may specifically include the following steps:

[0060] (1) The geological environment correlation spectrum is segmented according to the time series, and the correlation spectrum data in each time period is divided into subspaces to form a time-varying feature subspace;

[0061] (2) Perform singular value decomposition on the data in the time-varying feature subspace, extract the main feature vectors, and generate the feature basis matrix;

[0062] (3) Reorganize the feature basis matrix according to the spatial location information to construct a spatial feature field and extract spatial feature components;

[0063] (4) Perform tensor decomposition on spatial feature components and temporal feature vectors to extract spatiotemporal coupling features;

[0064] (5) Classify and aggregate the spatiotemporal coupling features according to the geological element types, and establish an element correlation matrix;

[0065] (6) Perform hierarchical decomposition and feature reconstruction on the element correlation matrix, organize the reconstructed features in a chain according to the sky-ground synergy relationship, perform multi-level feature mapping and correlation analysis, and form a geological element feature chain through feature transfer and feature fusion.

[0066] Specifically, the correlation spectral data is segmented according to fixed time intervals (e.g., 30 days), and subspace partitioning is performed on each segment. Principal component analysis is used for subspace partitioning, projecting high-dimensional data onto a low-dimensional feature space to form a time-varying feature subspace. This time-varying feature subspace contains the variation characteristics of different monitoring elements (such as surface deformation, groundwater level, rock mass stress, etc.) over different time periods.

[0067] Singular value decomposition is performed on the data in the time-varying feature subspace. Its mathematical expression is:

[0068]

[0069] Where: D(t) represents the time-varying feature subspace data matrix; U(t) represents the left singular matrix containing the time eigenvectors; Λ(t) represents the singular value diagonal matrix; V(t) represents the right singular matrix containing the spatial eigenvectors; μ i ω represents the weight coefficient of the i-th feature mode; i (t) represents the i-th time characteristic function; Let represent the i-th spatial feature function; r represents the number of selected feature modes.

[0070] The eigenvalue basis matrix obtained after singular value decomposition contains the main spatiotemporal variation features. Important eigenvectors are selected by the magnitude of the eigenvalues. The eigenvalue basis matrix is ​​then reorganized according to the spatial location information of the monitoring points to construct a spatial feature field. The spatial feature field reflects the correlation characteristics of different monitoring elements in spatial distribution.

[0071] The tensor decomposition of spatial feature components and temporal feature vectors is expressed mathematically as follows:

[0072]

[0073] Where: T(x,y,z,t) represents the four-dimensional spacetime tensor; ψ p (x) represents the spatial basis functions in the horizontal direction; θ q (y,z) represents the spatial basis functions in the vertical direction; η k (t) represents the time basis function; ρ pqk The coefficients represent the tensor decomposition coefficients; P, Q, and K represent the number of basis functions in the three directions, respectively.

[0074] The spatiotemporal coupling features extracted by tensor decomposition are classified and aggregated according to geological element types (such as structure, hydrology, lithology, etc.) to establish an element correlation matrix. Each element in the element correlation matrix represents the degree of correlation between different geological elements. The element correlation matrix is ​​then hierarchically decomposed using a multi-level network analysis method, organizing the geological elements into three levels: sky, surface, and subsurface. During feature reconstruction, expert experience and knowledge are incorporated to determine the correlation weights between elements at different levels, achieving chain-like transmission and fusion of features to form a geological element feature chain.

[0075] For example, the monitoring data includes multi-source data such as surface deformation, ground tilt monitoring, and underground displacement monitoring from satellite InSAR monitoring. The year's monitoring data is divided into 12 data blocks with 30-day time windows. Subspace partitioning is performed on each data block to extract the main change features. Singular value decomposition yields temporal and spatial feature vectors; analysis shows that the first three feature modes account for the majority of the total variance. The feature vectors are recombined to construct a spatial feature field, reflecting the spatial distribution of deformation. Subsequently, tensor decomposition is performed on the four-dimensional spatiotemporal data to obtain basis functions in the horizontal, vertical, and temporal directions. Based on the decomposition results, the monitoring elements are organized into three levels: space-based InSAR monitoring, ground-based GPS monitoring, and underground displacement monitoring, constructing a multi-level feature transmission network. By calculating the inter-layer correlation strength, feature transmission paths are determined, forming a complete geological feature chain. This feature chain fully reflects the deformation characteristics and evolution process of the landslide body from deep to shallow.

[0076] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0077] (1) Divide the geological element feature chain into regions according to spatial distribution characteristics, perform time-series analysis on the feature data of each region, and form a regional dynamic feature matrix;

[0078] (2) Spatial interpolation and time series decomposition are performed on the regional dynamic feature matrix to extract periodic and trend change components and construct a change feature field;

[0079] (3) Classify the changing characteristic field according to the geological element type, calculate the coupling coefficient between different elements, and generate a dynamic field of sky-ground coupling.

[0080] (4) Perform spatiotemporal decomposition on each component of the sky-ground coupled dynamic field, extract the interaction relationship between the fields, and establish the coupling strength matrix;

[0081] (5) Perform eigenvalue decomposition on the coupling strength matrix, extract the dominant coupling modes, and construct the evolution feature space;

[0082] (6) The evolution feature space is decomposed and reorganized at multiple levels, key evolution features are extracted and combined in time and space. Through coupling relationship transmission and feature association analysis, the feature combination is parameterized according to the sky-ground coupling relationship to generate a set of geological environment evolution parameters.

[0083] Specifically, based on the geological element characteristic chain and combined with the spatial distribution characteristics of topography, geological structure, etc., a spatial clustering algorithm is used to divide the monitoring area into several sub-regions. The spatial clustering uses the density-based DBSCAN algorithm, which groups spatially close and similarly characterized areas into one class by calculating the Euclidean distance between monitoring points and setting a density threshold. Time-series analysis is performed on the feature data of each sub-region, extracting time-series features based on the sliding time window method to form a regional dynamic feature matrix. For the regional dynamic feature matrix, Kriging interpolation is used to grid the spatially discrete point data to obtain continuous spatial distribution characteristics. Time-series decomposition is performed on the gridded data, using wavelet analysis to decompose the time-series data into periodic variation components and trend variation components. The periodic variation components reflect seasonal and diurnal cycle patterns, while the trend variation components reflect long-term evolutionary trends. A change feature field is constructed based on the decomposition results, which includes change features at different time scales.

[0084] The change characteristic field is classified according to geological element types (such as topographic change, hydrological change, tectonic change, etc.), and the coupling coefficient between different types of elements is calculated using canonical correlation analysis. The coupling coefficient quantifies the interaction strength between different elements, and a dynamic field of space-ground coupling is generated by constructing a coupling relationship network. This dynamic field integrates space-based remote sensing, surface monitoring, and subsurface exploration data, reflecting the collaborative characteristics of multi-source data.

[0085] The spatiotemporal decomposition of the coupled dynamic field between the sky and the ground is expressed mathematically as follows:

[0086]

[0087] Where: F(x,y,z,t) represents a four-dimensional coupled dynamic field; χ i (x,y) represents the characteristic function of the horizontal surface; ζj (z) represents the characteristic function in the vertical direction; ν k (t) represents the time characteristic function; ξ ijk The coupling strength coefficient is represented by m, n, and l, which represent the number of characteristic functions in the three directions, respectively.

[0088] By decomposing and extracting inter-field interactions, the coupling strength between different monitoring levels is quantified, and a coupling strength matrix is ​​established. Eigenvalue decomposition is performed on the coupling strength matrix to extract dominant coupling modes. These dominant coupling modes reflect the most significant coupling mechanisms, and an evolutionary feature space is constructed based on these modes. The evolutionary feature space is then decomposed into multiple levels using multi-scale analysis. Key evolutionary features are extracted at each scale level, and the correlations between features are determined through spatiotemporal combination analysis. Information transfer and fusion between features at different levels are achieved through the transmission of coupling relationships. Based on the sky-ground synergy, the feature combinations are parameterized to generate a set of geological environment evolution parameters.

[0089] For example, in geological environment monitoring in mining areas, based on InSAR deformation data, surface GPS monitoring data, and underground microseismic monitoring data, the monitoring area is divided into three sub-regions: the goaf area, the surrounding influence area, and the stable area. Time series analysis is performed on each sub-region to extract time series of characteristic parameters such as surface subsidence, groundwater level, and microseismic frequency. Kriging interpolation is used to extend the discrete monitoring data to the entire area, and wavelet analysis is used to separate annual seasonal variations and long-term trend variations. Canonical correlation coefficients are calculated for different types of monitoring data to establish a space-air-ground coupled dynamic field. Spatiotemporal decomposition extracts the main coupled modes, revealing significant interactions between surface subsidence, groundwater level changes, and microseismic activity. Based on these coupling relationships, a multi-level evolutionary feature space is constructed, forming an evolutionary parameter set containing parameters such as the rate of geological environment evolution, trends, and interaction strength.

[0090] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0091] (1) Divide the set of geological environment evolution parameters into multiple assessment units according to spatial distribution characteristics, and normalize the parameters of each assessment unit to form a standardized parameter matrix.

[0092] (2) Perform correlation analysis on the standardized parameter matrix, calculate the weight coefficients between parameters, and construct the parameter weight vector;

[0093] (3) Combine the parameter weight vector with the standardized parameter matrix to generate the integrated air-ground risk assessment matrix;

[0094] (4) Perform eigenvalue decomposition on the integrated air-ground risk assessment matrix, extract the main risk factors, and establish a risk measurement space;

[0095] (5) Divide the risk measurement space into multiple risk level intervals, calculate the threshold range of each interval, and form a hierarchical standard system;

[0096] (6) Conduct multi-level combination analysis of the graded standard system, determine the risk transmission relationship through risk factor correlation analysis, establish early warning level classification criteria, divide risks into zones based on spatial distribution characteristics, determine the risk level of each zone and set early warning thresholds, and generate a risk early warning indicator system.

[0097] Specifically, the geological environment evolution parameters include multiple evolution parameters such as surface deformation rate, groundwater level change, and rock mass stress change. Based on spatial distribution characteristics (e.g., topographic and geomorphic units, geological structural zones, hydrogeological zones, etc.), the monitoring area is divided into multiple assessment units using the geographic unit method. Parameters in each assessment unit are normalized using the maximum-minimum standardization method, unifying parameters of different dimensions into the [0,1] interval to form a standardized parameter matrix. Correlation analysis is performed on the standardized parameter matrix, using the Pearson correlation coefficient method to calculate the correlation between parameters. Based on the correlation analysis results, the analytic hierarchy process (AHP) is used to calculate parameter weights. The AHP method establishes a judgment matrix, determines the importance ratio between parameters through expert scoring, and then calculates eigenvalues ​​and eigenvectors to obtain the weight coefficients of each parameter, constructing a parameter weight vector.

[0098] The parameter weight vector and standardized parameter matrix are combined and calculated using a weighted summation method to obtain the comprehensive risk value for each assessment unit, generating an integrated sky-ground risk assessment matrix. This risk assessment matrix reflects the degree of geological environmental risk at different locations within the monitoring area. Eigenvalue decomposition is performed on the risk assessment matrix, and principal component analysis (PCA) is used to extract the main risk factors. During PCA, the main eigenvectors are selected based on their cumulative contribution rate; these eigenvectors form the basis of the risk measurement space. The risk measurement space is a multidimensional feature space, and its dimension is determined by the number of principal components selected.

[0099] Based on historical monitoring data and expert experience, the risk measurement space is divided into multiple risk level intervals. Cluster analysis methods (such as K-means clustering) are used to group risk values ​​and determine the threshold range for each level interval. The grading standard system includes quantitative boundaries for different risk levels. A multi-level combination analysis is conducted on the grading standard system to construct a risk indicator system, including multiple levels such as surface deformation indicators, groundwater indicators, and geological structure indicators. Correlation analysis is used to determine the relationships between indicators at different levels, establishing a risk transmission network. Based on the risk transmission network, early warning level classification criteria are set, dividing risk levels into four levels: red (dangerous), orange (alert), yellow (caution), and green (normal). Combining spatial distribution characteristics, a regional zoning method is used to divide the monitoring area into different risk zones, and a corresponding early warning threshold is set for each zone.

[0100] For example, in open-pit mines, the area is divided into assessment units based on factors such as stope slopes, deposits, and surrounding impact zones. Monitoring data for each unit (including surface displacement monitored by InSAR, deformation rate monitored by GPS, and groundwater level observations) are normalized. The weights of each parameter are determined using the analytic hierarchy process (AHP), such as surface displacement weight 0.4, deformation rate weight 0.3, and groundwater level weight 0.3. The calculated risk assessment matrix is ​​then subjected to principal component analysis, extracting three main risk factors: cumulative deformation, water level fluctuation, and tectonic activity. A risk measurement space is constructed based on these risk factors, and cluster analysis is used to classify risk values ​​into different level intervals. Risk transmission analysis reveals a significant correlation between surface deformation and groundwater level changes, while tectonic activity affects both. Based on this, a tiered early warning indicator system is established: an orange alert is triggered when the annual cumulative surface deformation exceeds a set threshold and groundwater level fluctuations are abnormal; a red alert is activated when tectonic activity is frequent and accelerated surface deformation occurs. This early warning system enables accurate identification and timely warning of different risk levels.

[0101] In one specific embodiment, the process of executing step S106 may specifically include the following steps:

[0102] (1) Classify the risk warning indicator system according to the risk level, extract the features of the indicators under each risk level, and form a risk feature sequence;

[0103] (2) Organize the attributes of governance scheme data in the historical case database, extract key governance parameters, and construct a scheme feature matrix;

[0104] (3) Calculate the similarity between the risk feature sequence and the scheme feature matrix, select historical cases with high matching degree, and generate an initial scheme set;

[0105] (4) Optimize the parameters of the initial set of schemes, combine schemes for different governance objectives, and form a sky-ground collaborative governance strategy;

[0106] (5) Divide the sky-ground collaborative governance strategy into zones according to spatial distribution, adjust the governance parameters of each zone, and establish a zone governance matrix;

[0107] (6) Conduct multi-level analysis and parameter combination of the regional governance matrix, optimize the governance parameters in multiple dimensions of sky and ground, combine risk warning level and governance effect evaluation, dynamically adjust and optimize the governance parameters, construct the governance process sequence, and generate geological environment governance plan.

[0108] Specifically, based on the warning levels (red, orange, yellow, and green), features are extracted from indicators (including surface deformation rate, groundwater level change rate, and tectonic activity intensity) at each level. Feature extraction employs statistical analysis methods to calculate the mean, standard deviation, and coefficient of variation for each indicator, and combines this with time-series analysis to extract trend and periodic features, forming a risk feature sequence. The governance scheme data in the historical case database is systematically organized to establish a data structure containing multi-dimensional attributes such as governance methods, engineering parameters, and implementation conditions. Key governance parameters are extracted, including quantitative indicators such as project scale, construction technology, and material parameters, as well as qualitative indicators such as geological conditions and environmental constraints. Through data standardization and coding transformation, these parameters are organized into a structured scheme feature matrix.

[0109] A similarity calculation method was used to match and analyze the risk feature sequence and the scheme feature matrix. The similarity calculation combined cosine similarity and Euclidean distance, considering both quantitative and qualitative indicators. By setting a similarity threshold, historical cases with high matching degrees were selected to form an initial scheme set. Parameter optimization of the initial scheme set was then performed using a multi-objective optimization method, taking governance effect, engineering cost, and environmental impact as optimization objectives. For different governance objectives (such as slope stability, groundwater remediation, and surface subsidence control), the scheme parameters were combined and optimized. The combination optimization process employed a genetic algorithm, generating new scheme combinations through crossover and mutation operations, gradually optimizing the scheme parameters to form a sky-ground collaborative governance strategy.

[0110] The integrated air-ground governance strategy is divided into zones based on spatial distribution characteristics, considering factors such as topography, engineering geology, and hydrogeology, thus dividing the governance area into multiple sub-regions. Governance parameters for each zone are adjusted accordingly, including engineering layout, construction parameters, and monitoring point deployment, establishing a zoned governance matrix. Multi-level analysis is performed on the zoned governance matrix, using the analytic hierarchy process (AHP) to assess the importance and priority of different governance measures. Governance parameters are dynamically adjusted based on risk warning levels. A rolling optimization strategy is employed for parameter optimization, dynamically adjusting the optimal parameter combinations based on real-time evaluation results of the governance effects. A governance process sequence including construction steps, parameter configuration, and monitoring plans is constructed, forming a complete geological environment governance plan.

[0111] For example, in landslide mitigation, risk warning indicators (including surface displacement rate, groundwater level, and crack development level) are categorized according to warning levels, and the spatiotemporal variation characteristics of each indicator are extracted. Historical case databases are consulted to extract key parameters for mitigation schemes such as anti-slide piles, drainage works, and anchoring works. Similarity calculations are used to identify historical cases with similar geological conditions and deformation characteristics. Initial schemes are optimized and combined, such as combining anti-slide piles with drainage works and optimizing pile placement and drainage network layout. Based on site conditions, the mitigation area is divided into primary landslide zone, secondary landslide zone, and affected zone, and mitigation measures are formulated for each. Anti-slide pile arrays are deployed and deep wells are installed for drainage in the primary landslide zone; anchor frame beam reinforcement is implemented in the secondary landslide zone; and a monitoring network is established in the affected zone. The mitigation effect is continuously monitored and evaluated, and construction parameters are dynamically adjusted.

[0112] In one specific embodiment, the process of performing parameter optimization on the initial scheme set may specifically include the following steps:

[0113] (1) Group the initial scheme set according to the type of governance objective, and standardize the parameters of each scheme to form a standard parameter set;

[0114] (2) Perform correlation analysis on the parameters in the standard parameter set, calculate the correlation strength between each parameter, and construct a parameter correlation network;

[0115] (3) The parameter association network is layered according to the governance objectives, and the key parameter combinations of each layer are extracted to generate a multi-objective parameter space;

[0116] (4) Set constraints on the multi-objective parameter space, determine the feasible region through parameter boundary analysis, and form the parameter optimization interval;

[0117] (5) Combine and match the parameters within the parameter optimization range, filter them according to the target priority, and establish a sequence of combination schemes;

[0118] (6) Conduct multi-dimensional evaluation of the combination sequence of schemes, dynamically adjust the parameter combination through parameter sensitivity analysis and scheme effect evaluation, optimize and integrate the schemes in combination with the sky-ground coordination relationship, and form a sky-ground coordination governance strategy.

[0119] Specifically, the schemes are grouped according to the type of governance objective (e.g., surface deformation control, groundwater remediation, geological structure reinforcement, etc.). For each group, the engineering parameters (e.g., support structure dimensions, grouting pressure, anchor bolt length, etc.) are standardized by maximizing and minimizing their values, unifying parameters of different dimensions into the [0,1] interval to form a standard parameter set. Correlation analysis is performed on the parameters in the standard parameter set, using the Pearson correlation coefficient method to calculate the correlation strength between parameters. By setting a correlation coefficient threshold, parameter pairs with significant correlations are selected to construct a parameter correlation network. The parameter correlation network is represented by a graph structure, where nodes represent parameters, edges represent the correlation relationships between parameters, and edge weights represent the correlation strength.

[0120] The parameter association network is hierarchically processed according to governance objectives. A community discovery algorithm is used to identify parameter communities in the network, with each community corresponding to a governance objective level. Key parameter combinations are extracted at each level, with the selection of key parameters based on centrality analysis, including degree centrality, betweenness centrality, and eigenvector centrality. By combining these key parameters, a multi-objective parameter space is generated. Constraints are set for the multi-objective parameter space, including engineering and technical constraints (such as construction process limitations and equipment capacity limitations), resource constraints (such as material supply and construction period), and environmental constraints (such as terrain and climate conditions). Through parameter boundary analysis, the value range of each parameter is determined, forming a parameter optimization interval.

[0121] Within the parameter optimization range, parameter combination matching is performed, and a non-dominated solution set is generated using a multi-objective optimization algorithm (such as the NSGA-II algorithm). Based on the priority of governance objectives (e.g., safety first, economy second), schemes are screened and ranked to establish a scheme combination sequence. The scheme combination sequence is evaluated from multiple dimensions, including technical feasibility assessment, economic rationality assessment, and environmental impact assessment. Parameter sensitivity analysis is used to identify key parameters that significantly affect the governance effect. Orthogonal experimental design is used to analyze the impact of different parameter combinations on the governance effect. Combining multi-source monitoring data from the sky, ground, and air, the parameter combinations are dynamically optimized and adjusted to form a sky-ground collaborative governance strategy.

[0122] For example, in karst collapse remediation, solutions including grouting, pile foundation reinforcement, and drainage are grouped according to remediation objectives. Grouting parameters (such as grout mix ratio, grouting pressure, and grout volume), pile foundation parameters (such as pile length, pile diameter, and pile spacing), and drainage parameters (such as pipeline layout and sump depth) are standardized. Correlation analysis reveals significant correlations between grouting pressure and grout diffusion radius, and between pile spacing and foundation bearing capacity. The parameter correlation network is divided into three levels: surface reinforcement layer, underground filling layer, and base reinforcement layer, and key parameter combinations are extracted. Considering constraints such as soil bearing capacity and construction equipment capacity, feasible parameter value ranges are determined. A multi-objective optimization method is used to generate a series of parameter combination schemes, and sensitivity analysis is used to determine the optimal parameter configuration, forming a comprehensive remediation strategy.

[0123] In one specific embodiment, the process of partitioning the sky-ground collaborative governance strategy according to spatial distribution may specifically include the following steps:

[0124] (1) The parameter data in the sky-ground collaborative governance strategy are processed into a grid according to spatial coordinates to generate a parameter spatial distribution map;

[0125] (2) Perform spatial autocorrelation analysis on the parameter spatial distribution map, identify the parameter variation characteristics, and form a spatial variation field;

[0126] (3) Divide the spatial variation field into regions according to parameter similarity, classify and statistically analyze the parameters in each region, and generate a partition parameter sequence;

[0127] (4) Perform boundary optimization on the partition parameter sequence to eliminate parameter abrupt changes between regions and construct a smooth transition zone;

[0128] (5) Perform joint analysis of the parameters of the smooth transition zone with the parameters inside the region, calculate the regional characteristic values, and generate the regional characteristic spectrum;

[0129] (6) Spatial overlay and correlation analysis of regional feature spectra are performed. Through parameter partitioning optimization and regional parameter adjustment, combined with spatial continuity analysis, the partition boundaries and internal parameters are dynamically corrected to form a partition governance matrix.

[0130] Specifically, the construction of the zonal governance matrix involves spatial processing of the parameter data in the sky-ground collaborative governance strategy. A regular grid method is used to divide the monitoring area into grid cells, with the grid size determined based on the monitoring accuracy. Interpolation calculations are performed on the governance parameters (including engineering parameters, monitoring parameters, and control parameters) within each grid cell. The Kriging interpolation method is used to extend the discrete parameter data to the entire spatial domain, generating a parameter spatial distribution map. Spatial autocorrelation analysis is performed on the parameter spatial distribution map, and the Moran index method is used to calculate the spatial correlation of the parameters. By constructing a spatial weight matrix, the parameter correlation degree between each grid cell and its neighboring cells is calculated. Based on variogram theory, the spatial variation patterns of the parameters are analyzed, including variation directionality, variation scale, and spatial structural characteristics, forming a spatial variation field. The spatial variation field reflects the variation characteristics of the parameters in their spatial distribution.

[0131] The spatial variation field is divided into regions based on parameter similarity, and fuzzy C-means clustering is used to classify the parameters. By setting cluster centers and fuzziness coefficients, grid cells with similar parameter characteristics are grouped into the same region. Statistical analysis is performed on the parameters within each region, calculating statistics such as mean and standard deviation to form a partitioned parameter sequence. This parameter sequence reflects the parameter characteristics of different regions. Boundary optimization is performed on the partitioned parameter sequence, using a boundary smoothing algorithm to handle abrupt parameter changes between regions. Transition zones are set between adjacent regions, with the width determined based on the parameter gradient. A weighted average method is used to interpolate the parameters within the transition zone to achieve a smooth parameter transition. The transition zone avoids managing spatial discontinuities in the parameters.

[0132] The parameters of the smooth transition zone are jointly analyzed with the parameters within the region, and principal component analysis is used to extract regional features. By calculating eigenvalues ​​and eigenvectors, the main parameter characteristics of the region are determined. A feature spectrum is generated for each region, containing the parameter distribution characteristics and variation patterns. Spatial overlay analysis is performed on the regional feature spectra, combining the feature spectra from different regions using spatial overlay operations. Correlation analysis is used to identify parameter relationships between regions, and spatial continuity analysis is used to assess the coherence of parameter distribution. Based on the analysis results, the parameters of the partition boundaries and internal parameters are dynamically corrected to optimize parameter configuration and form a partition governance matrix.

[0133] For example, in the remediation of goaf areas in mining areas, the remediation strategy, which includes filling parameters, support parameters, and monitoring parameters, is gridded with a grid size of 10 meters × 10 meters. Kriging interpolation is used to generate a spatial distribution map of the parameters, reflecting the distribution of remediation parameters at different locations. Spatial autocorrelation analysis shows that the parameters have obvious directional characteristics, exhibiting different variation patterns along the strike and dip. Based on parameter similarity, the area is divided into three zones: goaf area, transition zone, and stable zone. The goaf area is primarily remediated with filling, the transition zone uses composite support, and the stable zone focuses on monitoring and early warning. A 20-meter-wide transition zone is set at the zone boundaries, using a weighted decreasing approach to achieve a smooth parameter transition. Principal component analysis is used to extract the feature spectra of each zone, reflecting the remediation characteristics of different areas. Finally, combined with spatial continuity analysis, the configuration of zone parameters is optimized to form a complete zone remediation matrix.

[0134] The above describes the integrated sky-ground geological environment monitoring and assessment method in the embodiments of this application. The following describes the integrated sky-ground geological environment monitoring and assessment system in the embodiments of this application. Please refer to [link / reference]. Figure 4 One embodiment of the integrated air-ground geological environment monitoring and assessment system in this application includes:

[0135] The acquisition module 201 is used to integrate pre-acquired sky remote sensing parameters, surface monitoring parameters and underground detection data into the sky-ground integrated monitoring matrix, and generate a raw geological environment database after hierarchical outlier processing.

[0136] The generation module 202 is used to construct an integrated space-air-ground spatiotemporal standard field based on the original geological environment database, and generate a geological environment correlation spectrum through spatiotemporal registration and data fusion.

[0137] Decomposition module 203 is used to perform multi-dimensional feature decomposition on the geological environment correlation spectrum and generate geological element feature chains through sky-ground collaborative analysis.

[0138] The coupling module 204 is used to construct a dynamic field of sky-ground coupling based on the geological element feature chain, and to obtain a set of geological environment evolution parameters through multi-field coupling analysis.

[0139] The hierarchical module 205 is used to establish an integrated air-ground risk assessment matrix based on the set of geological environment evolution parameters, and to generate a risk warning indicator system through quantitative hierarchical analysis.

[0140] Design module 206 is used to design a sky-ground collaborative solution based on the risk warning indicator system and combined with the historical case library, and generate a geological environment governance solution.

[0141] Through the collaborative efforts of the aforementioned components, by integrating pre-collected sky remote sensing parameters, surface monitoring parameters, and underground detection data into an integrated sky-ground monitoring matrix, collaborative acquisition of multi-source data was achieved, ensuring the comprehensiveness and systematic nature of data collection. The resulting geological environment raw database, after hierarchical outlier processing, possesses high data quality. Based on this raw geological environment database, an integrated sky-ground spatiotemporal standard field was constructed. Through spatiotemporal registration and data fusion, a geological environment correlation spectrum was generated, achieving deep fusion and spatiotemporal alignment of multi-source heterogeneous data. Multidimensional feature decomposition was performed on the geological environment correlation spectrum, and through sky-ground collaboration… This method analyzes and generates geological element feature chains, extracts key features of geological elements, and establishes the correlations between elements. Based on these feature chains, a space-air-ground coupled dynamic field is constructed. Through multi-field coupling analysis, a set of geological environment evolution parameters is obtained, enabling a precise characterization of the dynamic evolution of the geological environment. An integrated space-air-ground risk assessment matrix is ​​established based on this parameter set. The risk warning indicator system generated through quantitative hierarchical analysis demonstrates strong scientific validity and operability. Using this risk warning indicator system as a benchmark, and combining it with a historical case database, a space-air-ground collaborative solution is designed, resulting in a geological environment governance plan that is highly targeted and practical. This method organically integrates monitoring data from the sky, surface, and subsurface levels, establishing a complete data processing, analysis, assessment, and governance process, significantly improving the accuracy and reliability of geological environment monitoring and assessment.

[0142] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for integrated air-ground geological environment monitoring and assessment, characterized in that, The integrated air-ground geological environment monitoring and assessment method includes: The pre-collected sky remote sensing parameters, surface monitoring parameters and underground detection data are integrated into the sky-ground integrated monitoring matrix, and after layered outlier processing, a raw geological environment database is generated. Based on the original geological environment database, an integrated space-air-ground spatiotemporal standard field is constructed, and a geological environment correlation spectrum is generated through spatiotemporal registration and data fusion. A multidimensional feature decomposition of the geological environment correlation spectrum is performed, and a geological element feature chain is generated through sky-ground collaborative analysis. This includes: segmenting the geological environment correlation spectrum according to a time series; dividing the correlation spectrum data within each time period into subspaces to form time-varying feature subspaces; performing singular value decomposition on the data in the time-varying feature subspaces to extract main feature vectors and generate feature basis matrices; reorganizing the feature basis matrices according to spatial location information to construct a spatial feature field and extract spatial feature components; performing tensor decomposition on the spatial feature components and time feature vectors to extract spatiotemporal coupling features; classifying and aggregating the spatiotemporal coupling features according to geological element types to establish an element correlation matrix; performing hierarchical decomposition and feature reconstruction on the element correlation matrix; organizing the reconstructed features in a chain according to sky-ground collaborative relationships; performing multi-level feature mapping and correlation analysis; and forming the geological element feature chain through feature transfer and feature fusion. A dynamic field coupled between the sky and the ground is constructed based on the characteristic chain of geological elements, and a set of geological environment evolution parameters is obtained through multi-field coupling analysis; An integrated air-ground risk assessment matrix is ​​established based on the set of geological environment evolution parameters, and a risk warning indicator system is generated through quantitative hierarchical analysis. Based on the risk warning indicator system, and combined with the historical case database, a sky-ground collaborative solution is designed to generate a geological environment governance plan.

2. The integrated air-ground geological environment monitoring and assessment method according to claim 1, characterized in that, The process involves integrating pre-collected sky remote sensing parameters, surface monitoring parameters, and underground detection data into a sky-ground integrated monitoring matrix. After layered outlier processing, a raw geological environment database is generated, including: The sky remote sensing parameters are divided into three data subsets: surface deformation, surface cover, and geological structure. Each data subset is standardized to form a sky layer data matrix. The surface monitoring parameters are classified according to point and surface elements. Spatial interpolation is performed on the point element data, and grid resampling is performed on the surface element data to generate a surface layer data matrix. The underground exploration data is layered according to depth, and the parameters of the data at different depths are normalized and spatially registered to construct an underground layer data matrix. The sky layer data matrix, surface layer data matrix and underground layer data matrix are aligned with time series, and the integrated sky-ground monitoring matrix is ​​generated through three-dimensional tensor operations. Hierarchical statistical analysis is performed on each monitoring parameter in the integrated sky-ground monitoring matrix, anomaly thresholds are set, and abnormal data points are removed. The integrated sky-ground monitoring matrix, after removing outlier data points, undergoes data interpolation and smoothing to generate the original geological environment database.

3. The integrated air-ground geological environment monitoring and assessment method according to claim 1, characterized in that, The construction of an integrated sky-ground spatiotemporal standard field based on the original geological environment database, followed by spatiotemporal registration and data fusion to generate a geological environment correlation spectrum, includes: Timestamp information and spatial location information are extracted from the original geological environment database to establish a spatiotemporal reference grid, and discrete data points are mapped to a unified spatiotemporal coordinate system. The monitoring data in the spatiotemporal reference grid is segmented according to time windows, and the data in each time window is standardized to form a time series dataset; Based on the time-series dataset, a spatiotemporal correlation matrix is ​​constructed, the correlation coefficients between different monitoring points are calculated, and the integrated sky-ground spatiotemporal standard field is generated. Spatiotemporal registration is performed on the integrated sky-ground spatiotemporal standard field to align data items with different sampling frequencies and fill in missing data areas; The registered data are grouped according to geological element attributes, the correlation strength between elements within the group is calculated, and a correlation strength matrix is ​​constructed. The correlation strength matrix is ​​divided into multiple feature sub-matrices according to hierarchical relationships. Spectral decomposition and frequency domain analysis are performed on each feature sub-matrix. The decomposition results are then merged after hierarchical clustering. Based on the clustering results, a correlation spectrum is constructed and spectral features are extracted to form the geological environment correlation spectrum.

4. The integrated air-ground geological environment monitoring and assessment method according to claim 1, characterized in that, The aforementioned method constructs a sky-ground coupled dynamic field based on geological element characteristic chains, and derives a set of geological environment evolution parameters through multi-field coupling analysis, including: The geological element feature chain is divided into regions according to spatial distribution characteristics, and the feature data of each region is analyzed in time series to form a regional dynamic feature matrix. Spatial interpolation and time series decomposition are performed on the dynamic feature matrix of the region to extract periodic and trend change components and construct a change feature field; The changing characteristic field is classified according to geological element type, the coupling coefficient between different elements is calculated, and the sky-ground coupled dynamic field is generated. Spatiotemporal decomposition is performed on each component of the sky-ground coupled dynamic field to extract the inter-field interaction relationship and establish the coupling strength matrix; The coupling strength matrix is ​​decomposed into eigenvalues ​​to extract the dominant coupling modes and construct an evolutionary feature space. The evolutionary feature space is decomposed and reorganized at multiple levels, key evolutionary features are extracted and combined in time and space, and the feature combination is parameterized according to the sky-ground coupling relationship through coupling relationship transmission and feature association analysis to generate the geological environment evolution parameter set.

5. The integrated air-ground geological environment monitoring and assessment method according to claim 1, characterized in that, The aforementioned method establishes an integrated air-ground risk assessment matrix based on a set of geological environmental evolution parameters, and generates a risk early warning indicator system through quantitative hierarchical analysis, including: The geological environment evolution parameter set is divided into multiple evaluation units according to spatial distribution characteristics. The parameters of each evaluation unit are normalized to form a standardized parameter matrix. Correlation analysis is performed on the standardized parameter matrix to calculate the weight coefficients between parameters and construct a parameter weight vector; The parameter weight vector and the standardized parameter matrix are combined to generate the integrated sky-ground risk assessment matrix. The integrated air-ground risk assessment matrix is ​​decomposed into eigenvalues ​​to extract the main risk factors and establish a risk measurement space. The risk measurement space is divided into multiple risk level intervals, and the threshold range of each interval is calculated to form a hierarchical standard system. The graded standard system is subjected to multi-level combination analysis. Risk transmission relationship is determined through risk factor correlation analysis. Early warning level classification criteria are established. Risk zoning is carried out in combination with spatial distribution characteristics. Risk level is determined and early warning threshold is set for each zoning, thereby generating the risk early warning indicator system.

6. The integrated air-ground geological environment monitoring and assessment method according to claim 1, characterized in that, The aforementioned approach, based on a risk warning indicator system and incorporating a historical case database, designs a collaborative air-ground solution to generate a geological environment remediation plan, including: The risk warning indicator system is classified according to risk level, and features are extracted from the indicators under each risk level to form a risk feature sequence. The governance scheme data in the historical case database is organized by attributes, key governance parameters are extracted, and a scheme feature matrix is ​​constructed. The risk feature sequence and the scheme feature matrix are compared with each other to calculate similarity, and historical cases with high matching degree are selected to generate an initial scheme set. The initial set of solutions is optimized by parameters, and solutions are combined for different governance objectives to form a sky-ground collaborative governance strategy. The aforementioned sky-ground collaborative governance strategy is divided into zones according to spatial distribution, and the governance parameters of each zone are adjusted to establish a zone governance matrix. The zonal governance matrix is ​​subjected to multi-level analysis and parameter combination. Through multi-dimensional governance parameter optimization of sky, ground and air, combined with risk warning level and governance effect evaluation, the governance parameters are dynamically adjusted and optimized to construct a governance process sequence and generate the geological environment governance plan.

7. The integrated air-ground geological environment monitoring and assessment method according to claim 6, characterized in that, The process of optimizing the parameters of the initial set of solutions and combining solutions for different governance objectives to form a sky-ground collaborative governance strategy includes: The initial set of schemes is grouped according to the type of governance objective, and the parameters of each group of schemes are numerically standardized to form a standard parameter set; Correlation analysis is performed on the parameters in the standard parameter set to calculate the correlation strength between each parameter and construct a parameter correlation network; The parameter association network is layered according to the governance objectives, and key parameter combinations of each layer are extracted to generate a multi-objective parameter space. Constraints are set for the multi-objective parameter space, and the feasible region is determined through parameter boundary analysis to form the parameter optimization interval; The parameters within the optimization range are combined and matched, and then filtered according to the target priority to establish a sequence of solution combinations; The proposed scheme combination sequence is evaluated from multiple dimensions. Through parameter sensitivity analysis and scheme effect evaluation, the parameter combination is dynamically adjusted. The schemes are optimized and integrated in combination with the sky-ground coordination relationship to form the sky-ground coordinated governance strategy.

8. The integrated air-ground geological environment monitoring and assessment method according to claim 6, characterized in that, The step of dividing the sky-ground collaborative governance strategy into spatial regions, adjusting the governance parameters of each region, and establishing a regional governance matrix includes: The parameter data in the aforementioned sky-ground collaborative governance strategy are processed into a grid according to spatial coordinates to generate a parameter spatial distribution map. Spatial autocorrelation analysis is performed on the spatial distribution map of the parameters to identify the variation characteristics of the parameters and form a spatial variation field; The spatial variation field is divided into regions according to parameter similarity, and the parameters within each region are classified and statistically analyzed to generate a partition parameter sequence. The partition parameter sequence is subjected to boundary optimization processing to eliminate parameter abrupt changes between regions and construct a smooth transition zone; The parameters of the smooth transition zone are jointly analyzed with the parameters within the region to calculate the regional characteristic values ​​and generate the regional characteristic spectrum. Spatial overlay and correlation analysis are performed on the regional feature spectrum. Through parameter partitioning optimization and regional parameter adjustment, combined with spatial continuity analysis, the partition boundaries and internal parameters are dynamically corrected to form the partition governance matrix.

9. A space-ground integrated geological environment monitoring and assessment system, used to implement the space-ground integrated geological environment monitoring and assessment method as described in any one of claims 1-6, characterized in that, The integrated air-ground geological environment monitoring and assessment system includes: The data acquisition module is used to integrate pre-collected sky remote sensing parameters, surface monitoring parameters and underground detection data into the sky-ground integrated monitoring matrix, and generate a raw geological environment database after hierarchical outlier processing. The generation module is used to construct an integrated space-air-ground spatiotemporal standard field based on the original geological environment database, and generate a geological environment correlation spectrum through spatiotemporal registration and data fusion. The decomposition module is used to perform multi-dimensional feature decomposition on the geological environment correlation spectrum and generate a geological element feature chain through sky-ground collaborative analysis. This includes: segmenting the geological environment correlation spectrum according to a time series; dividing the correlation spectrum data within each time period into subspaces to form time-varying feature subspaces; performing singular value decomposition on the data in the time-varying feature subspaces to extract the main feature vectors and generate a feature basis matrix; reorganizing the feature basis matrix according to spatial location information to construct a spatial feature field and extract spatial feature components; performing tensor decomposition on the spatial feature components and time feature vectors to extract spatiotemporal coupling features; classifying and aggregating the spatiotemporal coupling features according to geological element types to establish an element correlation matrix; performing hierarchical decomposition and feature reconstruction on the element correlation matrix; organizing the reconstructed features in a chain according to sky-ground collaborative relationships; performing multi-level feature mapping and correlation analysis; and forming the geological element feature chain through feature transfer and feature fusion. The coupling module is used to construct a dynamic field of sky-ground coupling based on the feature chain of geological elements, and to obtain a set of geological environment evolution parameters through multi-field coupling analysis; The hierarchical module is used to establish an integrated air-ground risk assessment matrix based on the set of geological environment evolution parameters, and to generate a risk warning indicator system through quantitative hierarchical analysis. The design module is used to design a sky-ground collaborative solution based on the risk warning indicator system and combined with the historical case library, and generate a geological environment governance solution.

Images