Land space data mining and management system based on big data analysis

Abstract

The invention relates to the technical field of spatial data analysis and management, in particular to a territorial spatial data mining and management system based on big data analysis, comprising a spatial topology fingerprint generation module used for receiving original territorial spatial data; the cross-modal association network construction module is used for constructing a spatial association network; the space-time abnormal conduction link mining module is used for identifying a conduction link with a space-time abnormal amplification effect; the dynamic planning conflict pre-judgment module is used for generating a visual pre-judgment report containing a conflict type, a conduction path and an influence range; the scheme self-correction execution module is used for outputting space adjustment scheme sets with different optimization dimensions; according to the method, intelligent management and optimization adjustment of territorial spatial data are realized through multi-modal spatial data mining, space-time anomaly conduction analysis and dynamic planning conflict pre-judgment, and the accuracy of spatial conflict identification and the scientificity of planning scheme adjustment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of spatial data analysis, and particularly to a national land spatial data mining and management system based on big data analysis. Background Art

[0002] With the continuous deepening of the development and utilization of national land space, the increase in surface building density, the complexity of underground facility layout, and the protection requirements of ecologically sensitive areas have made it necessary for national land space planning to take into account multiple factors such as buildings, transportation, environment, and infrastructure at the same time; in practical applications, the multi-modal fusion and mining analysis of national land space data have become important supporting technologies for space planning, risk assessment, and optimization decision-making; in order to improve the scientificity of national land space planning, modern planning methods generally adopt big data analysis, geographic information system (GIS), and spatial topology modeling technologies to achieve precise analysis of spatial data; however, the current national land space data management and analysis methods are mainly based on a single-mode data processing method, which is difficult to take into account the information association of different spatial levels at the same time, resulting in insufficient comprehensive impact assessment of the surface, underground, and ecological environment during the planning decision-making process, increasing spatial conflicts and potential risks.

[0003] The existing technologies have the following problems in national land space data mining and management: First, there is a lack of a modeling method based on topological relationships and multi-level spatial element associations, which cannot effectively identify the implicit association paths between different spatial layers, resulting in insufficient ability to identify abnormal risk conduction paths; second, in terms of spatial conflict prediction, the existing methods only rely on static data for spatial adaptability analysis, and fail to establish a complete spatio-temporal dynamic conduction mechanism, making it difficult to predict the spatio-temporal cascade effects that may be caused by planning changes; third, during the process of optimizing the planning scheme, the current adjustment strategies are mostly empirical methods, lacking quantitative optimization means based on the adjustment of spatial topological structures, resulting in insufficient pertinence and fineness of the scheme adjustment. Therefore, there is an urgent need for a national land spatial data mining and management system based on big data analysis to solve the above problems. Summary of the Invention

[0004] Based on the above purposes, the present invention provides a national land spatial data mining and management system based on big data analysis.

[0005] The national land spatial data mining and management system based on big data analysis includes a spatial topology fingerprint generation module, a cross-modal association network construction module, a spatio-temporal abnormal conduction link mining module, a dynamic planning conflict prediction module, and a scheme self-correction execution module; among them:

[0006] The spatial topology fingerprint generation module: is used to receive the original national land spatial data, and generate a topologically unique fingerprint code through a non-Euclidean geometric feature extraction algorithm, and output a multi-dimensional spatial data set with topological fingerprint identification;

[0007] Cross-modal association network construction module: Based on multidimensional spatial data sets, cross-layer association analysis is performed on surface building density, underground facility distribution, and ecologically sensitive area boundaries to construct a spatial association network containing explicit connections and implicit conduction paths;

[0008] Spatiotemporal abnormal conduction link mining module: used to implement bidirectional conduction simulation in spatial correlation networks, identify conduction links with spatiotemporal abnormal amplification effects, and output a set of conduction paths with risk level labels;

[0009] Dynamic planning conflict prediction module: used to receive the draft of the national land planning from external input, conduct three-dimensional space conflict rehearsal in combination with the transmission path set, and generate a visual prediction report including the conflict type, transmission path and impact range;

[0010] Scheme self-correction execution module: used to call the preset topology structure adjustment rule library according to the conflict type in the prediction report, and output at least three sets of space adjustment schemes with different optimization dimensions.

[0011] Optionally, the spatial topology fingerprint generation module includes a data receiving unit, a feature processing unit, a fingerprint code generation unit and a data output unit; wherein:

[0012] Data receiving unit: used to receive original land space data, wherein the original land space data includes vector boundaries, remote sensing features and underground pipe network topology;

[0013] Feature processing unit: used to normalize the data of the vector boundary, remote sensing ground feature and underground pipe network topology respectively, and quantitatively analyze the curvature, connectivity and topological adjacent relationship based on the non-Euclidean geometric feature extraction algorithm to obtain a feature matrix that can characterize the difference of spatial structure;

[0014] Fingerprint code generation unit: used to extract topological fingerprint information with spatial uniqueness according to the feature matrix, and digitally encode the topological fingerprint information according to a preset encoding rule to generate a topological fingerprint code;

[0015] The data output unit is used to associate and map the topological fingerprint code with the normalized original land space data, and output a multidimensional spatial data set with topological fingerprint identification.

[0016] Optionally, the feature processing unit includes:

[0017] Data normalization processing: After receiving vector boundaries, remote sensing features and underground pipe network topology data, the minimum-maximum normalization algorithm is used to perform numerical uniform processing on the data;

[0018] Curvature quantification analysis: For the normalized vector boundary and remote sensing feature data of ground objects, the discrete curvature calculation method is used to obtain the curvature value K of the spatial structure;

[0019] Connectivity quantification analysis: For the normalized topological data of underground pipe networks, based on the topological graph theory analysis algorithm, the connectivity index value C between the nodes of the underground pipe networks is calculated;

[0020] Topological adjacency relationship quantification analysis: For the vector boundary, remote sensing feature of ground objects and topological data of underground pipe networks after data normalization processing, the topological adjacency relationship matrix A is calculated based on the topological adjacency matrix analysis method;

[0021] Feature matrix generation: According to the respectively calculated curvature value K, connectivity index value C and topological adjacency relationship matrix A, a complete spatial structure difference feature matrix M is constructed, and its expression is: M = [K, C, A].

[0022] Optionally, the cross-modal association network construction module includes a cross-layer data mapping unit, a spatial association analysis unit, a hidden conduction path mining unit and a network structure generation unit; among them:

[0023] Cross-layer data mapping unit: Used to receive the multi-dimensional spatial data set with topological fingerprint identification, and based on the topological fingerprint coding, perform unified coordinate projection on the surface building density, underground facility distribution and ecological sensitive area boundary to obtain the cross-modal spatial data mapping result;

[0024] Spatial association analysis unit: Used to perform similarity analysis on the mapping result generated by the cross-layer data mapping unit, calculate the spatial correlation coefficient between the surface building density and the underground facility distribution, and perform topological matching on the adjacent relationship between the ecological sensitive area boundary and the underground facilities;

[0025] Hidden conduction path mining unit: Based on the spatial correlation coefficient and adjacent relationship result output by the spatial association analysis unit, and using the shortest path conduction algorithm to determine the hidden conduction path that is not directly connected but has conduction between the surface building unit, underground facility unit and ecological sensitive area boundary;

[0026] Network structure generation unit: Used to construct a spatial association network according to the explicit connection relationship and the hidden conduction path, where the nodes are spatial units with topological fingerprint identification, and the edges are composed of the explicit connection relationship and the hidden conduction path, and perform differential marking on the explicit connection relationship and the hidden conduction path to generate a spatial association network including the explicit connection and the hidden conduction path.

[0027] Optionally, the hidden conduction path mining unit includes:

[0028] Node and edge construction: Consider surface building units, underground facility units, and the boundaries of ecological sensitive areas as nodes respectively. Establish connectable edges between the nodes according to topological adjacency or potential influence relationships to form a graph structure for conduction analysis;

[0029] Edge weight determination: For any two connectable nodes, determine the edge weight for measuring the conduction cost by comprehensively considering the feature similarity between the nodes and geographical or topological distance factors;

[0030] Shortest path search: In the constructed graph structure, for any two nodes that are not directly connected, traverse all feasible paths based on the shortest path conduction algorithm and find the path with the minimum total conduction cost as the shortest conduction distance between them;

[0031] Latent conduction determination: Compare the obtained shortest conduction distance with a pre-set conduction threshold. When the shortest conduction distance does not exceed the conduction threshold, it is determined that there is a latent conduction path between the two nodes.

[0032] Optionally, the spatio-temporal abnormal conduction link mining module includes a bidirectional conduction simulation unit, an abnormal amplification effect evaluation unit, a risk level determination unit, and a conduction path output unit; where:

[0033] Bidirectional conduction simulation unit: Used to receive a cross-modal association network and, through a pre-set spatio-temporal conduction model, take any node in the network as the initial abnormal source and conduct forward and reverse simulation conduction to adjacent nodes along the explicit connection relationship and the latent conduction path respectively, calculate the cumulative value of node abnormal amplification in each path, and generate an abnormal conduction diffusion result;

[0034] Abnormal amplification effect evaluation unit: Used to analyze the abnormal conduction diffusion result of the bidirectional conduction simulation unit, calculate the conduction amplification coefficient based on the ratio of the abnormal intensity of the path start node to the intensity of the path end node after receiving the abnormality, and quantify the abnormal amplification effect of each conduction path;

[0035] Risk level determination unit: Used to determine the risk level of each conduction path according to the conduction amplification coefficient of the abnormal amplification effect evaluation unit. Specifically, compare the conduction amplification coefficient with a pre-set risk level threshold, and clarify the risk level mark corresponding to the path according to the interval where the ratio is located;

[0036] Conduction path output unit: Used to output a set of conduction paths with risk level marks, where each conduction path records a complete node sequence, conduction direction, and corresponding risk level mark.

[0037] Optionally, the bidirectional conduction simulation unit includes:

[0038] Initial abnormal source setting: Select any node in the cross-modal association network as the initial abnormal source, and assign an abnormal intensity value S0 to this node;

[0039] Forward conduction simulation: According to the explicit connection relationship and implicit conduction path, starting from the initial abnormal source node, conduct abnormal conduction to all directly connected adjacent nodes, and calculate the abnormal intensity S received by each target node i ;

[0040] Reverse conduction simulation: Use the end node of the target path as the reverse initial abnormal source, trace back and calculate the abnormal conduction path along the explicit connection relationship and implicit conduction path, and determine the abnormal intensity S received by the node during the reverse conduction i ';

[0041] Calculation of abnormal amplification value: Combine the abnormal intensities calculated in the forward and reverse conduction processes to calculate the cumulative abnormal amplification value of each path. The formula is: where F represents the cumulative abnormal amplification value of the path, N2 is the total number of nodes included in the path, S j and S j ' are the forward and reverse abnormal intensities of the jth node on the path respectively.

[0042] Optionally, the abnormal amplification effect evaluation unit includes:

[0043] Extraction of path abnormal intensity: Select a conduction path in the spatial association network, and obtain the abnormal intensity S0 of the starting node of the path and the abnormal intensity S received by the end node of the path N ;

[0044] Calculation of path abnormal loss: Calculate the cumulative abnormal loss value on the path, which is used to measure the degree of abnormal attenuation during the abnormal conduction process. The calculation formula is: L = S0 - S N , where L represents the total loss of the abnormal on the corresponding path;

[0045] Calculation of conduction amplification coefficient: Calculate the conduction amplification coefficient of the path, which is used to quantify the amplification or attenuation effect of the abnormal intensity of the starting node of the path relative to the abnormal intensity received by the end node of the path. The calculation formula is: where G represents the conduction amplification coefficient of the path. If G > 1, it means that the abnormal is amplified on the path, and the amplification degree increases with the increase of G; if 0 < G < 1, it means that the abnormal attenuates during the conduction on the path, and the attenuation degree increases with the decrease of G; if G = 1, it means that there is no loss during the conduction of the abnormal on the path.

[0046] Optionally, the dynamic programming conflict prediction module includes a data input and parsing unit, a conflict conduction simulation unit, a spatial conflict identification unit, a conflict path tracing unit, and a conflict prediction report generation unit; among them:

[0047] Data input and parsing unit: It is used to receive the draft of the national land plan input externally, extract the spatial location data and construction type data in the plan draft, and project the extracted data onto three-dimensional spatial coordinates to form a spatial planning dataset;

[0048] Conflict conduction simulation unit: It is used to combine the set of conduction paths marked with risk levels output by the spatio-temporal anomaly conduction link mining module, superimpose the planning dataset onto the spatial association network, and use the newly added or adjusted nodes in the planning dataset as the starting points, and simulate the influence diffusion process on the surrounding nodes according to the explicit connection relationship and implicit conduction paths to determine the potential abnormal diffusion influence range of the newly added nodes in the plan;

[0049] Spatial conflict identification unit: It is used to compare the three-dimensional spatial position relationship between each spatial node in the spatial planning dataset and the current spatial nodes, determine spatial overlap, spatial distance conflict and conduction influence conflict, and clarify the conflict types of each conflict node, including position overlap conflict, distance conflict and abnormal conduction conflict;

[0050] Conflict path tracing unit: It is used to trace the node sequence of the abnormal conduction link for the planning node determined to have a conduction influence conflict, identify and mark the starting node of the abnormal influence, the intermediate conduction node and the final influence node, and determine the complete conflict conduction path;

[0051] Conflict pre-judgment report generation unit: It is used to integrate the identified conflict types, the corresponding complete conduction paths and the abnormal diffusion influence range, and generate a conflict pre-judgment report that intuitively displays the spatial conflict area, conduction path and risk level in a three-dimensional visualization form.

[0052] Optionally, the self-correcting execution module of the solution includes a rule matching unit, a solution generation unit and a solution output unit; among them:

[0053] Rule matching unit: It is used to call the preset topological structure adjustment rule library according to the conflict type and retrieve the adjustment rule set corresponding to the conflict type;

[0054] Solution generation unit: Based on the adjustment rule set output by the rule matching unit, perform topological structure optimization in the three dimensions of land use layout, facility distribution and ecological boundary respectively to form at least three spatial adjustment solutions with different optimization dimensions;

[0055] Solution output unit: It is used to output the spatial adjustment solution output by the solution generation unit in a visual form, and attach the comparison result of the topological structure before and after optimization for decision-makers to choose.

[0056] Advantages of the present invention:

[0057] The present invention can simulate the abnormal bidirectional conduction effect through the spatio-temporal abnormal conduction link mining module, quantify the abnormal amplification degree of each path, and combine with the dynamic programming conflict prediction module to accurately identify the possible spatial conflicts and abnormal diffusion risks in the draft national land plan. At the same time, the present invention adopts a self-correcting execution mechanism based on the topological structure adjustment rule library, which can automatically generate optimization solutions according to different conflict types, thereby improving the intelligence level of the plan adjustment.

[0058] The present invention constructs a topological fingerprint code, enabling multi-dimensional spatial data to be matched and analyzed under the same topological benchmark, enhancing the structured management ability of the data. The conflict prediction method based on spatio-temporal dynamic analysis enables accurate identification of high-risk areas before the implementation of the plan adjustment, avoiding the chain risk diffusion caused by spatial conflicts, and providing effective support for the refined management of national land space. Brief Description of the Drawings

[0059] 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 drawings in the following description are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0060] Figure 1 It is a schematic diagram of the national land space data mining and management system according to the embodiment of the present invention;

[0061] Figure 2 It is a schematic diagram of the spatio-temporal abnormal conduction link mining module according to the embodiment of the present invention. Detailed Embodiments

[0062] The present invention will be described in detail below with reference to the drawings and specific embodiments. At the same time, 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; and the drawing part is only for more specifically describing the embodiments, and is not intended to specifically limit the present invention.

[0063] It should be pointed out that in the specification, it is mentioned that "an embodiment", "embodiment", "exemplary embodiment", "some embodiments", etc. indicate that the described embodiments may include specific features, structures or characteristics, but not necessarily every embodiment includes such specific features, structures or characteristics. In addition, when combining embodiments to describe specific features, structures or characteristics, the implementation of such features, structures or characteristics in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0064] Generally, terms can be understood at least in part from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or property in a singular sense, or can be used to describe a combination of features, structures, or properties 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 can alternatively, depending at least in part on the context, allow for the existence of other factors that are not necessarily explicitly described.

[0065] As Figure 1 - Figure 2 shown, the national territorial space data mining and management system based on big data analysis includes a spatial topology fingerprint generation module, a cross-modal association network construction module, a spatio-temporal anomaly conduction link mining module, a dynamic programming conflict prediction module, and a scheme self-correction execution module; wherein:

[0066] Spatial topology fingerprint generation module: used to receive the original national territorial space data, and generate a topological fingerprint code with spatial uniqueness through a non-Euclidean geometric feature extraction algorithm, and output a multi-dimensional spatial data set with topological fingerprint identification;

[0067] Cross-modal association network construction module: based on the multi-dimensional spatial data set, conduct cross-layer association analysis on the surface building density, underground facility distribution, and ecological sensitive area boundary, and construct a spatial association network including explicit connection and implicit conduction paths;

[0068] Spatio-temporal anomaly conduction link mining module: used to perform two-way conduction simulation in the spatial association network, identify the conduction links with spatio-temporal anomaly amplification effect, and output a set of conduction paths marked with risk levels;

[0069] Dynamic programming conflict prediction module: used to receive the externally input national territorial planning draft, combine the set of conduction paths to conduct a three-dimensional spatial conflict preview, and generate a visual prediction report including conflict types, conduction paths, and influence ranges;

[0070] Scheme self-correction execution module: used to call the preset topological structure adjustment rule library according to the conflict types in the prediction report, and output at least three sets of spatial adjustment schemes with different optimization dimensions.

[0071] The spatial topology fingerprint generation module includes a data receiving unit, a feature processing unit, a fingerprint coding generation unit, and a data output unit; wherein:

[0072] Data receiving unit: used to receive the original national territorial space data, and the original national territorial space data includes vector boundaries, remote sensing ground object features, and underground pipe network topologies;

[0073] Feature processing unit: used to perform data normalization on vector boundaries, remote sensing feature objects, and underground pipe network topologies respectively, and conduct quantitative analysis on curvature, connectivity, and topological adjacency relationships based on non-Euclidean geometric feature extraction algorithms to obtain a feature matrix that can characterize spatial structure differences;

[0074] Fingerprint encoding generation unit: used to extract topographical fingerprint information with spatial uniqueness from the feature matrix, and digitally encode this topographical fingerprint information according to a preset encoding rule to generate a topographical fingerprint code;

[0075] The above-mentioned preset encoding rule includes:

[0076] Encoding rule 1: According to the topological adjacency relationship matrix in the feature matrix, sort the elements in the topological adjacency relationship matrix in descending order of the number of node connections, and assign a unique topological adjacency relationship serial number to each node;

[0077] Encoding rule 2: Set curvature levels according to the numerical range of curvature values, map the curvature values to corresponding curvature level encodings, and sort them by spatial unit;

[0078] Encoding rule 3: Set a threshold based on the connectivity index value, mark the underground pipe network topology nodes exceeding the threshold as high connectivity levels, and sequentially perform connectivity level encoding;

[0079] Encoding rule 4: Adopt a spatial position hashing algorithm to combine the above topological adjacency relationship serial numbers, curvature level encodings, and connectivity index encodings to generate a topographical fingerprint information code with spatial uniqueness.

[0080] Data output unit, used to perform associated mapping on the topographical fingerprint code and the original national land spatial data after normalization processing, and output a multi-dimensional spatial data set with topographical fingerprint identification; through the above units, the spatial topographical fingerprint generation module can ensure the spatial uniqueness of national land spatial data during structure analysis, realize the accurate identification and marking of vector boundaries, feature objects, and underground pipe network topologies in complex geographical environments, thereby improving the accuracy and consistency of subsequent spatial data analysis, and providing high-quality input data for the construction of cross-modal association networks.

[0081] The feature processing unit includes:

[0082] Data normalization processing: After receiving vector boundary, remote sensing feature object, and underground pipe network topology data, use the min-max normalization algorithm to perform numerical unification processing on the data. The calculation formula is: Among them, X represents the original data to be normalized, X min and X max respectively represent the minimum and maximum values of this type of original data, X normRepresents the data after normalization;

[0083] Curvature quantization analysis: For the normalized vector boundary and remote sensing feature data of ground objects, the discrete curvature calculation method is used to obtain the curvature value K of the spatial structure. The formula is: where r is the position vector of the spatial point on the vector boundary or the curve of the remote sensing feature of the ground object, T represents the unit tangent vector of the spatial curve, s represents the arc length parameter, is the differential of the unit tangent vector with respect to the arc length parameter, is the differential of the position vector with respect to the arc length parameter;

[0084] Connectivity quantization analysis: For the normalized topological data of the underground pipe network, based on the topological graph theory analysis algorithm, the connectivity index value C between the nodes of the underground pipe network is calculated. The calculation formula is: where e represents the number of actually existing edges in the topological graph of the underground pipe network, and n1 represents the total number of nodes in the topological graph of the underground pipe network;

[0085] Topological adjacency relationship quantization analysis: For the vector boundary, remote sensing feature of the ground object and topological data of the underground pipe network after data normalization, the topological adjacency relationship matrix A is calculated based on the topological adjacency matrix analysis method. The calculation formula is: A = [a ij n×n ;

[0086] where,

[0087] The matrix element a ij represents the topological adjacency relationship between spatial elements. It takes the value of 1 when the spatial topology is adjacent and 0 when the spatial topology is not adjacent; n represents the total number of spatial elements;

[0088] Feature matrix generation: According to the respectively calculated curvature value K, connectivity index value C and topological adjacency relationship matrix A, a complete spatial structure difference feature matrix M is constructed. Its expression is: M = [K, C, A]. This feature matrix M combines the above calculated curvature value, connectivity index and topological adjacency relationship by columns, so as to form a feature matrix for quantitatively describing the spatial structure difference; Through the specific steps of the above feature processing unit, accurate quantitative analysis of the spatial structure data can be realized, so as to provide clear and quantifiable basic data support for the subsequent generation of unique topological fingerprint codes in space, and ensure the accuracy and reliability of the generation of topological fingerprint codes.

[0089] The cross-modal association network construction module includes a cross-layer data mapping unit, a spatial association analysis unit, a hidden conduction path mining unit and a network structure generation unit; among them:

[0090] ​Cross - layer data mapping unit: It is used to receive a multi - dimensional space data set with topological fingerprint identification, and perform unified coordinate projection on the surface building density, underground facility distribution, and ecological sensitive area boundary based on the topological fingerprint coding to obtain a cross - modal space data mapping result;

[0091] Spatial correlation analysis unit: It is used to perform similarity analysis on the mapping result generated by the cross - layer data mapping unit, calculate the spatial correlation coefficient between the surface building density and the underground facility distribution, and perform topological matching on the adjacent relationship between the ecological sensitive area boundary and the underground facilities; The spatial correlation coefficient is calculated using the following formula: Where, R ij represents the spatial correlation coefficient between the i - th surface building unit and the j - th underground facility unit, X i,k represents the density eigenvalue of the i - th surface building unit in the k - th dimension, Y j,k represents the distribution eigenvalue of the j - th underground facility unit in the k - th dimension, and respectively represent the mean values of their respective eigenvalues, and m represents the number of dimensions of the eigenvector;

[0092] Latent conduction path mining unit: Based on the spatial correlation coefficient and adjacent relationship results output by the spatial correlation analysis unit, and using the shortest path conduction algorithm to determine the latent conduction paths that are not directly connected but have conduction between the surface building units, underground facility units, and ecological sensitive area boundary, and record the node order of this path in the network;

[0093] Network structure generation unit: It is used to construct a spatial correlation network according to the explicit connection relationship and the latent conduction path, where the nodes are spatial units with topological fingerprint identification, and the edges are composed of the explicit connection relationship and the latent conduction path, and perform differential marking on the explicit connection relationship and the latent conduction path to generate a spatial correlation network containing explicit connections and latent conduction paths; Through the above units, the cross - modal association network construction module can enable different - modality spatial information to achieve structured cross - layer correlation analysis, thereby establishing an accurate spatial conduction relationship network and improving the accuracy and stability of subsequent spatial anomaly propagation link mining and conflict prediction.

[0094] The network structure generation unit executes the following steps:

[0095] Node construction: Receive spatial units with topological fingerprint identification, establish a node set with surface building units, underground facility units, and ecological sensitive area boundaries as independent nodes respectively, and use the corresponding topological fingerprint coding as the unique identifier for each node;

[0096] Explicit connection establishment: Based on the explicit connection relationships determined by the spatial association analysis unit, connect node pairs with direct spatial adjacency or a spatial correlation coefficient exceeding a preset threshold, and mark each explicit connection relationship as an explicit edge;

[0097] Implicit conduction path establishment: Based on the implicit conduction paths determined by the implicit conduction path mining unit, connect all the nodes on the path in the order of path conduction, and mark each implicit conduction path as an implicit edge;

[0098] Network fusion: Integrate the explicit edges and implicit edges to form a unified spatial association network structure, where all nodes, explicit edges, and implicit edges carry clear topological fingerprint identifiers and type labels; Through the above steps, it can be ensured that the explicit connection relationships and implicit conduction paths are unified and integrated into a complete spatial association network, forming a clear and structured network expression, thereby improving the subsequent recognition and analysis efficiency for spatio-temporal anomalies and spatial conflicts.

[0099] The implicit conduction path mining unit includes:

[0100] Node and edge construction: Consider the surface building units, underground facility units, and ecological sensitive area boundaries as nodes respectively, and establish connectable edges between the nodes according to topological adjacency or potential influence relationships to form a graph structure for conduction analysis;

[0101] Edge weight determination: For any two connectable nodes, determine the edge weight used to measure the conduction cost by comprehensively considering the feature similarity between the nodes and geographical or topological distance factors. The greater the edge weight, the higher the conduction difficulty between the nodes;

[0102] Shortest path search: In the constructed graph structure, for any two non-directly connected nodes, traverse all feasible paths based on the shortest path conduction algorithm and find the path with the minimum total conduction cost as the shortest conduction distance between them;

[0103] Implicit conduction determination: Compare the obtained shortest conduction distance with a pre-set conduction threshold. When the shortest conduction distance does not exceed the conduction threshold, it is determined that there is an implicit conduction path between the two nodes.

[0104] Specifically, first consider the surface building units, underground facility units, and ecological sensitive area boundaries as nodes respectively to form a node set N; and consider the node pairs that satisfy topological adjacency or have potential influence relationships as connectable edges to form an edge set E, thereby obtaining a graph structure G=(N, E) for performing conduction analysis;

[0105] Then, for any edge (p,q)∈E in graph G, calculate the conduction weight w of this edge p,q, the conduction weight is used to measure the conduction cost between node p and node q, and the calculation formula is: w p,q = α × (1 - R p,q + β × D p,q , where p and q respectively represent any two adjacent nodes in graph G, α and β represent the weighted coefficients determined in the present invention, R p,q represents the feature similarity between node p and node q, and D p,q represents the geographical or topological distance between node p and node q;

[0106] Next, for any two non-directly connected nodes r and s, the shortest path conduction algorithm is used to search all path sets P(r, s) from node r to node s on graph G, and the following formula is used to determine the shortest conduction distance dist(r, s) between node r and node s: where p represents a feasible path from node r to node s, (u, v) represents an adjacent node pair in path p, and w u,v represents the conduction weight calculated in the above steps;

[0107] Finally, for the obtained shortest conduction distance dist(r, s), it is compared with the set conduction threshold T 阈值 When dist(r, s) ≤ T 阈值 , it is determined that there is an absolute conduction path between node r and node s, and the node sequence included in this shortest path is recorded;

[0108] Through the step-by-step analysis and threshold determination of the above shortest path conduction algorithm, the hidden conduction paths that are not directly connected but have potential risk impacts among the surface building units, underground facility units, and ecological sensitive area boundaries can be accurately identified, avoiding the limitation of the traditional method that only focuses on the explicit adjacent relationship and misses the indirect conduction channels, and providing more in-depth and extensive support data for subsequent spatial anomaly detection and planning conflict prediction.

[0109] The spatio-temporal anomaly conduction link mining module includes a bidirectional conduction simulation unit, an anomaly amplification effect evaluation unit, a risk level determination unit, and a conduction path output unit; among them:

[0110] Bidirectional conduction simulation unit: used to receive the cross-modal association network, and through a preset spatio-temporal conduction model, taking any node in the network as the initial anomaly source, perform forward and reverse simulation conduction along the explicit connection relationship and the hidden conduction path to adjacent nodes respectively, calculate the cumulative value of node anomaly amplification in each path, and generate the anomaly conduction diffusion result;

[0111] Abnormal amplification effect evaluation unit: used to analyze the abnormal conduction diffusion results of the bidirectional conduction simulation unit, calculate the conduction amplification coefficient based on the ratio of the abnormal intensity of the path start node to the intensity after the path end node receives the abnormality, and quantify the abnormal amplification effect of each conduction path;

[0112] Risk level determination unit: used to determine the risk level of each conduction path according to the conduction amplification coefficient of the abnormal amplification effect evaluation unit. Specifically, compare the conduction amplification coefficient with the preset risk level threshold, and clarify the risk level mark corresponding to the path according to the interval where the ratio is located;

[0113] Conduction path output unit: used to output the set of conduction paths with risk level marks, where each conduction path records the complete node sequence, conduction direction, and corresponding risk level mark for subsequent use by the dynamic programming conflict prediction module; Through the above units, the spatio-temporal abnormal conduction link mining module can clearly identify and mark the abnormal conduction paths and risk levels in the spatial association network, intuitively reveal the diffusion process and risk degree of various abnormalities between different spatial units, and provide a reliable and clear analysis basis for dynamic programming conflict prediction.

[0114] The bidirectional conduction simulation unit includes:

[0115] Initial abnormal source setting: Select any node in the cross-modal association network as the initial abnormal source, and assign the abnormal intensity value S0 to this node. This abnormal intensity value is set according to the geographical environment characteristics, spatial structure characteristics, or historical abnormal data of the node;

[0116] Forward conduction simulation: Starting from the initial abnormal source node, according to the explicit connection relationship and implicit conduction path, conduct abnormal conduction to all directly connected adjacent nodes, and calculate the abnormal intensity S received by each target node i , and the calculation formula is as follows: Among them, S i represents the abnormal intensity received by the i-th adjacent node; γ 0,i is the abnormal conduction ratio from the initial abnormal source node to the i-th node; λ is the abnormal attenuation coefficient; d 0,i is the spatial distance from the initial abnormal source node to the i-th node; Continue along the explicit connection relationship and implicit conduction path, for all nodes that receive abnormalities, use them as new abnormal sources, and calculate the abnormal intensity of the next-level nodes according to the forward conduction formula until the abnormal conduction calculations of all nodes in the network are completed, forming a complete set of abnormal propagation paths;

[0117] Reverse conduction simulation: Use the target path end node as the reverse initial abnormal source, and trace back and calculate the abnormal conduction path along the explicit connection relationship and implicit conduction path to determine the abnormal intensity S received by the node during the reverse conduction i', and the calculation formula is as follows: Among them, S i ' represents the abnormal intensity received by the i-th node during the reverse conduction process, and S i+1 ' is the abnormal intensity of its adjacent node, δ i+1,i is the reverse conduction ratio, μ is the reverse abnormal attenuation coefficient, and d i+1,i is the spatial distance from node i + 1 to node i;

[0118] Abnormal amplification value calculation: Combine the abnormal intensities calculated in the forward and reverse conduction processes to calculate the cumulative abnormal amplification value of each path. The formula is: Among them, F represents the cumulative abnormal amplification value of the path, N2 is the total number of nodes included in the path, S j and S j ' are the forward and reverse abnormal intensities of the j-th node on the path respectively; Through the above process of two-way conduction simulation steps, the propagation path and influence range of the anomaly in the spatial correlation network can be accurately calculated. Combining the calculation of the cumulative abnormal amplification value helps to identify the conduction links that have a greater impact on the surrounding area.

[0119] The abnormal amplification effect evaluation unit includes:

[0120] Path abnormal intensity extraction: Select a conduction path in the spatial correlation network, and obtain the abnormal intensity S0 of the starting node of the path and the abnormal intensity S N received by the end node of the path;

[0121] Path abnormal loss calculation: Calculate the cumulative abnormal loss value on the path, which is used to measure the degree of abnormal attenuation caused by factors such as topological attenuation, medium damping, and spatial distance during the abnormal conduction process. The calculation formula is: L = S0 - S N , where L represents the total loss of the anomaly on the corresponding path. If L = 0, it means that the abnormal conduction has not decayed. If L > 0, it means that the anomaly has weakened during the conduction process;

[0122] Conduction amplification coefficient calculation: Calculate the conduction amplification coefficient of the path, which is used to quantify the amplification or attenuation effect of the abnormal intensity of the starting node of the path relative to the abnormal intensity received by the end node of the path. The calculation formula is: Among them, G represents the conduction amplification coefficient of the path. If G > 1, it means that the anomaly is amplified on the path, and the amplification degree increases with the increase of G; If 0 < G < 1, it means that the anomaly decays during the conduction on the path, and the attenuation degree increases with the decrease of G; If G = 1, it means that there is no loss during the conduction of the anomaly on the path; Through the above steps, the amplification or attenuation of the anomaly in the spatial correlation network can be accurately quantified. Combining the calculation results of the conduction amplification coefficient provides a quantitative basis for subsequent risk assessment and conflict prediction.

[0123] The risk level determination rules in the risk level determination unit are as follows:

[0124] When G > T H then it is determined that the path belongs to a high-risk path, and the risk level is marked as R3;

[0125] When T M ≤ G ≤ T H then it is determined that the path belongs to a medium-risk path, and the risk level is marked as R2;

[0126] When T L ≤ G < T M then it is determined that the path belongs to a low-risk path, and the risk level is marked as R1;

[0127] When G < T L then it is determined that the path belongs to a safe path, and the risk level is marked as RO;

[0128] Among them, T H is the high-risk threshold, T M is the medium-risk threshold, T L is the low-risk threshold.

[0129] The dynamic programming conflict prediction module includes a data input and parsing unit, a conflict conduction simulation unit, a spatial conflict identification unit, a conflict path tracing unit, and a conflict prediction report generation unit; among them:

[0130] The data input and parsing unit: is used to receive the externally input national land planning draft, extract the spatial location data and construction type data in the planning draft, and project the extracted data into three-dimensional space coordinates to form a spatial planning data set;

[0131] The conflict conduction simulation unit: is used to combine the set of conduction paths marked with risk levels output by the spatio-temporal anomaly conduction link mining module, overlay the planning data set onto the spatial association network, and use the newly added or adjusted nodes in the planning data set as the starting points, and simulate the influence diffusion process on the surrounding nodes according to the explicit connection relationship and the implicit conduction path to determine the potential abnormal diffusion influence range of the newly added nodes in the planning;

[0132] The spatial conflict identification unit: is used to compare the three-dimensional spatial position relationship between each spatial node in the spatial planning data set and the current spatial nodes, determine spatial overlap, spatial distance conflict, and conduction influence conflict, and clarify the conflict types of each conflict node, including position overlap conflict, distance conflict, and abnormal conduction conflict;

[0133] Conflict path tracing unit: For the planning nodes determined to have conduction influence conflicts, trace back the node sequence of the abnormal conduction link, identify and mark the starting node of the abnormal influence, the intermediate conduction nodes, and the final influence node, and determine the complete conflict conduction path;

[0134] Conflict pre-judgment report generation unit: Used to integrate the identified conflict types, the corresponding complete conduction paths, and the scope of abnormal diffusion influence, and generate a conflict pre-judgment report that intuitively displays the spatial conflict area, conduction path, and risk level in a three-dimensional visualization form for the program self-correction execution module to call; Through the gradual implementation of the above dynamic programming conflict pre-judgment module, it is possible to accurately simulate and identify the spatial location conflicts and abnormal conduction risks that may be caused by the planning scheme, clearly and intuitively present the conflict types and influence scope, and form a clear three-dimensional visualization pre-judgment report, providing accurate decision-making basis for the subsequent optimization and adjustment of the territorial space plan.

[0135] The program self-correction execution module includes a rule matching unit, a program generation unit, and a program output unit; among them:

[0136] Rule matching unit: Used to call the preset topological structure adjustment rule library according to the conflict type, and retrieve the adjustment rule set corresponding to the conflict type. The adjustment rules include specific topological structure correction strategies for location overlap conflicts, distance conflicts, and abnormal conduction conflicts;

[0137] Program generation unit: Based on the adjustment rule set output by the rule matching unit, perform topological structure optimization in the three dimensions of land use layout, facility distribution, and ecological boundary respectively to form at least three spatial adjustment plans with different optimization dimensions. Each plan retains the core requirement constraints on the original planning draft, and focuses on correcting high-risk nodes in combination with the conduction path analysis results;

[0138] Program output unit: Used to output the spatial adjustment plan output by the program generation unit in a visual form, and attach the comparison results of the topological structures before and after optimization for decision-makers to choose.

[0139] The preset topological structure adjustment rule library specifically includes the following rules:

[0140] Adjustment rule 1: When there is a spatial overlap conflict between the surface building unit and the boundary of the ecological sensitive area, move or shrink the building boundary towards the non-ecological sensitive area until the safety distance from the boundary of the ecological sensitive area meets the minimum distance requirement of the preset buffer zone;

[0141] Adjustment Rule 2: When there is a potential conflict between the underground facility unit and the boundary of the ecological sensitive area or important surface buildings, by adjusting the topological layout of the underground facility, its path is bypassed around the boundary of the sensitive area or the direct influence range of important buildings, ensuring that the distance between the two types of nodes is greater than the preset safety distance of the present invention;

[0142] Adjustment Rule 3: When the density of surface building units exceeds a predetermined threshold and may lead to a significant increase in the spatio-temporal anomaly amplification effect, by reducing the planned density of building units in a local area to no higher than the set density threshold to reduce the overall spatial conduction risk;

[0143] Adjustment Rule 4: When the facility construction in the planning draft may have an indirect impact on the ecological sensitive area, a buffer protection area is set outside the boundary of the ecological sensitive area. By increasing the width of the buffer protection area, the distance between the construction area and the boundary of the ecological sensitive area meets the preset safety buffer distance of the present invention, thereby blocking the formation of the abnormal conduction link;

[0144] Adjustment Rule 5: When the density of surface building units is too high and there is a conflict conduction with underground facilities, by diluting the density of the area with high building density, dispersing or adjusting the positions of building units, density balance is achieved, and the mutual influence intensity between spatial units is reduced;

[0145] Adjustment Rule 6: When the risk of explicit connection or implicit conduction path in the nodes involved in the planning draft exceeds the set risk level threshold, by appropriately translating the positions of high-risk nodes, their spatial positions are beyond the influence range of the abnormal conduction path or the topological distance from high-risk nodes is increased, thereby blocking or weakening the abnormal conduction;

[0146] Adjustment Rule 7: For areas where multiple nodes are highly dense and the anomaly amplification effect is obvious, some topological nodes are deleted or merged to reduce the number of topological connections, so as to reduce the cumulative amplification effect of local anomalies;

[0147] Adjustment Rule 8: When the topological structure of the conduction path significantly leads to an overly strong anomaly amplification effect, by changing the direction or topological relationship of the topological path, a new conduction path is constructed, increasing the length of the abnormal conduction path or exacerbating the conduction attenuation, thereby reducing the abnormal propagation efficiency.

[0148] The present invention covers any substitutions, modifications, equivalent methods, and solutions made within 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. However, those skilled in the art can fully understand the present invention without the description of these details. Additionally, to avoid unnecessary confusion to the essence of the present invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0149] 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 modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A territorial space data mining and management system based on big data analysis, characterized in that, It includes a spatial topology fingerprint generation module, a cross-modal association network construction module, a spatio-temporal abnormal conduction link mining module, a dynamic programming conflict prediction module, and a scheme self-correction execution module; among which: Spatial topology fingerprint generation module: It is used to receive the original national land spatial data, and generate a topologically unique fingerprint code through a non-Euclidean geometric feature extraction algorithm, and output a multi-dimensional spatial data set with a topology fingerprint identifier; Cross-modal association network construction module: Based on the multi-dimensional spatial data set, it conducts cross-layer association analysis on the surface building density, underground facility distribution, and ecological sensitive area boundary, and constructs a spatial association network including explicit connection and implicit conduction paths; Spatio-temporal abnormal conduction link mining module: It is used to perform two-way conduction simulation in the spatial association network, identify the conduction links with spatio-temporal abnormal amplification effects, and output a set of conduction paths marked with risk levels; Dynamic programming conflict prediction module: It is used to receive the externally input national land planning draft, combine the set of conduction paths to conduct a three-dimensional space conflict preview, and generate a visual prediction report including conflict types, conduction paths, and influence ranges; Scheme self-correction execution module: It is used to call the preset topology structure adjustment rule library according to the conflict types in the prediction report, and output at least three sets of spatial adjustment schemes with different optimization dimensions.

2. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that The spatial topology fingerprint generation module includes a data reception unit, a feature processing unit, a fingerprint code generation unit, and a data output unit; among which: Data reception unit: It is used to receive the original national land spatial data, and the original national land spatial data includes vector boundaries, remote sensing ground object features, and underground pipe network topologies; Feature processing unit: It is used to perform data normalization on the vector boundary, remote sensing ground object features, and underground pipe network topology respectively, and conduct quantitative analysis on curvature, connectivity, and topological adjacency relationship based on a non-Euclidean geometric feature extraction algorithm to obtain a feature matrix that can characterize the spatial structure difference; Fingerprint code generation unit: It is used to extract topologically unique fingerprint information according to the feature matrix, and digitally encode the fingerprint information according to the preset encoding rules to generate a topology fingerprint code; Data output unit: It is used to perform associative mapping on the topology fingerprint code and the original national land spatial data after normalization processing, and output a multi-dimensional spatial data set with a topology fingerprint identifier.

3. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that The feature processing unit includes: Data normalization processing: After receiving the vector boundary, remote sensing ground object features, and underground pipe network topology data, it uses the min-max normalization algorithm to perform numerical unification processing on the data; Curvature quantitative analysis: For the normalized vector boundary and remote sensing ground object feature data, it uses the discrete curvature calculation method to obtain the curvature value K of the spatial structure; Connectivity quantitative analysis: For the normalized underground pipe network topology data, based on the topological graph theory analysis algorithm, it calculates the connectivity index value C between the underground pipe network nodes; Topological adjacency relationship quantitative analysis: For the vector boundary, remote sensing ground object features, and underground pipe network topology data after data normalization processing, it calculates the topological adjacency relationship matrix A based on the topological adjacency matrix analysis method; Feature matrix generation: Based on the respectively calculated curvature value K, connectivity index value C, and topological adjacency relation matrix A, a complete spatial structure difference feature matrix M is constructed, and its expression is: M = [K, C, A].

4. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that The cross-modal association network construction module includes a cross-layer data mapping unit, a spatial association analysis unit, a hidden conduction path mining unit, and a network structure generation unit; among them: Cross-layer data mapping unit: Used to receive a multi-dimensional spatial data set with topological fingerprint identifiers, and perform unified coordinate projection on the surface building density, underground facility distribution, and ecological sensitive area boundary based on topological fingerprint coding to obtain a cross-modal spatial data mapping result; Spatial association analysis unit: Used to perform similarity analysis on the mapping result generated by the cross-layer data mapping unit, calculate the spatial correlation coefficient between the surface building density and the underground facility distribution, and perform topological matching on the adjacent relationship between the ecological sensitive area boundary and the underground facilities; Hidden conduction path mining unit: Based on the spatial correlation coefficient and adjacent relationship results output by the spatial association analysis unit, and using the shortest path conduction algorithm to determine the hidden conduction path that is not directly connected but has conduction between the surface building unit, underground facility unit, and ecological sensitive area boundary; Network structure generation unit: Used to construct a spatial association network according to the explicit connection relationship and the hidden conduction path, where the nodes are spatial units with topological fingerprint identifiers, and the edges are composed of the explicit connection relationship and the hidden conduction path, and perform differential marking on the explicit connection relationship and the hidden conduction path to generate a spatial association network including the explicit connection and the hidden conduction path.

5. The land spatial data mining and management system based on big data analysis according to claim 4, characterized in that The hidden conduction path mining unit includes: Node and edge construction: Respectively regard the surface building unit, underground facility unit, and ecological sensitive area boundary as nodes, and establish connectable edges between the nodes according to the topological adjacency or potential influence relationship to form a graph structure for conduction analysis; Edge weight determination: For any two connectable nodes, by comprehensively considering the feature similarity between the nodes and the geographical or topological distance elements, determine the edge weight used to measure the conduction cost; Shortest path search: In the constructed graph structure, for any two nodes that are not directly connected, traverse all feasible paths based on the shortest path conduction algorithm, and find the path with the smallest total conduction cost as the shortest conduction distance between the two; Hidden conduction determination: Compare the obtained shortest conduction distance with the pre-set conduction threshold. When the shortest conduction distance does not exceed the conduction threshold, it is determined that there is a hidden conduction path between the two nodes.

6. The land spatial data mining and management system based on big data analysis according to claim 5, characterized in that The spatio-temporal abnormal conduction link mining module includes a bidirectional conduction simulation unit, an abnormal amplification effect evaluation unit, a risk level determination unit, and a conduction path output unit; among them: Bidirectional conduction simulation unit: Used to receive the cross-modal association network, and through a pre-set spatio-temporal conduction model, take any node in the network as the initial abnormal source, and perform forward and reverse simulation conduction along the explicit connection relationship and the hidden conduction path to adjacent nodes respectively, calculate the cumulative value of node abnormal amplification in each path, and generate an abnormal conduction diffusion result; Abnormal Amplification Effect Evaluation Unit: It is used to analyze the abnormal conduction diffusion results of the bidirectional conduction simulation unit, calculate the conduction amplification coefficient based on the ratio of the abnormal intensity of the path start node to the intensity after the path end node receives the abnormality, and quantify the abnormal amplification effect of each conduction path; Risk Level Determination Unit: It is used to determine the risk level of each conduction path according to the conduction amplification coefficient of the Abnormal Amplification Effect Evaluation Unit. Specifically, it compares the conduction amplification coefficient with the preset risk level threshold, and determines the risk level mark corresponding to the path according to the interval where the ratio is located; Conduction Path Output Unit: It is used to output the set of conduction paths with risk level marks, where each conduction path records the complete node sequence, conduction direction, and corresponding risk level mark.

7. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that The bidirectional conduction simulation unit includes: Initial Abnormal Source Setting: Select any node in the cross-modal association network as the initial abnormal source, and assign the abnormal intensity value S0 to this node; Forward conduction simulation: Based on the explicit connection relationship and the implicit conduction path, starting from the initial abnormal source node, abnormal conduction is carried out to all directly connected adjacent nodes, and the abnormal intensity S received by each target node is calculated i ; Reverse conduction simulation: Using the end node of the target path as the reverse initial abnormal source, trace back along the explicit connection relationship and the implicit conduction path to calculate the abnormal conduction path, and determine the abnormal intensity S received by the node during the reverse conduction i ′; Abnormal amplification value calculation: Combine the abnormal intensities calculated in the forward and reverse conduction processes to calculate the cumulative abnormal amplification value of each path. The formula is: where F represents the cumulative abnormal amplification value of the path, N2 is the total number of nodes included in the path, S j and S j ′ are the forward and reverse abnormal intensities of the j-th node on the path, respectively.

8. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that The Abnormal Amplification Effect Evaluation Unit includes: Path anomaly intensity extraction: Select a conduction path in the spatial correlation network, and obtain the anomaly intensity S0 of the starting node of the path and the anomaly intensity S received by the end node of the path N ; Path Abnormal Loss Calculation: Calculate the cumulative abnormal loss value on the path, which is used to measure the degree of abnormal attenuation during the abnormal conduction process. The calculation formula is: L = S0 - S N , where L represents the total loss of the abnormality on the corresponding path; Conduction amplification factor calculation: Calculate the conduction amplification factor of a path, which is used to quantify the amplification or attenuation effect of the anomaly intensity at the starting node of the path relative to the anomaly intensity received at the ending node of the path. The calculation formula is as follows: Among them, G represents the conduction amplification factor of the path. If G > 1, it means that the anomaly is amplified on the path, and the amplification degree increases with the increase of G; if 0 < G < 1, it means that the anomaly attenuates during conduction on the path, and the attenuation degree increases with the decrease of G; if G = 1, it means that there is no loss during the conduction of the anomaly on the path.

9. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that The dynamic programming conflict prediction module includes a data input and parsing unit, a conflict conduction simulation unit, a spatial conflict identification unit, a conflict path tracing unit, and a conflict prediction report generation unit; among them: Data Input and Parsing Unit: It is used to receive the externally input national land planning draft, extract the spatial location data and construction type data in the planning draft, and project the extracted data into three-dimensional space coordinates to form a spatial planning data set; Conflict Conduction Simulation Unit: It is used to combine the set of conduction paths with risk level marks output by the spatio-temporal abnormal conduction link mining module, overlay the planning data set onto the spatial association network, and use the newly added or adjusted nodes in the planning data set as the starting points, and simulate the influence diffusion process on the surrounding nodes according to the explicit connection relationship and implicit conduction paths to determine the potential abnormal diffusion influence range of the newly added nodes in the plan; Spatial Conflict Identification Unit: It is used to compare the three-dimensional spatial position relationship between each spatial node in the spatial planning data set and the current spatial nodes, determine spatial overlap, spatial distance conflict, and conduction influence conflict, and clarify the conflict type of each conflict node, including position overlap conflict, distance conflict, and abnormal conduction conflict; Conflict Path Tracing Unit: It is used to trace the node sequence of the abnormal conduction link for the planning node determined to have a conduction influence conflict, identify and mark the abnormal influence start node, intermediate conduction nodes, and final influence nodes, and determine the complete conflict conduction path; Conflict Prediction Report Generation Unit: It is used to integrate the identified conflict types, corresponding complete conduction paths, and abnormal diffusion influence ranges, and generate a conflict prediction report that visually displays the spatial conflict area, conduction path, and risk level in a three-dimensional visualization form.

10. The land spatial data mining and management system based on big data analysis according to claim 1, characterized in that, The scheme self-correction execution module includes a rule matching unit, a scheme generation unit, and a scheme output unit; among them: Rule Matching Unit: It is used to call the preset topological structure adjustment rule library according to the conflict type, and retrieve the adjustment rule set corresponding to the conflict type; Scenario Generation Unit: Based on the adjustment rule set output by the Rule Matching Unit, perform topology optimization in three dimensions: land use layout, facility distribution, and ecological boundary, respectively, to form at least three spatial adjustment scenarios with different optimization dimensions; Scenario Output Unit: Used to output the spatial adjustment scenarios output by the Scenario Generation Unit in a visual form, along with the comparison results of the topology before and after optimization, for decision-makers to select.

Images