A method for low-altitude multi-source environmental stress field and geographic space coupling risk prediction and decision optimization

Abstract

The application discloses a low-altitude multi-source environmental stress field and geographic space coupling risk prediction and decision optimization method, and belongs to the technical field of low-altitude equipment environmental adaptability evaluation and space decision. The application obtains multi-dimensional environmental stress parameters in the stereoscopic space of a region to be warned, independently quantifies single damage eigenvalues caused by various environmental stresses such as atmospheric corrosion, electromagnetic interference and acoustic vibration fatigue by using multiple solving branches of built-in physical / chemical dynamics models. On this basis, subintervals with different environmental stress combination characteristics are identified based on a space clustering algorithm, and weights are adaptively assigned, single damage eigenvalues are weighted and fused into a comprehensive environmental stress damage risk index, and then a stereoscopic risk distribution field is reconstructed by three-dimensional interpolation. The application further performs space coordinate registration and layer superposition on the stereoscopic risk distribution field and basic geographic information data, establishes a mapping relationship between the risk distribution field and the real geographic space, and generates decision feedback information such as flight route avoidance strategies, facility site selection suitability classification and differentiated protection level allocation. The application can provide quantitative decision basis for low-altitude aircraft flight route planning, ground facility site selection and infrastructure differentiation protection.

Description

Technical Field

[0001] This invention relates to the field of environmental adaptability assessment and spatial decision-making technology for low-altitude equipment, and particularly to a risk prediction and decision optimization method for low-altitude multi-source environmental stress field coupled with geospatial factors. Background Technology

[0002] With the booming development of the low-altitude economy, the application of drones, electric vertical takeoff and landing (eVTOL) aircraft, and various low-altitude operation equipment is becoming increasingly widespread. At the same time, the environmental stresses experienced by fixed facilities such as towers, high-rise buildings, power grids, and communication base stations at vertical heights exhibit significant spatial heterogeneity. During their service life, these equipment and facilities simultaneously endure the complex coupling effects of multiple environmental stresses, including meteorological erosion (temperature, humidity, sunlight, salt spray), electromagnetic radiation (communication base stations, high-voltage transmission lines), and mechanical vibration and noise, posing severe challenges to their safety and reliability.

[0003] Existing environmental impact assessment methods have the following main shortcomings: First, the assessment dimension is limited to a two-dimensional plane. Existing atmospheric environmental testing standards (such as GB / T 2423) are mainly for ground or near-ground (1-2 meters) environments, which cannot reflect the vertical distribution differences of environmental stress within the low-altitude layer of 50-200 meters, resulting in inaccurate assessment of the environmental adaptability of low-altitude operation equipment.

[0004] Second, the assessment object is singular. Traditional methods often focus on the distribution of a single environmental factor (such as material corrosion caused by temperature and humidity) or a single physical field (temperature field, humidity field), failing to incorporate multiple categories of environmental stresses, such as meteorological, electromagnetic, and acoustic-vibration stresses, into a unified assessment framework. However, high humidity environments may accelerate material corrosion, strong electromagnetic fields may cause performance drift in electronic devices, and high sound pressure level environments may induce structural acoustic-vibration fatigue—single-dimensional assessments cannot truly reflect the comprehensive damage risk of equipment under the coupled effects of multiple environmental stresses.

[0005] Third, the assessment results are disconnected from engineering decisions. Most existing methods stop at the scientific demonstration of environmental effects (such as drawing corrosion distribution maps), failing to further translate the assessment results into usable engineering decision-making information. Even after receiving a color rendering, decision-makers still cannot directly answer specific engineering questions such as "How should aircraft plan their routes to avoid high-risk areas?", "Where should the factory be located?", and "What level of protection should be taken for towers in different locations?"

[0006] Fourth, the assessment methods are static. Traditional environmental effect distribution maps are mostly static displays and cannot be updated in real time with the dynamic changes in environmental parameters. This makes it difficult to meet the needs of real-time early warning and dynamic decision-making in the rapidly changing low-altitude environment and the rapid movement of equipment.

[0007] To address the aforementioned issues, this invention provides a comprehensive solution that integrates low-altitude multi-source environmental stress assessment, three-dimensional risk distribution field reconstruction, geospatial coupling, and intelligent decision mapping into a complete chain solution. Summary of the Invention

[0008] This invention discloses a risk prediction and decision optimization method for low-altitude multi-source environmental stress field coupled with geospatial factors, comprising: Acquire multidimensional environmental stress data at multiple vertical height layers within the three-dimensional space of the area to be warned; Each type of environmental stress is used to independently calculate the individual damage value caused by each type of environmental stress using a damage evolution model corresponding to that type of environmental stress. Based on the spatial distribution characteristics of the multidimensional environmental stress data, the area to be warned is divided into several sub-intervals, and the fusion weight of each individual damage value is adaptively determined for each sub-interval. Based on the aforementioned fusion weights, the individual damage values ​​are weighted and fused to obtain a comprehensive environmental stress damage risk index, which is then reconstructed into a three-dimensional risk distribution field. The three-dimensional risk distribution field is spatially registered and overlaid with basic geographic information data to establish a mapping relationship between the risk distribution field and geographic space. Based on the mapping relationship, at least one type of decision feedback information for low-altitude operations is generated.

[0009] Optionally, the damage evolution model includes at least: Atmospheric corrosion damage model, which calculates the degree of material corrosion or aging based on temperature, humidity, salt spray deposition and solar radiation; Electromagnetic interference damage model, which calculates the electromagnetic induction effect of equipment based on electric field strength, magnetic field strength and spectrum distribution; The acoustic vibration fatigue damage model calculates the cumulative fatigue damage of the structure based on the noise sound pressure level, spectrum, and vibration acceleration.

[0010] Preferably, the method for adaptively determining the fusion weights includes Preset the initial baseline weights for each individual damage value; The existence of a strong coupling state is determined based on the statistical correlation between different environmental stresses within each sub-interval. When a strong coupling state exists, the correlation weights are dynamically adjusted based on the degree to which the measured values ​​of environmental stress deviate from the historical average. The corrected weights are normalized so that the sum of the weights in each sub-interval is 1.

[0011] Optionally, the sub-intervals are divided using a clustering algorithm, and the sample data for the clustering algorithm is obtained in the following way: In the area to be warned, monitoring points are set up in a spatially evenly distributed manner to collect multidimensional environmental stress data and corresponding spatial coordinates of each point; Simultaneously collect performance degradation parameters of target equipment or materials under environmental stress; The sensitive environmental stress factors with the highest correlation to performance degradation were selected through correlation analysis and used as cluster feature vectors.

[0012] Optionally, the method for reconstructing the three-dimensional risk distribution field is as follows: input the spatial coordinates of each monitoring point and the corresponding comprehensive environmental stress damage risk index into the geographic information system, create a volumetric grid through three-dimensional spatial interpolation, and generate a continuous three-dimensional risk distribution field.

[0013] Optionally, the spatial registration and overlay includes: unifying the basic geographic information data to the target coordinate system, performing coordinate matching with the three-dimensional risk distribution field, and achieving visual and semantic fusion between the risk distribution field and the geographic space through at least one of the following methods: voxel grid overlay, isosurface extraction, or texture mapping semi-transparent mixing.

[0014] Optionally, the decision feedback information includes: The route avoidance strategy is generated as follows: The aircraft waypoints are mapped to a three-dimensional risk distribution field to read the risk index; the comprehensive risk value of the route is calculated; when the risk exceeds a safety threshold, a three-dimensional grid search algorithm is used along the gradient descent direction of the risk distribution field to automatically output the optimal alternative route and suggestions for elevation or lateral offset; or... The method for generating facility site suitability classification is as follows: mapping candidate site locations to a three-dimensional risk distribution field to read risk indices, integrating the geographical attribute information of the site locations, and outputting classified site selection recommendations through a weighted comprehensive scoring model; or The method for generating differentiated protection level allocation is as follows: map infrastructure nodes to a three-dimensional risk distribution field, read the risk index, and automatically match the corresponding protection process parameters according to the risk level.

[0015] Optionally, the decision feedback information may also include at least one of the following: take-off and landing point suitability assessment, emergency obstacle avoidance path generation, infrastructure inspection cycle optimization, low-altitude road network automatic planning, and insurance actuarial-assisted pricing.

[0016] Optionally, the correlation analysis employs grey relational analysis to calculate the correlation coefficient and correlation degree, and then sorts and filters sensitive environmental stress factors according to their correlation degree. Optionally, the clustering algorithm is K-means clustering, which uses the cosine distance between sensitive environmental stress factor vectors as the clustering criterion and iteratively updates the cluster centers until convergence.

[0017] Due to the adoption of the above technical solutions, the present invention has the following beneficial effects: 1. This invention expands the evaluation dimension from a two-dimensional plane on the ground to a three-dimensional space, filling the technical gap in the environmental adaptability evaluation of low-altitude equipment.

[0018] 2. By using a multi-branch parallel solution architecture with built-in physical / chemical dynamics models, environmental stresses with different properties such as meteorology, electromagnetics, and acoustic vibration are uniformly incorporated into the evaluation system, solving the problem of cross-dimensional comprehensive evaluation of multi-physics fields.

[0019] 3. By using clustering algorithms to identify the environmental stress combination characteristics of different sub-intervals, and adaptively adjusting the weight ratio of each damage component based on measured data, the assessment results are more consistent with the actual environmental action mechanisms of different regions and different altitude layers.

[0020] 4. Spatial coordinate registration and layer overlay are performed between the three-dimensional environmental stress risk distribution field and basic geographic information such as digital surface model, water system, and road network, so that the abstract rendering map can establish a spatial mapping relationship with the physical entities in the real world.

[0021] 5. Through risk interpretation and decision mapping steps, it automatically outputs specific decision-making information that can directly guide engineering practice, such as route avoidance schemes, site selection classification maps, and differentiated protection schemes, thus solving the pain point of "disconnect between assessment results and decision-making" in traditional methods.

[0022] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0023] The accompanying drawings of this invention are described below.

[0024] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] A risk prediction and decision optimization method coupling low-altitude multi-source environmental stress field with geospatial data, such as Figure 1 As shown, the specific steps include: S1. Data Acquisition and Preprocessing Acquire multidimensional environmental stress raw data at multiple vertical height layers within the three-dimensional space of the area to be warned. The multidimensional environmental stress raw data includes at least meteorological data (temperature, humidity, air pressure, solar radiation, salt spray concentration, etc.), electromagnetic data (electric field strength, magnetic field strength, spectrum distribution, etc.), and mechanical data (noise pressure level, vibration acceleration, spectrum characteristics, etc.).

[0027] Data acquisition is achieved through low-altitude three-dimensional environmental monitoring: For the area to be warned, several monitoring points with different vertical heights and spatial distributions are selected; sensor equipment (air temperature and humidity sensors, total radiation sensors, salt spray deposition acquisition devices, three-dimensional electric field sensors, magnetic field sensors, noise sensors, vibration acceleration sensors, etc.) is deployed at each monitoring point to collect multi-dimensional environmental stress data and record spatial coordinates simultaneously; material samples or device test boards are placed simultaneously to obtain performance degradation characterization parameters (such as metal corrosion weight loss, electronic device signal attenuation rate, structural fatigue crack propagation length, etc.) under the coupling of multiple environmental stresses.

[0028] The collected data undergoes consistency checks, missing value imputation, outlier removal, and dimensionless preprocessing to form a standard sample database.

[0029] S2, Screening of Stress Factors in Sensitive Environments Typical performance degradation parameters of equipment (such as metal corrosion rate, increase in signal error rate of electronic devices, and increase in dielectric loss of insulating materials) were selected as indicators for estimating environmental effects. Grey relational analysis was used to conduct correlation analysis, and key sensitive environmental stress factors with the highest correlation to performance degradation were selected to form the feature vector for subsequent cluster analysis.

[0030] Specifically, for all environmental stress factor data X i After dimensionless transformation of (k) and the environmental effect estimation index Y(k), the correlation coefficient ξ is calculated. i (k) and relevance r: ξ i (k)=[min i m ink ||Y(k)-X i (k)||+ρ·max i m axk ||Y(k)-X i (k)||] / [||Y(k)-X i (k)||+ρ·max i m axk ||Y(k)-X i (k)||] r=(1 / n)·Σ k=1 n ξ i (k) Where ρ is the resolution coefficient (usually taken as 0.5), and ||·|| represents the absolute distance. Key sensitive environmental stress factors affecting equipment performance degradation are selected by sorting them from high to low according to their relevance r.

[0031] S3, Multi-branch parallel solution and single-factor damage quantization Based on the physical property categories of the original multidimensional environmental stress data, they are respectively sent to the corresponding effect calculation branches. Each effect calculation branch has a built-in damage evolution model constructed based on the principles of physical dynamics or chemical dynamics, which independently calculates the individual damage values ​​caused by various types of environmental stresses.

[0032] Each effect calculation branch should include at least: S31, Atmospheric Corrosion Effect Calculation Branch It incorporates the electrochemical corrosion kinetics equations for metals or the aging and degradation kinetics equations for polymer materials. Based on the input temperature, relative humidity, salt spray deposition rate, and solar irradiance, it calculates the material corrosion depth or performance degradation rate, which serves as the first single-item damage characteristic value. Example equation (metal corrosion rate): V_corr=K0·exp(-E a / (R·T))·f(S) Where V_corr is the corrosion rate (mm / a), T is the absolute temperature (K), S is the salt spray deposition rate, and E is the corrosion rate (mm / a). a Here, is the activation energy, R is the gas constant, K0 is the material-related proportionality constant, and f(S) is the salt spray effect function.

[0033] S32, Electromagnetic Interference Effect Calculation Branch It incorporates an electromagnetic field coupling transfer function or induction effect calculation equation. Based on the input electric field strength, magnetic field strength, and spectral energy distribution, it solves for the induced voltage, induced current, or signal bit error rate increment of the equipment or device, which serves as the second single-item damage characteristic value.

[0034] Example equation (induced electromotive force): E_ind=∫ l ( B / t·dl) Where E_ind is the induced electromotive force (V), B is the magnetic flux density (T), and l is the equivalent path length of the circuit (m).

[0035] S33, Solving branch for acoustic vibration fatigue effect The system incorporates a structural acoustic vibration fatigue cumulative damage model or resonance response calculation equation. Based on the input noise sound pressure level, spectral characteristics, and vibration acceleration, it solves for the fatigue cumulative damage coefficient or resonance risk margin of the structural component, which serves as the third individual damage characteristic value.

[0036] Example equation (Miner linear cumulative damage): D_vib=Σ i (n i / N i ) Where D_vib is the fatigue cumulative damage coefficient, n i N represents the actual number of cycles under the i-th stress amplitude. i The number of fatigue fracture cycles under the i-th stress amplitude (determined by the SN curve).

[0037] S4. Spatial Heterogeneity Partitioning and Adaptive Weight Allocation Based on the spatial coordinates of each monitoring point and the corresponding original multidimensional environmental stress data (or sensitive environmental stress factor vectors), a clustering algorithm (preferably K-means algorithm) is used to divide the area to be warned into several sub-intervals with different environmental stress combination characteristics.

[0038] The clustering process is as follows: Given a sample set D={x1, x2, …, x n}, where x i Let be the vector (p-dimensional) of the sensitive environmental stress factors at monitoring point i. Select the number of clusters k, and choose initial cluster centers {y1, y2, …, y}. k}. Calculate the cosine distance from each sample to the cluster center: d ij = (Σ l=1 p x il · y jl ) / ( √(Σ l=1 p x il ²) · √(Σ l=1 p y jl ²) ) Assign samples to the cluster set corresponding to the cluster center with the smallest distance. Iteratively update the cluster centers: y' j = (1 / ||C j ||) · Σ{x∈C j} x Continue until the cluster centers no longer change, thus obtaining the sub-intervals.

[0039] Based on this, adaptive allocation of coupling weights is performed, specifically as follows: Baseline weight assignment: For the service scenario of the target equipment or material, the initial baseline weight values ​​of each effect calculation branch are preset based on the analytic hierarchy process (AHP) or expert weighting method.

[0040] Coupling strength determination: Calculate the statistical correlation (such as Pearson correlation coefficient) or covariance matrix between different types of environmental stress parameters in the current sub-interval to determine whether there is a strong coupling effect of multiple factors (such as the absolute value of the correlation coefficient is greater than the preset threshold).

[0041] Dynamic weight adjustment: When a strong coupling effect is determined, the baseline weight of the related effect solution branch is adjusted up or down according to the degree of deviation of the measured amplitude of each environmental stress parameter in the current sub-interval from its historical mean (such as standardized deviation) by a preset correction function (such as sigmoid function or linear function).

[0042] Weight normalization output: The corrected weight values ​​of each branch are normalized to ensure that the sum of the weight coefficients in each sub-interval is always equal to 1, which is used as the weight coefficient for subsequent fusion.

[0043] S5, Integrated Operations Field Fusion Within each sub-interval, normalized weighting coefficients (e.g., α, β, and γ corresponding to corrosion, electromagnetic, and acoustic vibration branches, respectively) are used to weight and fuse the individual damage feature values. First, the individual damage feature values ​​are normalized (e.g., Min-Max normalization) to obtain dimensionless values. _corr、Ē_ind、 _vib, then calculate the comprehensive environmental stress damage risk index R_total: R_total = α · _corr + β · Ē_ind + γ · _vib For each monitoring point, a comprehensive risk index R_total is calculated.

[0044] S6. Reconstruction of Three-Dimensional Risk Distribution Field Input the spatial coordinates (X, Y, Z) of each monitoring point and the corresponding comprehensive environmental stress damage risk index R_total into a geographic information spatial analysis system (such as ArcGIS, QGIS or a self-developed platform), create a volumetric raster through a three-dimensional spatial interpolation algorithm (preferably Kriging interpolation or inverse distance weighted interpolation), establish a three-dimensional environmental risk display model, and reconstruct a continuous three-dimensional risk distribution field of the area to be warned in the vertical height.

[0045] S7, Geographic Information Coupling Obtain basic geographic information data for the area to be warned, including at least one or more of the following: digital surface model (DSM), water system distribution, road network layout, administrative divisions, and land use types.

[0046] The basic geographic information data is uniformly converted to the target spatial reference coordinate system (such as the CGCS2000 National Geodetic Coordinate System). Using the voxel grid as the reference, the converted geographic information data is spatially matched and layered with the three-dimensional risk distribution field to establish a spatial mapping relationship between the three-dimensional environmental stress risk distribution field and surface features.

[0047] The specific superposition method can adopt at least one of the following methods: Void Grid Overlay Method: Discretize the three-dimensional risk distribution field into a void grid, perform coordinate matching and overlay with the three-dimensional real scene model, assign a risk value to each void and render it by color; Isosurface extraction method: The Marching Cubes algorithm is used to extract isosurfaces with specific risk thresholds from the risk distribution field and automatically generate the risk envelope boundary; Texture mapping semi-transparent blending method: The risk distribution field color is mapped as a semi-transparent texture and superimposed on the surface of the 3D geographic model.

[0048] S8. Risk Interpretation and Decision Mapping Based on the mapping relationship between risk distribution fields and geographic space, and combined with preset risk tolerance thresholds, the risk level of target features is determined point-by-point or area-by-area, and at least one decision feedback information for low-altitude operations is generated. Specifically, this includes: S81. Route Avoidance Strategy: Obtain the pre-planned route of the aircraft (consisting of several waypoints), map each waypoint to a three-dimensional risk distribution field to read the risk index, accumulate the risk integral of the entire route or extract the maximum value, and calculate the comprehensive risk cost of the route. When the safety threshold is exceeded, based on the gradient descent direction of the three-dimensional risk distribution field, use three-dimensional grid discretization and A* algorithm or ant colony algorithm to automatically search for and output the alternative route with the lowest risk cost, and provide suggestions for elevation adjustment or lateral offset.

[0049] S82. Facility Site Selection Suitability Classification: Obtain the geographical boundary of the area to be selected and the coordinates of candidate site points, map them to the risk distribution field to read the risk index, integrate the geographical attribute information of candidate points (distance from water source, distance from industrial pollution source, altitude, vegetation cover, etc.), and output a classification map of "preferred", "condition passed" and "not recommended" based on a weighted comprehensive scoring model.

[0050] S83. Differentiated protection level allocation: Obtain spatial distribution data of target infrastructure (transmission towers, communication base stations, pipelines, elevated bridges, etc.), map each node to the risk distribution field to extract risk index, and automatically match protection process parameters (coating thickness, number of anti-corrosion coating layers, electromagnetic shielding effectiveness index, vibration damping coefficient, etc.) according to preset levels.

[0051] S84. Other derivative decision-making functions: may also include at least one of the following: take-off and landing point suitability assessment, emergency obstacle avoidance path generation, infrastructure inspection cycle optimization, low-altitude road network automatic planning, and insurance actuarial auxiliary pricing.

[0052] S9, Dynamic Real-Time Updates By continuously receiving real-time multidimensional environmental stress raw data from various monitoring points, periodically (e.g., every 10 minutes) iteratively calculating the comprehensive environmental stress damage risk index, dynamically updating the three-dimensional risk distribution field, and refreshing the aforementioned decision feedback information in real time, dynamic prediction and real-time decision support for low-altitude environmental risks are achieved.

[0053] This invention can be widely applied to flight path planning and dynamic obstacle avoidance of low-altitude aircraft (UAVs, eVTOLs), site selection and protection design of ground and elevated infrastructure (transmission towers, communication base stations, bridges), low-altitude logistics network planning, insurance actuarial science and other fields, and has clear industrial application value.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for risk prediction and decision optimization based on the coupling of low-altitude multi-source environmental stress field and geospatial data, characterized in that, include: Acquire multidimensional environmental stress data at multiple vertical height layers within the three-dimensional space of the area to be warned; Each type of environmental stress is used to independently calculate the individual damage value caused by each type of environmental stress using a damage evolution model corresponding to that type of environmental stress. Based on the spatial distribution characteristics of the multidimensional environmental stress data, the area to be warned is divided into several sub-intervals, and the fusion weight of each individual damage value is adaptively determined for each sub-interval. Based on the aforementioned fusion weights, the individual damage values ​​are weighted and fused to obtain a comprehensive environmental stress damage risk index, which is then reconstructed into a three-dimensional risk distribution field. The three-dimensional risk distribution field is spatially registered and overlaid with basic geographic information data to establish a mapping relationship between the risk distribution field and geographic space. Based on the mapping relationship, at least one type of decision feedback information for low-altitude operations is generated.

2. The method as described in claim 1, characterized in that, The damage evolution model includes at least: Atmospheric corrosion damage model, which calculates the degree of material corrosion or aging based on temperature, humidity, salt spray deposition and solar radiation; Electromagnetic interference damage model, which calculates the electromagnetic induction effect of equipment based on electric field strength, magnetic field strength and spectrum distribution; The acoustic vibration fatigue damage model calculates the cumulative fatigue damage of the structure based on the noise sound pressure level, spectrum, and vibration acceleration.

3. The method as described in claim 1, characterized in that, The method for adaptively determining fusion weights includes: Preset the initial baseline weights for each individual damage value; The existence of a strong coupling state is determined based on the statistical correlation between different environmental stresses within each sub-interval. When a strong coupling state exists, the correlation weights are dynamically adjusted based on the degree to which the measured values ​​of environmental stress deviate from the historical average. The corrected weights are normalized so that the sum of the weights in each sub-interval is 1.

4. The method as described in claim 1, characterized in that, The sub-intervals are divided using a clustering algorithm, and the sample data for the clustering algorithm is obtained in the following way: In the area to be warned, monitoring points are set up in a spatially evenly distributed manner to collect multidimensional environmental stress data and corresponding spatial coordinates of each point; Simultaneously collect performance degradation parameters of target equipment or materials under environmental stress; The sensitive environmental stress factors with the highest correlation to performance degradation were selected through correlation analysis and used as cluster feature vectors.

5. The method as described in claim 1, characterized in that, The method for reconstructing the three-dimensional risk distribution field is as follows: input the spatial coordinates of each monitoring point and the corresponding comprehensive environmental stress damage risk index into the geographic information system, create a volume element grid through three-dimensional spatial interpolation, and generate a continuous three-dimensional risk distribution field.

6. The method as described in claim 1, characterized in that, The spatial registration and overlay includes: unifying the basic geographic information data to the target coordinate system, matching the coordinates with the three-dimensional risk distribution field, and achieving visual and semantic fusion between the risk distribution field and the geographic space through at least one of the following methods: voxel grid overlay, isosurface extraction, or texture mapping semi-transparent mixing.

7. The method as described in claim 1, characterized in that, The decision feedback information includes: The route avoidance strategy is generated as follows: The aircraft waypoints are mapped to a three-dimensional risk distribution field to read the risk index; the comprehensive risk value of the route is calculated; when the risk exceeds a safety threshold, a three-dimensional grid search algorithm is used along the gradient descent direction of the risk distribution field to automatically output the optimal alternative route and suggestions for elevation or lateral offset; or... The method for generating facility site suitability classification is as follows: mapping candidate site locations to a three-dimensional risk distribution field to read risk indices, integrating the geographical attribute information of the site locations, and outputting classified site selection recommendations through a weighted comprehensive scoring model; or The method for generating differentiated protection level allocation is as follows: map infrastructure nodes to a three-dimensional risk distribution field, read the risk index, and automatically match the corresponding protection process parameters according to the risk level.

8. The method as described in claim 1, characterized in that, The decision feedback information also includes at least one of the following: take-off and landing point suitability assessment, emergency obstacle avoidance path generation, infrastructure inspection cycle optimization, low-altitude road network automatic planning, and insurance actuarial pricing assistance.

9. The method as described in claim 4, characterized in that, The correlation analysis adopts the grey relational analysis method to calculate the correlation coefficient and correlation degree, and screens sensitive environmental stress factors according to the correlation degree.

10. The method as described in claim 4, characterized in that, The clustering algorithm is K-means clustering, which uses the cosine distance between sensitive environmental stress factor vectors as the clustering criterion and iteratively updates the cluster centers until convergence.

Images