Hydraulic ring surveying and mapping data acquisition method and system based on geographic space information

Abstract

The invention relates to the technical field of hydraulic ring surveying and mapping, in particular to a hydraulic ring surveying and mapping data acquisition method and system based on geographic space information, and the method comprises the following steps: S1, data acquisition and preliminary data set generation: generating an initial geographic space data set; s2, space-time alignment and coupling feature extraction: constructing a preliminary geographic space model; s3, dynamic verification and cross verification: generating a dynamic verification parameter set; s4, three-dimensional geologic modeling and parameter fusion: generating a comprehensive three-dimensional geologic model; s5, environment interference correction and data optimization: analyzing the offset of an environment interference factor to sensor data through an adaptive learning algorithm, and outputting corrected hydraulic ring surveying and mapping data; and S6, safety threshold comparison and early warning map generation: generating a visual early warning map. According to the method, deep support is provided for underground structure analysis, and the comprehensive capability of hydraulic ring surveying and mapping is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogeology, engineering geology and environmental geology surveying and mapping, and particularly to a method and system for collecting hydrogeology, engineering geology and environmental geology surveying and mapping data based on geospatial information. Background Art

[0002] With the acceleration of the urbanization process and the continuous advancement of infrastructure construction, the application of hydrogeology, engineering geology and environmental geology surveying and mapping in engineering construction, environmental protection and disaster prevention is becoming more and more extensive. By obtaining important parameters such as hydrology, rock stratum structure and ground surface deformation, hydrogeology, engineering geology and environmental geology surveying and mapping provides important data for groundwater management, geological exploration and environmental monitoring. However, in complex geographical environments, traditional hydrogeology, engineering geology and environmental geology surveying and mapping methods mostly rely on a single data source, such as ground sensors or remote sensing data, which results in low timeliness and accuracy of data. Especially in the case of extensive monitoring areas and complex environmental conditions, the difficulty of data integration and analysis is relatively large.

[0003] At present, existing hydrogeology, engineering geology and environmental geology surveying and mapping technologies mostly adopt independent data collection and analysis methods, lacking multi-source data fusion and dynamic verification mechanisms. For example, some systems only rely on traditional ground sensors or single remote sensing data to obtain parameters, ignoring the environmental interference and spatio-temporal inconsistency problems of sensor data, resulting in inaccurate measurement results. At the same time, there is a lack of effective model optimization means in the existing technology, and it is impossible to perform real-time correction and verification on a large amount of collected data, resulting in difficult-to-guarantee data quality and inability to effectively support safety assessment and decision-making in engineering construction. In addition, the early warning mechanisms of existing technologies often rely on simple threshold settings and fail to fully consider the interaction and time-varying characteristics of multiple factors.

[0004] The purpose of the present invention is to provide a method and system for collecting hydrogeology, engineering geology and environmental geology surveying and mapping data based on geospatial information, realizing high-precision and high-efficiency collection and analysis of hydrogeology, engineering geology and environmental geology surveying and mapping data, improving the risk assessment and decision-making support capabilities in engineering construction, and thus better meeting the requirements of engineering safety and environmental monitoring. Summary of the Invention

[0005] The present invention provides a method and system for collecting hydrogeology, engineering geology and environmental geology surveying and mapping data based on geospatial information.

[0006] The method for collecting hydrogeology, engineering geology and environmental geology surveying and mapping data based on geospatial information includes the following steps:

[0007] S1, data collection and generation of a preliminary data set: By using an unmanned aerial vehicle equipped with a multi-spectral sensor, a ground distributed sensor network and satellite remote sensing data, synchronously collect hydrological parameters, rock stratum structure parameters and ground surface deformation parameters of a target area to generate an initial geospatial data set;

[0008] S2, Spatiotemporal Alignment and Coupling Feature Extraction: Perform spatiotemporal alignment processing on the initial geospatial dataset, extract hydrogeological coupling feature parameters, and construct a preliminary geospatial model;

[0009] S3, Dynamic Verification and Cross-Validation: Based on the preliminary geospatial model, perform cross-validation on groundwater level monitoring points and surface deformation monitoring points through a dynamic verification algorithm to generate a dynamic verification parameter set;

[0010] S4, 3D Geological Modeling and Parameter Fusion: Inject the dynamic verification parameter set into a 3D geological modeling engine, fuse the strike parameters of rock layer fractures and the pore water pressure gradient parameters to generate a comprehensive 3D geological model;

[0011] S5, Environmental Interference Correction and Data Optimization: Based on the comprehensive 3D geological model, analyze the offset of environmental interference factors on sensor data through an adaptive learning algorithm, and output the corrected hydrogeological survey data;

[0012] S6, Safety Threshold Comparison and Early Warning Atlas Generation: Compare the corrected hydrogeological survey data with the preset engineering safety threshold to generate a visual early warning atlas.

[0013] Optionally, the data acquisition and preliminary dataset generation in S1 include:

[0014] S11, Sensor Cooperative Deployment and Parameter Configuration: Configure the band range (400 - 2500nm) of the drone multispectral sensor, set the deployment density of the ground distributed sensor network to be positively correlated with the degree of rock weathering, and simultaneously receive the acquisition time window of satellite remote sensing data;

[0015] S12, Multimodal Data Synchronous Acquisition: The drone flies along a preset Z-shaped path and triggers data acquisition synchronously, specifically including:

[0016] Collect surface reflection spectral data through the multispectral sensor and label it as hydrogeological parameters;

[0017] Real-time upload the micro-vibration frequency of the rock layer and the groundwater level fluctuation data through the ground sensor network and label it as rock layer structure parameters;

[0018] Obtain the surface InSAR deformation interference map through satellite remote sensing data and label it as surface deformation parameters;

[0019] S13, Spatiotemporal Reference Alignment Processing: Adopt an anti-electromagnetic interference clock synchronization protocol to impose a unified spatiotemporal coordinate system on the hydrogeological parameters, rock layer structure parameters, and surface deformation parameters, and generate an initial geospatial dataset through weighted fusion of the hydrogeological parameters, rock layer structure parameters, and surface deformation parameters.

[0020] Optionally, the spatio-temporal reference alignment processing in S13 includes:

[0021] S131, time alignment: Based on the anti-electromagnetic interference clock synchronization protocol, perform deviation compensation on the timestamps of different data sources;

[0022] S132, weighted fusion: Generate the initial geospatial dataset G by weighted fusion of the hydrological parameters, rock formation structure parameters, and surface deformation parameters after time alignment.

[0023] Optionally, the spatio-temporal alignment and coupling feature extraction in S2 includes:

[0024] S21, extraction of hydro-geological coupling feature parameters: Based on the generated initial geospatial dataset G, calculate the coupling parameters, including the permeability coefficient matrix K(x,y,z) and the fracture-seepage correlation degree β;

[0025] S22, construction of a preliminary geospatial model: Inject the coupling parameters into a three-dimensional grid engine to generate a preliminary geospatial model.

[0026] Optionally, the construction of the preliminary geospatial model in S22 includes:

[0027] S221, grid division: Use an adaptive octree grid, and the condition for the grid side length L is expressed as:

[0028] L = 10m (β ≥ 0.5);

[0029] L = 20m (β < 0.5);

[0030] S222, model parameterization: Assign values at the grid vertices, including the permeability coefficient matrix K(x,y,z) and the fracture-seepage correlation degree β, and calculate the pore water pressure gradient

[0031] S223, model verification: Test the spatial correlation through the variogram γ(h).

[0032] Optionally, the dynamic verification and cross-validation in S3 includes:

[0033] S31, dynamic residual calculation and anomaly detection: Based on the permeability coefficient matrix and pore water pressure gradient in the preliminary geospatial model, predict the theoretical value h pred of the groundwater level and the theoretical value S pred of the surface deformation, and calculate the residuals, including the groundwater level residual ∈ h and the surface deformation residual ∈ S ;

[0034] S32, multi-parameter joint optimization verification: Use the Kalman filter algorithm to dynamically correct the residuals, specifically including:

[0035] State equation construction:

[0036] Wherein, is the state vector, A is the state transition matrix, Bu k is the input control term, w k , v k are the process noise and the observation noise, x k-1 is the value of the state vector at the previous moment, H is the observation matrix, z k is the observation value;

[0037] Parameter optimization: Update the elastic modulus E and the porosity μ;

[0038] S33, Generation of dynamic verification parameter set: Generate a dynamic verification parameter set, including the verification weight W, the time-varying reliability index γ(t), and the coupling correction factor η.

[0039] Optionally, the 3D geological modeling and parameter fusion in S4 include:

[0040] S41, Dynamic generation of fracture network: Based on the formation strike variance δ and the maximum principal stress direction θ, use Monte Carlo simulation to generate a fracture network, and regulate the fracture density ρ fracture , and generate the fracture strike;

[0041] S42, Multi-parameter fusion modeling: Integrate the dynamic verification parameter set with the physical field parameters, correct the pore water pressure gradient field, and generate a stress-seepage coupling equation;

[0042] S43, Adaptive grid modeling and parameter assignment: Perform meshing through Delaunay triangulation. The grid size l is adjusted according to the fracture density. When ρ fracture ≥50, l takes 0.5m. When ρ fracture <50, l takes 2m, and assign the permeability coefficient, the pore water pressure gradient, and the fracture density to the grid nodes;

[0043] S44, Model iteration optimization and verification: Check the difference between the prediction result and the actual measurement value through the convergence criterion. At the same time, update the fracture damage variable by simulating the fracture propagation, and correct the fracture evolution according to the fracture energy.

[0044] Optionally, the environmental interference correction and data optimization in S5 include:

[0045] S51, Quantification of environmental interference factors: Calculate the environmental interference factors, including the electromagnetic interference intensity I EM and the temperature fluctuation effect ΔT;

[0046] S52, Adaptive learning model construction: Construct an adaptive learning model, including an input layer, a hidden layer, and an output layer;

[0047] S53, Model training and optimization: Optimize the adaptive learning model by introducing a loss function. The loss function measures the difference between the predicted value and the true value, and cross-validation is used to avoid overfitting;

[0048] S54, Data correction and output: Correct the predicted values of the groundwater level, surface deformation, and permeability coefficient by compensating for the offset. The corrected predicted values are used to evaluate the correction effect through the residual ratio.

[0049] Optionally, the safety threshold comparison and warning map generation in S6 include:

[0050] S61, Dynamic setting of safety threshold: Set the groundwater level warning threshold h th and the surface deformation warning threshold S th ;

[0051] S62, Multi-parameter joint warning analysis: Based on the groundwater level warning index WI and the surface deformation warning index SI, and divide the comprehensive warning level L. When WI = 0 and SI = 0, the comprehensive warning level L is safe. When 0 < WI ≤ 0.2 or 0 < SI ≤ 0.1, the comprehensive warning level L is low risk. When 0.2 < WI ≤ 0.5 or 0.1 < SI ≤ 0.3, the comprehensive warning level L is medium risk. When WI > 0.5 or SI > 0.3, the comprehensive warning level L is high risk;

[0052] S63, Thermal layer stacking and visualization: Generate the groundwater level thermal map C h and the surface deformation thermal map C S 。

[0053] The hydrogeological environmental survey data acquisition system based on geospatial information is used to implement the above-mentioned hydrogeological environmental survey data acquisition method based on geospatial information, and includes the following modules:

[0054] Data acquisition and preliminary data set generation module: Through the multi-spectral sensor carried by the drone, the ground distributed sensor network, and the satellite remote sensing data, the hydrological parameters, rock layer structure parameters, and surface deformation parameters of the target area are collected in real time and synchronized to generate an initial geospatial data set;

[0055] Spatio-temporal alignment and coupling feature extraction module: Perform spatio-temporal alignment processing on the initial geospatial data set, extract the hydrogeological coupling feature parameters, and construct a preliminary geospatial model;

[0056] Dynamic verification and cross-validation module: Based on the preliminary geospatial model, cross-validate the groundwater level monitoring points and the surface deformation monitoring points through the dynamic verification algorithm to generate a dynamic verification parameter set;

[0057] 3D Geological Modeling and Parameter Fusion Module: Inject the dynamic verification parameter set into the 3D geological modeling engine, fuse the parameters of the rock layer fracture trend and the pore water pressure gradient, and generate a comprehensive 3D geological model;

[0058] Environmental Interference Correction and Data Optimization Module: Based on the comprehensive 3D geological model, analyze the offset of the environmental interference factors on the sensor data through the adaptive learning algorithm, and output the corrected hydrogeological and environmental survey data;

[0059] Safety Threshold Comparison and Early Warning Atlas Generation Module: Compare the corrected hydrogeological and environmental survey data with the preset engineering safety threshold, and generate a visual early warning atlas.

[0060] Advantages of the present invention:

[0061] In the present invention, through the data acquisition and preliminary data set generation module, the hydrological, rock layer structure and surface deformation parameters of the target area can be obtained in real time through the collaborative operation of drones, multispectral sensors, ground distributed sensor networks and satellite remote sensing data, and the initial geospatial data set can be quickly generated. In addition, the spatio-temporal alignment and coupling feature extraction module ensures the spatio-temporal consistency of the data, effectively extracts the hydrological and geological coupling features, and enhances the accuracy and precision of the geospatial model.

[0062] In the present invention, the cross-verification of the groundwater level and surface deformation monitoring data is carried out through the dynamic verification algorithm to generate a dynamic verification parameter set, thereby further optimizing the accuracy of the model. This process effectively reduces the influence caused by measurement errors and data deviations, making the model more in line with the actual situation. Through the 3D geological modeling and parameter fusion module, key information such as the rock layer fracture trend and the pore water pressure gradient can be fused to generate a comprehensive 3D geological model, providing in-depth support for underground structure analysis and enhancing the comprehensive ability of hydrogeological and environmental surveys.

[0063] In the present invention, the influence of environmental interference factors on sensor data is corrected through the adaptive learning algorithm to ensure the high accuracy of the survey data. By real-time correcting the data offset and comparing the corrected data with the preset safety threshold, the safety threshold comparison and early warning atlas generation module can generate a visual early warning atlas to provide support for engineering safety assessment and decision-making. Description of the Drawings

[0064] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only for the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0065] Figure 1 Schematic flow chart of the acquisition method according to an embodiment of the present invention;

[0066] Figure 2 Schematic diagram of the system function modules according to an embodiment of the present invention. Detailed implementation manners

[0067] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0068] It should be pointed out that in the specification, when referring to "an embodiment", "embodiment", "exemplary embodiment", "some embodiments", etc., it indicates that the described embodiment may include specific features, structures or characteristics, but not necessarily every embodiment includes such specific features, structures or characteristics. Additionally, when combining an embodiment to describe a specific feature, structure or characteristic, implementing such feature, structure or characteristic in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0069] Generally, terms can be understood at least in part from their use in context. For example, at least in part depending on the context, the term "one or more" as used herein can be used to describe any feature, structure or characteristic in a singular sense, or can be used to describe a combination of features, structures or characteristics in a plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but rather, at least in part depending on the context, can allow for the existence of other factors that may not be explicitly described.

[0070] As Figure 1 shown, the hydrogeological and environmental survey data acquisition method based on geospatial information includes the following steps:

[0071] S1, Data acquisition and generation of a preliminary data set: Through drones equipped with multi-spectral sensors, a ground distributed sensor network, and satellite remote sensing data, synchronously acquire the hydrological parameters, rock formation structure parameters, and surface deformation parameters of the target area, and generate an initial geospatial data set;

[0072] S2, Spatiotemporal alignment and extraction of coupled features: Perform spatiotemporal alignment processing on the initial geospatial data set, extract the hydrogeological coupling feature parameters, and construct a preliminary geospatial model;

[0073] S3, Dynamic Verification and Cross-Verification: Based on the preliminary geospatial model, cross-verify the groundwater level monitoring points and surface deformation monitoring points through a dynamic verification algorithm to generate a dynamic verification parameter set;

[0074] S4, 3D Geological Modeling and Parameter Fusion: Inject the dynamic verification parameter set into the 3D geological modeling engine, fuse the parameters of the rock layer fracture trend and the pore water pressure gradient to generate a comprehensive 3D geological model;

[0075] S5, Environmental Interference Correction and Data Optimization: Based on the comprehensive 3D geological model, analyze the offset of environmental interference factors on sensor data through an adaptive learning algorithm, and output the corrected hydrogeological survey data;

[0076] S6, Safety Threshold Comparison and Early Warning Atlas Generation: Compare the corrected hydrogeological survey data with the preset engineering safety threshold to generate a visual early warning atlas;

[0077] Through the above content, it is possible to efficiently and comprehensively collect multi-dimensional information such as hydrology, rock layer structure, and surface deformation in the target area, ensuring the comprehensiveness and accuracy of the data. The spatio-temporal alignment processing and hydro-geological coupling feature extraction can effectively eliminate the spatio-temporal deviation between different data sources, improve data consistency. The introduction of dynamic verification and 3D geological modeling technology further optimizes data verification and model accuracy, ensuring a high degree of consistency between groundwater level monitoring and surface deformation. By correcting the offset of environmental interference factors through an adaptive learning algorithm, the finally output hydrogeological survey data is more reliable, thus improving engineering safety, and providing an intuitive risk assessment tool for decision-makers through the visual early warning atlas, which helps in real-time monitoring and early warning. This not only improves the intelligence level of data collection and processing, but also significantly improves the accuracy and efficiency of hydrogeological survey work.

[0078] The data collection and preliminary dataset generation in S1 include:

[0079] S11, Sensor Collaborative Deployment and Parameter Configuration: Configure the band range (400 - 2500nm) of the drone multi-spectral sensor, set the deployment density of the ground distributed sensor network to be positively correlated with the degree of rock weathering, and at the same time receive the acquisition time window of satellite remote sensing data;

[0080] S12, Multi-modal Data Synchronous Acquisition: The drone flies along a preset Z-shaped path and synchronously triggers data acquisition, specifically including:

[0081] Collect surface reflection spectral data through the multi-spectral sensor and label it as hydrological parameters;

[0082] Real-time upload the micro-vibration frequency of the rock layer and the groundwater level fluctuation data through the ground sensor network and label it as rock layer structure parameters;

[0083] Obtain the surface InSAR deformation interference map through satellite remote sensing data, which is marked as the surface deformation parameter;

[0084] S13. Space-time reference alignment processing: Adopt a clock synchronization protocol resistant to electromagnetic interference, apply a unified space-time coordinate system to the hydrological parameters, rock formation structure parameters, and surface deformation parameters, ensure that the data timestamp error does not exceed 1 ms, and generate an initial geospatial dataset by the weighted fusion method for the hydrological parameters, rock formation structure parameters, and surface deformation parameters;

[0085] Through the above content, high-precision hydrogeological, engineering geological, and environmental geological survey data acquisition can be achieved. The collaborative deployment and parameter configuration of sensors ensure the efficient cooperation between different data sources, enabling the synchronous acquisition of multi-dimensional data such as hydrology, rock formation structure, and surface deformation, so as to comprehensively reflect the environmental state of the target area. Through the Z-shaped path flight of the unmanned aerial vehicle and real-time data upload, key parameters such as surface reflection spectrum, micro-vibration frequency of rock formations, and underground water level fluctuations can be accurately obtained, ensuring the timeliness and comprehensiveness of the data. The space-time reference alignment processing and the clock synchronization protocol resistant to electromagnetic interference effectively solve the space-time error problem between different data sources, making the finally generated dataset have higher consistency and accuracy.

[0086] The space-time reference alignment processing in S13 includes:

[0087] S131. Time alignment: Based on the clock synchronization protocol resistant to electromagnetic interference, perform deviation compensation on the timestamps of different data sources, expressed as:

[0088]

[0089] where Δt is the time deviation, t ground,i is the timestamp of the i-th ground sensor data point, t uav,i is the timestamp of the i-th unmanned aerial vehicle data point, and n is the number of samples participating in the time alignment calculation;

[0090] S132. Weighted fusion: Generate an initial geospatial dataset G by the weighted fusion method for the hydrological parameters, rock formation structure parameters, and surface deformation parameters after time alignment, expressed as:

[0091] G = λ W ·W + λ R ·R + λ D ·D;

[0092] where W is the hydrological parameter, R is the rock formation structure parameter, D is the surface deformation parameter, and λ W 、λ R 、λ D are the corresponding weight coefficients respectively;

[0093] Through the above, the time difference deviation between different data sources can be effectively eliminated, ensuring the synchronization of various sensor data under the same time benchmark, thereby improving the timeliness and consistency of the data. Secondly, by adopting the weighted fusion method to fuse different types of parameters such as hydrology, rock stratum structure, and ground deformation, the data combination can be optimized according to the weight of each parameter, enhancing the accuracy and reliability of the fusion result.

[0094] The spatio-temporal alignment and coupling feature extraction in S2 include:

[0095] S21, extraction of hydro-geological coupling characteristic parameters: Based on the generated initial geospatial dataset G, calculate the coupling parameters, including the permeability coefficient matrix K(x, y, z) and the fracture-flow correlation β, which are expressed as:

[0096]

[0097] where T is the transmissivity coefficient, calculated from the groundwater level fluctuation data, b is the aquifer thickness, is the groundwater level gradient, R is the rock mass integrity index (0 - 1), from the geological layer of the initial dataset;

[0098]

[0099] where F i is the fracture density at the i-th monitoring point, ΔS i is the ground deformation at the corresponding point, from the ground displacement vector field, and m is the number of monitoring points;

[0100] S22, construction of the preliminary geospatial model: Inject the coupling parameters into the three-dimensional grid engine to generate a preliminary geospatial model;

[0101] Through the above, the coupling characteristics extracted based on the initial geospatial dataset, such as the permeability coefficient matrix and the fracture-flow correlation, can effectively capture the interaction between hydrology and geology, thereby providing accurate parameter support for subsequent geological assessment and hydrological simulation. Injecting these coupling parameters into the three-dimensional grid engine to generate a preliminary geospatial model enables the intuitive display of the correlation between spatial distribution and physical properties.

[0102] The construction of the preliminary geospatial model in S22 includes:

[0103] S221, grid division: Adopt an adaptive octree grid, and the condition for the grid side length L is expressed as:

[0104] L = 10m (β ≥ 0.5);

[0105] L = 20m (β < 0.5);

[0106] S222, Model parameterization: Assign values at the grid vertices, including the permeability coefficient matrix K(x, y, z) and the fracture-flow correlation β, and calculate the pore water pressure gradient Expressed as:

[0107]

[0108] Where ρ w is the water density, g is the acceleration due to gravity, is the ground surface deformation acceleration, is the groundwater level gradient;

[0109] S223, Model verification: Examine the spatial correlation through the variogram γ(h), expressed as:

[0110]

[0111] Where λ i is the weight coefficient of the i-th data point, N(h) is the number of samples participating in the calculation at the spatial interval h, Z(x i ) is the known data value (weighted fusion data value) at the i-th location, Z(x i +h) is the known data value (weighted fusion data value) at the location with a displacement of h, γ(h) is the variogram. When the coefficient of determination R 2 ≥0.85, output the preliminary geospatial model;

[0112] Through the above content, the high precision and high reliability of the model are ensured. The adaptive octree grid division enables the model to flexibly adjust the resolution according to the complexity of the geospatial area, accurately reflecting the characteristics of different regions. Assigning key parameters such as the permeability coefficient matrix, fracture-flow correlation, and pore water pressure gradient at the grid vertices can accurately simulate factors such as groundwater flow and rock layer permeability. The application of the variogram verifies the spatial correlation of the model, ensuring the consistency and accuracy of the model in spatial distribution.

[0113] The dynamic verification and cross-validation in S3 include:

[0114] S31, Dynamic residual calculation and anomaly detection: Based on the permeability coefficient matrix and pore water pressure gradient in the preliminary geospatial model, predict the theoretical groundwater level value h pred and the theoretical ground surface deformation value S pred , and calculate the residuals, including the groundwater level residual ∈ h and the ground surface deformation residual ∈ S , expressed as:

[0115]

[0116] Among them, h0 is the initial groundwater level, K is the permeability coefficient matrix, μ is the porosity of the rock formation, t0 is the initial time, and t is the current time;

[0117]

[0118] Among them, E is the elastic modulus of the geological body, σ i is the stress change at the i-th monitoring point, β i is the fracture-seepage correlation degree at the corresponding point, and Δt is the time change;

[0119] ∈ h =|h meas -h pred |;

[0120] ∈ S =||ΔS meas -S pred ||2;

[0121] Among them, h meas is the measured groundwater level value, and ΔS meas is the measured surface deformation value;

[0122] S32, multi-parameter joint optimization verification: The Kalman filter algorithm is used to dynamically correct the residuals, specifically including:

[0123] State equation construction:

[0124] Among them, is the state vector, A is the state transition matrix, Bu k is the input control term, w k 、v k are the process noise and the observation noise, x k-1 is the value of the state vector at the previous moment, H is the observation matrix, and z k is the observed value;

[0125] Parameter optimization: Update the elastic modulus E and the porosity μ, expressed as:

[0126]

[0127] μ new =μ·exp(-0.1∈ h );

[0128] Among them, E new is the updated elastic modulus, and μ new is the updated porosity;

[0129] S33, Generation of dynamic verification parameter set: Generate a dynamic verification parameter set, including verification weight W, time-varying reliability index γ(t), and coupling correction factor η, expressed as:

[0130]

[0131] where T' is the length of the sliding time window (24 hours), ∈ h (τ) is the value of the groundwater level residual at time τ, ∈ S (τ) is the value of the surface deformation residual at time τ, h meas (τ) is the measured value of the groundwater level at time τ, S meas (τ) is the measured value of the surface deformation at time τ, W i is the verification weight of the i-th monitoring point;

[0132] Through the above content, residual calculation and anomaly detection ensure that the difference between the predicted value and the actual measured value of the model can be monitored in real time and adjusted in a timely manner, avoiding large deviations in the model. Through multi-parameter joint optimization verification, the prediction of groundwater level and surface deformation is dynamically corrected using the Kalman filter algorithm, optimizing parameters such as porosity and elastic modulus, improving the adaptability of the model. By introducing the time-varying reliability index and coupling correction factor, the reliability of the model can be evaluated and adjusted, making the prediction of the model more accurate and credible under different time and environmental conditions.

[0133] The 3D geological modeling and parameter fusion in S4 include:

[0134] S41, Dynamic generation of fracture network: Based on the variance of the rock layer strike δ and the direction of the maximum principal stress θ, use Monte Carlo simulation to generate a fracture network and regulate the fracture density ρ fracture , and generate the fracture strike, specifically including:

[0135] Fracture density regulation:

[0136] where ρ fracture is the fracture density, and ρ0 is the reference fracture density;

[0137] Fracture strike generation: Obey the probability density function of the Fisher distribution, expressed as:

[0138]

[0139] where, is the probability density function of the fracture strike, is the fracture strike angle, and κ is the concentration parameter, indicating the degree of concentration of the fracture strike;

[0140] S42, Multi-parameter fusion modeling: Fuse the dynamic verification parameter set with the physical field parameters, correct the pore water pressure gradient field, and generate the stress-seepage coupling equation, specifically including:

[0141] Correction of the pore water pressure gradient field:

[0142] Among them, is the corrected pore water pressure gradient, is the original pore water pressure gradient, is the time variation of the verification weight matrix;

[0143] Stress-seepage coupling equation:

[0144] Among them, is the head gradient, S s is the storage coefficient, α is the Biot coefficient, ∈ v is the volumetric strain, σ ij,j is the component of the stress tensor, ρ is the density, g i is the acceleration due to gravity;

[0145] S43, Adaptive grid modeling and parameter assignment: Perform meshing through Delaunay triangulation. The mesh size l is adjusted according to the fracture density. When ρ fracture ≥ 50, l takes 0.5 m. When ρ fracture < 50, l takes 2 m, and assign the permeability coefficient, pore water pressure gradient, and fracture density to the grid nodes, expressed as:

[0146] Permeability coefficient K(x, y, z) → Grid seepage property;

[0147]

[0148] Fracture density ρ fracture → Grid damage factor;

[0149] S44, Model iteration optimization and verification: Check the difference between the predicted result and the actual measured value through the convergence criterion. At the same time, update the fracture damage variable by simulating the fracture propagation, and correct the fracture evolution according to the fracture energy, specifically including:

[0150] Convergence criterion: And

[0151] Among them, h model,i is the predicted groundwater level of the model, h meas,i is the actually measured groundwater level, S meas is the measured value of the surface deformation, and N is the total number of monitoring points used to calculate the error;

[0152] Fracture propagation simulation: The phase field method is used to solve the fracture evolution equation, which is expressed as:

[0153]

[0154] where d is the damage variable, g(d)=(1 - d) 2 , G c is the fracture energy, obtained by β correction, M is a constant, and l0 is the initial length scale;

[0155] Through the above content, complex phenomena such as groundwater flow, surface deformation, and fracture propagation can be accurately reflected. By using adaptive meshing and parameter space mapping, the calculation accuracy in different regions is improved, enabling the model to better adapt to changes in geological and hydrological characteristics. At the same time, the dynamic correction of fracture density and pore water pressure gradient makes the model more consistent with actual observation data, improving the reliability of predictions.

[0156] The environmental interference correction and data optimization in S5 include:

[0157] S51, Quantification of environmental interference factors: Calculate environmental interference factors, including electromagnetic interference intensity I EM and the influence of temperature fluctuation ΔT, which is expressed as:

[0158]

[0159] where B i is the measured magnetic field intensity of the i-th sensor, B0 is the background magnetic field intensity, d i is the distance between the sensor and the interference source, λ is the attenuation coefficient, and P is the number of sensors;

[0160]

[0161] where T j is the reading of the j-th temperature sensor, T ref is the reference temperature (20 °C), is the temperature coefficient of rock mass density, and Q is the number of temperature sensors;

[0162] S52, Construction of an adaptive learning model: Construct an adaptive learning model, including an input layer, a hidden layer, and an output layer, which is expressed as:

[0163] Input layer: 5 nodes, which is expressed as:

[0164]

[0165] where X is the input feature vector;

[0166] Hidden layer: 2 layers, with 16 nodes in each layer, and the activation function is ReLU;

[0167] Output layer: 3 nodes, with a linear activation function, expressed as:

[0168] Y = [Δh offset , ΔS offset , ΔK offset T ;

[0169] where Y is the output target vector, and Δh offset , ΔS offset , ΔK offset are the offsets of the groundwater level, surface deformation, and permeability coefficient respectively;

[0170] S53, Model training and optimization: Optimize the adaptive learning model by introducing a loss function. The loss function measures the difference between the predicted value and the true value, and cross-validation is used to avoid overfitting, specifically including:

[0171] Loss function design:

[0172] where L is the loss function, Z is the number of samples, Y i is the i-th target value, is the i-th predicted value, and W is the verification weight matrix;

[0173] Training data augmentation: Apply Gaussian noise (σ = 0.05) to X and use K-fold cross-validation (K = 5) to prevent overfitting;

[0174] S54, Data correction and output: Correct the predicted values of the groundwater level, surface deformation, and permeability coefficient by compensating for the offsets to ensure data accuracy. The corrected predicted values are evaluated for the correction effect through the residual ratio to ensure that the final results meet the actual requirements, specifically including:

[0175] Offset compensation:

[0176] where h corrected is the corrected groundwater level, S corrected is the corrected surface deformation, K corrected is the corrected permeability coefficient, h raw is the original groundwater level, S raw is the original surface deformation, and K raw is the original permeability coefficient;

[0177] Evaluation of correction effect: Calculate the residual ratio r before and after correction, expressed as:

[0178]

[0179] ​Among them, r is the residual ratio before and after correction. When r ≤ 0.3, it is determined that the correction is effective. Y corrected is the corrected target value, Y true is the true target value, Y raw is the original target value;

[0180] Through the above content, the prediction accuracy of the model is effectively improved. The impacts of electromagnetic interference and temperature fluctuations on sensor data are accurately quantified, and the impacts of these environmental interference factors are eliminated through the correction formula. An adaptive learning model is used to compensate for the offset of the data to ensure more accurate prediction results of the groundwater level, surface deformation, and permeability coefficient. Through the training methods of loss function and data augmentation, the performance of the model is further optimized, and higher stability and generalization ability are ensured. The correction effect evaluation index ensures the effectiveness of data correction, making the final correction result more reliable and accurate in practical applications.

[0181] The safety threshold comparison and early warning map generation in S6 include:

[0182] S61, dynamic setting of safety threshold: Set the groundwater level early warning threshold h th and the surface deformation early warning threshold S th , expressed as:

[0183]

[0184] Among them, h th is the groundwater level early warning threshold, h base is the reference groundwater level threshold, is the groundwater level change rate, Δt is the monitoring time interval, and γ(t) is the time-varying reliability index;

[0185]

[0186] Among them, S th is the surface deformation early warning threshold, S base is the reference deformation threshold, is the corrected pore water pressure gradient, is the maximum pore water pressure gradient;

[0187] S62, multi-parameter joint early warning analysis: Based on the groundwater level early warning index WI and the surface deformation early warning index SI, and divide the comprehensive early warning level L. When WI = 0 and SI = 0, the comprehensive early warning level L is safe. When 0 < WI ≤ 0.2 or 0 < SI ≤ 0.1, the comprehensive early warning level L is low risk. When 0.2 < WI ≤ 0.5 or 0.1 < SI ≤ 0.3, the comprehensive early warning level L is medium risk. When WI > 0.5 or SI > 0.3, the comprehensive early warning level L is high risk;

[0188] The groundwater level warning index WI is expressed as:

[0189]

[0190] Where h corrected is the corrected groundwater level;

[0191] The surface deformation warning index SI is expressed as:

[0192]

[0193] Where S corrected is the corrected surface deformation;

[0194] S63, Thermal layer stacking and visualization: Generate the groundwater level thermal map C h and the surface deformation thermal map C S , which is expressed as:

[0195]

[0196] Through the above content, the real-time monitoring and assessment of potential risks are realized. The calculation of the warning index enables the deviation degree of the groundwater level and surface deformation to be intuitively quantified, providing a reliable basis for risk assessment. At the same time, through the thermal map and visualization functions, the areas with different risk levels are displayed in the form of color coding, enabling decision-makers to quickly identify high-risk areas and take corresponding measures.

[0197] Such as Figure 2 shown, the hydrogeological and environmental survey data acquisition system based on geospatial information is used to implement the above-mentioned hydrogeological and environmental survey data acquisition method based on geospatial information, and includes the following modules:

[0198] Data acquisition and preliminary dataset generation module: Through the multi-spectral sensor carried by the unmanned aerial vehicle, the ground distributed sensor network and satellite remote sensing data, the hydrological parameters, rock layer structure parameters and surface deformation parameters of the target area are collected in real-time and synchronously, and the initial geospatial dataset is generated;

[0199] Spatio-temporal alignment and coupling feature extraction module: Perform spatio-temporal alignment processing on the initial geospatial dataset, extract the hydrogeological coupling feature parameters, and construct a preliminary geospatial model;

[0200] Dynamic verification and cross-validation module: Based on the preliminary geospatial model, cross-validate the groundwater level monitoring points and surface deformation monitoring points through the dynamic verification algorithm, and generate a dynamic verification parameter set;

[0201] 3D Geological Modeling and Parameter Fusion Module: Inject the dynamic verification parameter set into the 3D geological modeling engine, fuse the parameters of the rock layer fracture trend and the pore water pressure gradient, and generate a comprehensive 3D geological model;

[0202] Environmental Interference Correction and Data Optimization Module: Based on the comprehensive 3D geological model, analyze the offset of the environmental interference factor on the sensor data through an adaptive learning algorithm, and output the corrected hydrogeological, engineering geological and environmental geological survey data;

[0203] Safety Threshold Comparison and Early Warning Atlas Generation Module: Compare the corrected hydrogeological, engineering geological and environmental geological survey data with the preset engineering safety threshold, and generate a visual early warning atlas.

[0204] The present invention covers any alternatives, modifications, equivalent methods and solutions made on the essence and scope of the present invention. For the public to have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, and those skilled in the art can fully understand the present invention without these detailed descriptions. In addition, well-known methods, processes, procedures, components and circuits are not described in detail to avoid unnecessary confusion to the essence of the present invention.

[0205] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information, characterized in that, It includes the following steps: S1, Data acquisition and initial dataset generation: By using drones equipped with multispectral sensors, ground distributed sensor networks, and satellite remote sensing data, synchronously acquire hydrological parameters, rock formation structure parameters, and surface deformation parameters of the target area to generate an initial geospatial dataset; S2, Spatiotemporal alignment and coupled feature extraction: Perform spatiotemporal alignment processing on the initial geospatial dataset, extract hydro-geological coupled feature parameters, and construct a preliminary geospatial model; S3, Dynamic verification and cross-validation: Based on the preliminary geospatial model, perform cross-validation on groundwater level monitoring points and surface deformation monitoring points through a dynamic verification algorithm to generate a dynamic verification parameter set; S4, 3D geological modeling and parameter fusion: Inject the dynamic verification parameter set into a 3D geological modeling engine, fuse the rock fracture strike parameters and pore water pressure gradient parameters to generate a comprehensive 3D geological model; S5, Environmental interference correction and data optimization: Based on the comprehensive 3D geological model, analyze the offset of environmental interference factors on sensor data through an adaptive learning algorithm, and output the corrected hydrogeological engineering survey data; S6, Safety threshold comparison and warning map generation: Compare the corrected hydrogeological engineering survey data with the preset engineering safety threshold to generate a visual warning map.

2. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 1, wherein The data acquisition and initial dataset generation in S1 include: S11, Sensor collaborative deployment and parameter configuration: Configure the band range of the drone multispectral sensor, set the deployment density of the ground distributed sensor network to be positively correlated with the degree of rock weathering, and at the same time receive the acquisition time window of satellite remote sensing data; S12, Multimodal data synchronous acquisition: The drone flies along a preset Z-shaped path and triggers data acquisition synchronously, specifically including: Collect surface reflection spectral data through the multispectral sensor and label it as hydrological parameters; Real-time upload the rock micro-vibration frequency and groundwater level fluctuation data through the ground sensor network and label it as rock formation structure parameters; Obtain the surface InSAR deformation interferogram through satellite remote sensing data and label it as surface deformation parameters; S13, Spatiotemporal reference alignment processing: Adopt an anti-electromagnetic interference clock synchronization protocol to apply a unified spatiotemporal coordinate system to the hydrological parameters, rock formation structure parameters, and surface deformation parameters, and generate an initial geospatial dataset by using the weighted fusion method for the hydrological parameters, rock formation structure parameters, and surface deformation parameters.

3. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 1, wherein The spatiotemporal reference alignment processing in S13 includes: S131, Time alignment: Based on the anti-electromagnetic interference clock synchronization protocol, perform deviation compensation on the timestamps of different data sources; S132, Weighted fusion: Generate an initial geospatial dataset G by using the weighted fusion method for the hydrological parameters, rock formation structure parameters, and surface deformation parameters after time alignment.

4. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 1, wherein, The spatiotemporal alignment and coupled feature extraction in S2 include: S21, Hydro-geological coupled feature parameter extraction: Based on the generated initial geospatial dataset G, calculate the coupling parameters, including the permeability coefficient matrix K(x, y, z) and the fracture-seepage correlation degree β; S22, Preliminary geospatial model construction: Inject the coupling parameters into a 3D grid engine to generate a preliminary geospatial model.

5. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 4, characterized in that, The preliminary geospatial model construction in S22 includes: S221, grid division: Adaptive octree grids are used. The condition that the grid side length L satisfies is expressed as: L = 10m (β ≥ 0.5); L = 20m (β < 0.5); S222, Model parameterization: Assign values at the grid vertices, including the permeability coefficient matrix K(x, y, z) and the fracture-flow correlation degree β, and calculate the pore water pressure gradient S223, model verification: The spatial correlation is tested through the variogram γ(h).

6. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 1, wherein The dynamic verification and cross-validation in S3 include: S31, Dynamic Residual Calculation and Anomaly Detection: Based on the permeability coefficient matrix and pore water pressure gradient in the preliminary geospatial model, predict the theoretical value h of the groundwater level pred and the theoretical value S of the surface deformation pred , and calculate the residuals, including the groundwater level residual ∈ h and the surface deformation residual ∈ S ; S32, multi-parameter joint optimization verification: The Kalman filter algorithm is used to dynamically correct the residuals, specifically including: State equation construction: Among them, is the state vector, A is the state transition matrix, Bu k is the input control term, w k , v k are the process noise and the observation noise, x k-1 is the value of the state vector at the previous moment, H is the observation matrix, z k is the observed value; Parameter optimization: Update the elastic modulus E and porosity μ; S33, generation of dynamic verification parameter set: Generate a dynamic verification parameter set, including verification weight W, time-varying reliability index γ(t), and coupling correction factor η.

7. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 6, wherein The 3D geological modeling and parameter fusion in S4 include: S41, Dynamic generation of fracture network: Based on the variance δ of the rock stratum strike and the maximum principal stress direction θ, use Monte Carlo simulation to generate a fracture network and regulate the fracture density ρ fracture , and generate the fracture strike; S42, multi-parameter fusion modeling: The dynamic verification parameter set is fused with physical field parameters to correct the pore water pressure gradient field and generate a stress-seepage coupling equation; S43, Adaptive grid modeling and parameter assignment: Meshing is carried out through Delaunay triangulation. The grid size l is adjusted according to the fracture density. When ρ fracture ≥ 50, l is taken as 0.5 m. When ρ fracture < 50, l is taken as 2 m, and the permeability coefficient, pore water pressure gradient and fracture density are assigned to the grid nodes; S44, model iteration optimization and verification: Check the difference between the prediction result and the actual measurement value through the convergence criterion. At the same time, by simulating the propagation of fractures, update the fracture damage variable and correct the fracture evolution according to the fracture energy.

8. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 7, characterized in that, The environmental interference correction and data optimization in S5 include: S51, Environmental interference factor quantification: Calculate the environmental interference factors, including the electromagnetic interference intensity I EM and the temperature fluctuation effect ΔT; S52, construction of adaptive learning model: Construct an adaptive learning model, including an input layer, a hidden layer, and an output layer; S53, model training and optimization: Optimize the adaptive learning model by introducing a loss function. The loss function measures the difference between the predicted value and the true value, and cross-validation is used to avoid overfitting; S54, data correction and output: Correct the predicted values of the groundwater level, surface deformation, and permeability coefficient by compensating for the offset. The corrected predicted values are used to evaluate the correction effect through the residual ratio.

9. The method for collecting hydrogeological, engineering geological and environmental geological survey data based on geospatial information according to claim 8, characterized in that, The safety threshold comparison and warning map generation in S6 include: S61, Dynamic setting of safety threshold: Set the warning threshold h of the groundwater level th and the warning threshold S of the surface deformation th ; S62, multi-parameter joint warning analysis: Based on the groundwater level warning index WI and the surface deformation warning index SI, and divide the comprehensive warning level L. When WI = 0 and SL = 0, the comprehensive warning level L is safe. When 0 < WI ≤ 0.2 or 0 < SI ≤ 0.1, the comprehensive warning level L is low risk. When 0.2 < WI ≤ 0.5 or 0.1 < Si ≤ 0.3, the comprehensive warning level L is medium risk. When WI > 0.5 or Si > 0.3, the comprehensive warning level L is high risk; S63, Thermal Layer Stacking and Visualization: Generating a Thermal Map of the Groundwater Level C h and a Thermal Map of the Land Surface Deformation C S .

10. A hydrogeological, engineering geological and environmental geological survey data acquisition system based on geospatial information, which is used to implement the hydrogeological, engineering geological and environmental geological survey data acquisition method based on geospatial information according to any one of claims 1-9, characterized in that, It includes the following modules: Data acquisition and preliminary dataset generation module: Through the multi-spectral sensor carried by the drone, the ground distributed sensor network, and satellite remote sensing data, synchronously collect the hydrological parameters, rock formation structure parameters, and surface deformation parameters of the target area in real time, and generate an initial geospatial dataset; Spatio-temporal alignment and coupling feature extraction module: Perform spatio-temporal alignment processing on the initial geospatial dataset, extract hydro-geological coupling feature parameters, and construct a preliminary geospatial model; Dynamic verification and cross-validation module: Based on the preliminary geospatial model, cross-validate the groundwater level monitoring points and surface deformation monitoring points through the dynamic verification algorithm to generate a dynamic verification parameter set; 3D Geological Modeling and Parameter Fusion Module: Inject the dynamic verification parameter set into the 3D geological modeling engine, fuse the parameters of the rock layer fracture trend and the pore water pressure gradient parameter to generate a comprehensive 3D geological model; Environmental Interference Correction and Data Optimization Module: Based on the comprehensive 3D geological model, analyze the offset of the environmental interference factor on the sensor data through an adaptive learning algorithm, and output the corrected hydrogeological, engineering geological and environmental geological survey data; Safety Threshold Comparison and Early Warning Atlas Generation Module: Compare the corrected hydrogeological, engineering geological and environmental geological survey data with the preset engineering safety threshold to generate a visual early warning atlas.

Images