A land space planning conflict detection and intelligent auxiliary correction method and system

Abstract

The application discloses a kind of national space planning conflict detection and intelligent auxiliary correction method and system, belong to data processing technical field, it includes obtaining national space monitoring data, generates conflict situation parameter;Acquire national space conflict knowledge graph, query national space conflict knowledge graph based on conflict situation parameter, generate dynamic preplan;According to the influence range in conflict situation parameter, divide out core conflict area, peripheral influence area and un-affected area;Based on dynamic preplan, for the influence area divided respectively start emergency-short-term optimization model, medium-term optimization model and long-term optimization model, generate optimization strategy combination;Execute optimization strategy combination, and monitor execution process to generate feedback data, according to feedback data adjustment model optimization model parameter;Feedback data is used to update national space conflict knowledge graph.The application adopts partition optimization model, combines conflict evolution rule and feedback learning mechanism, and can realize the detection and adaptive correction of national space conflict.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for detecting and intelligently assisting in the correction of conflicts in land and space planning. Background Technology

[0002] Territorial spatial planning is an administrative activity aimed at achieving sustainable spatial development. Its core lies in coordinating the layout of functional spaces such as agriculture, ecology, and urban areas, with the goal of optimizing the pattern of territorial spatial development and protection. In this process, due to the inherent competition between different development goals and land use demands, various planning conflicts frequently occur, such as conflicts between farmland protection and urban expansion, and between ecological protection and economic development. Therefore, effectively detecting, analyzing, and correcting these conflicts is a crucial link in ensuring the smooth implementation of planning and improving governance capabilities.

[0003] Among related technologies, Chinese invention patent CN120297558B discloses a method and system for optimizing and analyzing land spatial planning based on big data. The method includes: collecting spatiotemporal data, storing it through a sliding window, performing anomaly detection, missing data completion, and smoothing filtering, dynamically adjusting normalization parameters, merging overlapping data according to weights, and continuously optimizing preprocessing based on data quality feedback; employing spatiotemporal weighted fuzzy clustering to partition real-time collected multi-source data according to spatial coordinates and time attributes, constructing a regular objective function using the weighted distance between data points and regional centers, and iteratively updating membership degrees and center positions; constructing a spatiotemporal prediction model, updating prediction errors and uncertainties in real time, dynamically adjusting weights, and outputting prediction results through weighted average integration; transforming land planning objectives into a multi-objective optimization problem, constructing a comprehensive objective function covering economic costs, environmental protection, and social benefit indicators, dynamically adjusting the weights of each objective based on real-time feedback, using augmented Lagrange and metaheuristic algorithms for real-time adaptive optimization and dynamic adjustment of the planning scheme, and collecting real-time data after planning implementation for comparison and error analysis with pre-predicted values.

[0004] However, the aforementioned existing technical solutions suffer from the following technical deficiencies: Passivity in conflict detection and response: Existing technologies focus on data preprocessing, clustering prediction, and multi-objective optimization, but lack proactive identification and dynamic early warning mechanisms for planning conflicts. They rely on historical data and prediction results for post-event adjustments, failing to generate conflict situation parameters in real time, leading to delayed conflict response and an inability to adapt to the sudden and evolving characteristics of territorial spatial conflicts. Coarse-grained strategy generation and lack of regional specificity: Existing technologies generate global planning schemes through multi-objective optimization, but fail to differentiate them according to the scope of conflict impact. Their strategies lack hierarchical optimization models for short-term, medium-term, and long-term emergencies, making it difficult to achieve precise resource allocation for conflict governance and failing to balance local emergency response with overall stability. Lack of knowledge evolution and adaptive capabilities: While existing technologies introduce real-time feedback to adjust target weights, they lack a dynamic assembly mechanism for conflict knowledge graphs and strategy atoms, failing to solidify successful conflict handling experiences into reusable evolutionary rules. The system lacks self-learning capabilities, making it difficult to adapt to the dynamic changes in territorial spatial conflict types, resulting in the inability to continuously accumulate and optimize governance experience. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a method and system for detecting and intelligently correcting conflicts in land spatial planning. By employing a zoning optimization model and combining conflict evolution rules with a feedback learning mechanism, it can achieve the detection and adaptive correction of land spatial conflicts.

[0006] The above objectives can be achieved through the following approach:

[0007] A method and system for detecting and intelligently assisting in the correction of conflicts in territorial spatial planning includes: acquiring territorial spatial monitoring data and generating conflict situation parameters; acquiring a territorial spatial conflict knowledge graph, querying the territorial spatial conflict knowledge graph based on the conflict situation parameters, and generating dynamic contingency plans; dividing the area into core conflict zones, peripheral impact zones, and unaffected zones based on the impact range in the conflict situation parameters; based on the dynamic contingency plans, activating emergency-short-term optimization models, medium-term optimization models, and long-term optimization models for the divided impact zones respectively, and generating optimization strategy combinations; executing the optimization strategy combinations, monitoring the execution process to generate feedback data, adjusting the optimization model parameters based on the feedback data; and updating the territorial spatial conflict knowledge graph using the feedback data.

[0008] Optionally, the dynamic assembly of the strategy atoms to generate a dynamic contingency plan includes: parsing the conflict situation parameters to obtain the conflict type, and matching the conflict evolution rules from the territorial spatial conflict knowledge graph according to the conflict type; selecting multiple strategy atoms from a preset strategy atom library according to the conflict evolution rules, and calculating the compatibility weights between each pair of the multiple strategy atoms; combining the multiple strategy atoms based on the compatibility weights, and arranging them to generate a dynamic contingency plan.

[0009] Optionally, the generation of the optimization strategy combination includes: for the core conflict zone, using the emergency-short-term optimization model and inheriting the dynamic contingency plan to generate a first optimization strategy; for the peripheral impact zone, using the medium-term optimization model and taking the first optimization strategy as input to generate a second optimization strategy to buffer the impact; for the unaffected zone, using the long-term optimization model and taking the first and second optimization strategies as constraints to generate a third optimization strategy to ensure the global objective; and integrating the first optimization strategy, the second optimization strategy, and the third optimization strategy to form an optimization strategy combination.

[0010] Optionally, adjusting the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the feedback data includes: extracting conflict reduction rate and measure implementation parameters from the feedback data; obtaining the expected reduction rate threshold; comparing the conflict reduction rate with the expected reduction rate threshold; and adjusting the target weights in the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the comparison result; and relaxing or tightening the constraints in the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the measure implementation parameters to obtain the adjusted optimization model parameters.

[0011] Optionally, updating the territorial spatial conflict knowledge graph includes: calculating the effectiveness score of the dynamic plan based on the feedback data and the adjusted optimization model parameters; obtaining an effectiveness threshold, comparing the effectiveness score with the effectiveness threshold, and when the effectiveness score exceeds the effectiveness threshold, solidifying the strategy atomic combination in the dynamic plan into a new conflict evolution mode; adding the new conflict evolution mode to the territorial spatial conflict knowledge graph, and updating the conflict evolution rules.

[0012] Optionally, the step of generating a first optimization strategy for the core conflict zone using the emergency-short-term optimization model and inheriting the dynamic contingency plan includes: simplifying the objective function of the emergency-short-term optimization model to the target value of the conflict impact and removing non-core constraints; solving the simplified model to generate a candidate strategy set; verifying the feasibility of the candidate strategy set and selecting the target optimization candidate strategy set as the first optimization strategy from the verified candidate strategy set.

[0013] Optionally, the integration of the first optimization strategy, the second optimization strategy, and the third optimization strategy includes: calculating the conflict index among the first optimization strategy, the second optimization strategy, and the third optimization strategy, and aggregating them into a total conflict index; comparing the total conflict index with a preset tolerance; when the total conflict index exceeds the tolerance, initiating a coordination optimization process to adjust the strategy parameters and generate a coordinated strategy combination; and outputting the coordinated strategy combination as an optimized strategy combination.

[0014] Optionally, generating conflict situation parameters includes: extracting spatial distribution features and time series features from the land space monitoring data to form multi-source data features; calculating a conflict probability score based on the multi-source data features; and determining conflict events and encapsulating relevant features based on the conflict probability score and a preset probability threshold to form conflict situation parameters.

[0015] Optionally, the step of extracting spatial distribution features and time series features from the land space monitoring data to form multi-source data features includes: performing spatial grid division and spatial autocorrelation analysis on the land space monitoring data to obtain spatial clustering and heterogeneity indicators; performing time series decomposition and trend analysis on the land space monitoring data to obtain periodic and trend change indicators; and performing feature fusion and standardization processing on the spatial clustering and heterogeneity indicators and the periodic and trend change indicators to form multi-source data features.

[0016] Based on the same inventive concept, this invention also provides a land spatial planning conflict detection and intelligent auxiliary correction system. The system includes: a conflict situation analysis module for acquiring land spatial monitoring data and analyzing the data to generate conflict situation parameters; a dynamic contingency plan generation module for acquiring a land spatial conflict knowledge graph containing strategy atoms and conflict evolution rules, querying the knowledge graph based on the conflict situation parameters, and dynamically assembling the strategy atoms to generate a dynamic contingency plan; a zoning module for dividing the conflict area into core conflict zone, peripheral impact zone, and unaffected zone based on the impact range in the conflict situation parameters; an optimized strategy combination generation module for generating optimized strategy combinations by activating emergency-short-term optimization models, medium-term optimization models, and long-term optimization models for the core conflict zone, peripheral impact zone, and unaffected zone, respectively, based on the dynamic contingency plan; a strategy execution and model tuning module for executing the optimized strategy combination, monitoring the execution process to generate feedback data, and adjusting the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the feedback data; and a feedback learning and knowledge graph evolution module for updating the land spatial conflict knowledge graph using the feedback data.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] This invention achieves a systematic improvement in the land spatial planning management model by constructing a data-driven closed-loop intelligent control system. This method can proactively identify and quantify potential conflicts based on multi-source monitoring data, transforming passive conflict resolution into proactive risk warning and management. This enhances the scientific rigor and foresight of planning administration, ensuring that management decisions are based on objective data rather than simply relying on outdated experience-based judgments.

[0019] This invention improves the accuracy of planning correction schemes and the efficiency of resource allocation by creating a strategy generation mechanism based on spatial differentiation and temporal phases. Emergency, medium-term, and long-term optimization models are activated for different areas affected by conflicts. The resulting strategy combinations can quickly respond to urgent issues in the core area while also ensuring the stability of peripheral areas and the achievement of overall goals, thus realizing refined and differentiated administrative management.

[0020] This invention, by introducing feedback learning and knowledge evolution mechanisms, endows the planning management system with the ability to self-improve and dynamically adapt. This method can learn from the actual results of conflict resolution, automatically adjust and optimize model parameters, and update the knowledge graph. It solidifies successful management experiences into reusable knowledge, making the entire administrative decision support system intelligent and efficient in continuous application, thereby adapting to the ever-changing needs of national land space development.

[0021] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a method for detecting and intelligently assisting in the correction of conflicts in land and space planning according to an embodiment of the present invention.

[0024] Figure 2 This is a strategy atomic compatibility weight heatmap of an embodiment of the present invention.

[0025] Figure 3 This is a feedback adjustment curve of the conflict reduction deceleration rate according to an embodiment of the present invention.

[0026] Figure 4This is a schematic diagram of the structure of a land spatial planning conflict detection and intelligent auxiliary correction system according to an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Reference Figure 1 One embodiment of the present invention proposes a method for detecting and intelligently correcting conflicts in land spatial planning. It adopts a zoning optimization model and combines conflict evolution rules and feedback learning mechanisms to achieve the detection and adaptive correction of conflicts in land spatial planning.

[0029] The method described in this embodiment specifically includes:

[0030] S1. Acquire land space monitoring data and analyze the land space monitoring data to generate conflict situation parameters;

[0031] Optionally, the generated conflict situation parameters include:

[0032] Spatial distribution features and time series features are extracted from the aforementioned land space monitoring data to form multi-source data features;

[0033] Based on the multi-source data features, a conflict probability score is calculated.

[0034] Based on the conflict probability score and the preset probability threshold, conflict events are identified and related features are encapsulated to form conflict situation parameters.

[0035] Specifically, spatial distribution characteristics and time-series characteristics need to be extracted from the aforementioned land space monitoring data to form multi-source data characteristics. The specific operations include spatial gridding and spatial autocorrelation analysis of the monitoring data to obtain spatial clustering and heterogeneity indicators. These indicators describe the spatial concentration and differences in land use types or change trends within the same region. Simultaneously, time-series decomposition and trend analysis are performed on the data to obtain periodic and trend change indicators, used to characterize the seasonal patterns and long-term evolution directions of conflict occurrence. Next, these spatial clustering and heterogeneity indicators, along with the periodic and trend change indicators, are fused and standardized to form a unified multi-source data feature vector.

[0036] Based on the features of the multi-source data, a conflict probability score is calculated. This score is obtained by predicting the features of the multi-source data through a pre-trained classification model. Its value is between zero and one, and is a dimensionless numerical value that represents the possibility of territorial spatial conflict occurring under specific spatiotemporal conditions.

[0037] Based on the conflict probability score and a preset probability threshold, conflict events are identified and their relevant features are encapsulated to form conflict situation parameters. When the calculated conflict probability score is greater than or equal to the preset probability threshold, a conflict event is immediately identified. At this point, all relevant information, such as the spatial location of the conflict, intensity index, the type of planning involved, and timestamp, is encapsulated to form conflict situation parameters that can be directly used in subsequent steps.

[0038] For example, the land space monitoring data includes satellite remote sensing imagery of a city's fringe area and quarterly updated records of farmland changes over the past five years. First, the system extracts spatial distribution characteristics and time series characteristics from the land space monitoring data. Through spatial autocorrelation analysis, the spatial clustering index of construction land expansion within the region is derived as follows: This indicates a high concentration of construction activities. Time series analysis reveals a trend of 5% annual growth in the conversion of arable land into construction land. These indicators are then fused to form multi-source data features. Subsequently, based on these multi-source data features, a conflict probability score is calculated. Using a pre-trained conflict prediction model, the conflict probability score for the region is calculated as follows: Finally, the system determines the conflict event based on the conflict probability score and a preset probability threshold. The preset probability threshold is... .because Greater than The system determines that there is a conflict between "farmland protection and urban construction" in the area, and immediately encapsulates the precise coordinates of the conflict, conflict intensity indicators such as the expected value of farmland loss, the type of planning involved, such as permanent basic farmland and time, etc., to form conflict situation parameters and output them.

[0039] Optionally, the step of extracting spatial distribution features and time series features from the land space monitoring data to form multi-source data features includes:

[0040] Spatial grid division and spatial autocorrelation analysis were performed on the aforementioned land space monitoring data to obtain spatial clustering and heterogeneity indicators;

[0041] The land space monitoring data were subjected to time series decomposition and trend analysis to obtain periodic and trend change indicators;

[0042] The spatial clustering and heterogeneity indicators are combined with the periodic and trend change indicators through feature fusion and standardization to form multi-source data features.

[0043] Specifically, the land space monitoring data needs to be spatially gridded and subjected to spatial autocorrelation analysis. Spatial gridding discretizes the land space monitoring data, dividing a continuous spatial area into a series of spatial units with uniform size and shape. For example, a city area can be divided into uniform rectangular grids. The purpose is to provide a unified spatial reference benchmark for subsequent quantitative analysis. The subsequent spatial autocorrelation analysis is a statistical method used to measure the correlation between attribute values ​​in one spatial unit and attribute values ​​in adjacent spatial units. Through this analysis, spatial clustering, which characterizes the degree of concentration of a certain type of land use change or conflict risk in geographic space, and heterogeneity indicators can be obtained, which quantify the attribute differences or uneven spatial distribution among spatial units.

[0044] The land space monitoring data is decomposed into time series and analyzed for trends. By statistically processing historical monitoring data, patterns such as seasonal changes and periodic fluctuations are extracted to obtain periodic change indicators. By fitting the long-term evolution direction of the data, trend change indicators are obtained, such as the average annual change rate of illegal land use area.

[0045] The spatial clustering and heterogeneity indicators, along with the periodicity and trend change indicators, undergo feature fusion and standardization. Feature fusion integrates all extracted spatial and temporal indicators to form a complete, multi-dimensional feature vector. Standardization transforms all indicators in this feature vector to a uniform scale, eliminating the impact of differences in scales and ensuring that all features have equal weight and comparability in the subsequent conflict probability calculation model. Ultimately, this process forms multi-source data features.

[0046] For example, the monitoring data consists of monthly records of illegal sewage discharge and the expansion of surrounding industrial land in a key wetland protected area over the past five years. First, the system performs spatial gridding and spatial autocorrelation analysis on the aforementioned land spatial monitoring data. The wetland and surrounding area are divided into grids, and the intensity of illegal sewage discharge in each grid is calculated. Through spatial autocorrelation analysis, the spatial clustering degree of illegal sewage discharge behavior is determined. This indicates that sewage discharge is highly concentrated on one side of the wetland; and a heterogeneity index is obtained, indicating significant differences in sewage discharge intensity within the clustered area. Next, the system performs time-series decomposition and trend analysis on the aforementioned land spatial monitoring data. Analysis of five years of sewage discharge data reveals that sewage discharge intensity increases after the high-water season, yielding a periodicity index; simultaneously, through fitting, it is found that the total sewage discharge caused by industrial land expansion increases by 3% annually, yielding a trend index. Finally, the system performs feature fusion and standardization processing on the spatial clustering and heterogeneity indices, along with the periodicity and trend indices. All features, including clustering, heterogeneity, periodicity, and trend indices, are combined and uniformly standardized to eliminate numerical differences, ultimately forming a multi-source data feature vector and outputting it.

[0047] S2. Obtain a knowledge graph of territorial spatial conflicts containing strategy atoms and conflict evolution rules, query the knowledge graph of territorial spatial conflicts based on the conflict situation parameters, and dynamically assemble the strategy atoms to generate dynamic contingency plans.

[0048] Optionally, the dynamic assembly of the strategy atoms to generate dynamic plans includes:

[0049] The conflict type is obtained by analyzing the conflict situation parameters, and the conflict evolution rules are obtained by matching the conflict type from the territorial spatial conflict knowledge graph.

[0050] According to the conflict evolution rule, multiple policy atoms are selected from the preset policy atom library, and the compatibility weights between each pair of the multiple policy atoms are calculated.

[0051] The multiple strategy atoms are combined based on the compatibility weights and arranged to generate a dynamic plan.

[0052] Specifically, conflict status parameters are analyzed to determine the conflict type. These parameters, generated in previous steps through analysis of land and space monitoring data, contain relevant characteristics of conflict events. Analysis clarifies the specific category of the conflict currently facing land and space planning, such as "ecological protection zones being encroached upon by construction land" or "conflict between farmland protection and infrastructure construction land demand." Based on the obtained conflict type, conflict evolution rules are matched against a land and space conflict knowledge graph. This knowledge graph pre-stores knowledge containing strategy atoms and conflict evolution rules. These rules describe the possible development trends and response strategies for a particular conflict type under specific conditions.

[0053] Multiple strategy atoms are selected from a pre-defined strategy atom library based on the matched conflict evolution rules. Strategy atoms are the basic units that constitute a solution, such as "adjusting land use boundaries," "increasing ecological compensation standards," or "restricting development intensity." After selection, the pairwise compatibility weights between the multiple strategy atoms need to be calculated. A heatmap of strategy atom compatibility weights is shown below. Figure 2 As shown. Compatibility weight Represents the strategy atom and strategy atoms When applied simultaneously, the degree to which they promote or inhibit each other can be obtained through methods such as expert scoring, historical data analysis, or knowledge graph reasoning.

[0054] The multiple strategy atoms are combined based on the aforementioned compatibility weights, and a dynamic contingency plan is generated through their arrangement. When combining strategy atoms, it is necessary to ensure that the overall compatibility of the selected strategies is maximized, enabling the strategies to cooperate and improve the efficiency and effectiveness of conflict resolution. The dynamic contingency plan is a set of executable strategies, rapidly generated and arranged by priority or execution order, based on the current conflict situation and its evolutionary trend.

[0055] For example, after analyzing the land space monitoring data of a certain region, the generated conflict situation parameters, after parsing, determine that the current conflict type is "conflict caused by the occupation of basic farmland by non-agricultural construction." Based on this conflict type, the corresponding conflict evolution rule is matched from the land space conflict knowledge graph. This rule indicates that for this type of conflict, the focus should be on "restoring arable land" and "optimizing land use structure." According to this rule, three strategy atoms are selected from the strategy atom library: strategy atoms To adjust construction land indicators and strategy atoms To implement land consolidation and reclamation projects and strategic atom To establish a compensation mechanism for farmland protection, the next step is to calculate the pairwise compatibility weights. Assume the calculation results are: strategy atoms. and Compatibility weight A value of 0.8 indicates high compatibility, as adjusting the construction land quota can free up space for reclamation projects; strategy atom and Compatibility weight A value of 0.5 indicates medium compatibility; strategy atoms and Compatibility weight A compatibility score of 0.9 indicates extremely high compatibility, as compensating farmers after land reclamation would better facilitate the project. Based on these compatibility weights, dynamic contingency plans are generated by combining and arranging them. Considering compatibility and rule requirements, the final dynamic contingency plan might be: Step 1: Execute the strategy. Adjusting indicators to control incremental conflicts, the second step is to implement the strategy. Reclamation is carried out to reduce existing conflicts; the third step is to implement the strategy. Establish a long-term protection mechanism.

[0056] S3. Based on the influence range in the conflict situation parameters, divide the core conflict zone, the outer influence zone, and the unaffected zone;

[0057] Specifically, the process of dividing the conflict zone into core conflict zone, peripheral impact zone, and unaffected zone based on the influence range in the conflict situation parameters first requires resolving the spatial coordinates of the conflict source and the geometric boundary information of the conflict impact from the conflict situation parameters generated in the previous steps. Then, spatial analysis techniques are used to calculate the conflict impact index of each spatial unit within the planning area. The impact index of the conflict Characterizes location The degree of interference from conflict events is calculated by combining the intensity parameter of the conflict source with a spatial distance attenuation function, ensuring dimensional consistency. A first-level threshold is set. Second grade threshold ,in Greater than The calculated conflict impact index Compare each of these two thresholds one by one. If a spatial unit's... Greater than or equal to the first grade threshold If so, the unit is determined to belong to the core conflict zone, which is the area where the conflict is most intense and requires emergency intervention. If a spatial unit's Less than the first grade threshold And greater than or equal to the second-level threshold If so, the unit is determined to belong to the outer influence zone, which is affected by the radiation of the core conflict and requires medium-term strategies for buffering and guidance. If a certain spatial unit's Less than the second grade threshold If the condition is met, the unit is determined to be an unaffected area, and the area will be maintained under normal management.

[0058] S4. Based on the dynamic contingency plan, activate the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model for the core conflict zone, the peripheral impact zone, and the unaffected zone, respectively, and generate a combination of optimization strategies.

[0059] Optionally, the combination of generation optimization strategies includes:

[0060] For the core conflict zone, the first optimization strategy is generated by using the emergency-short-term optimization model and inheriting the dynamic contingency plan;

[0061] For the aforementioned peripheral influence zone, a second optimization strategy for buffering the influence is generated using the mid-term optimization model and taking the first optimization strategy as input.

[0062] For the unaffected region, the long-term optimization model is used, and a third optimization strategy is generated to ensure the global objective, constrained by the first and second optimization strategies.

[0063] The first optimization strategy, the second optimization strategy, and the third optimization strategy are integrated to form an optimization strategy combination.

[0064] Specifically, an emergency short-term optimization model is activated for the core conflict zone. This model directly inherits the strategy atoms from the dynamic contingency plan generated in the previous steps, aiming to maximize the speed of conflict mitigation, and generates the first optimization strategy by solving for local optima.

[0065] A mid-term optimization model is initiated for the peripheral impact zone. This model takes the first optimization strategy as an unchangeable input parameter, sets the carrying capacity threshold of the buffer zone, calculates governance measures that can withstand the pressure of the core area and prevent the spread of conflict, and thus generates a second optimization strategy.

[0066] For the unaffected area, a long-term optimization model is initiated. This step takes the first and second optimization strategies as known boundary constraints, and plans the long-term development path of the area under the premise of ensuring that the macro-planning goals of the region are not destroyed, and generates the third optimization strategy.

[0067] The three strategies described above are spatially and temporally combined to form a complete optimized strategy portfolio. To ensure coordination among the strategies, an overall policy utility function is defined:

[0068] ,

[0069] in, These represent the first, second, and third optimization strategies, respectively. The emergency response efficiency value of the first optimization strategy. The buffer performance value of the second optimization strategy. The three function values ​​are all normalized dimensionless values, ranging from 0 to 1, representing the global programming compliance value of the third optimization strategy. The corresponding weight coefficients, and satisfying These weights are dynamically set based on the urgency of the conflict. By maximizing this utility function, the overall superiority of the strategy combination can be verified. This method achieves a smooth transition from local emergency response to global stability through partitioned and hierarchical progressive optimization, effectively preventing secondary damage to the overall national spatial planning caused by single-dimensional governance measures.

[0070] For example, in a scenario where a chemical plant leak leads to an ecological conflict, the preliminary steps system has already generated a first optimization strategy for the core conflict zone of the leaking plant area. For example, emergency production shutdowns and pollution source containment are secondary optimization strategies targeting the surrounding farmland and river areas. Examples include: setting up interception dams, initiating soil remediation, and third optimization strategies for unaffected areas in other urban functional zones. Examples include adjusting industrial layout and relocating high-risk enterprises. To verify the superiority of this strategy combination, an overall strategy utility function is used for evaluation. The weighting coefficients are... The settings are based on the urgency and long-term impact of the conflict, respectively. ,satisfy The system obtains the utility value of each strategy by post-processing and normalizing the solution results of the three hierarchical optimization models: First optimization strategy Emergency response efficiency value for This value, derived from indicators such as conflict deceleration rate, indicates excellent emergency response effectiveness; the second optimization strategy buffer performance value for This indicates that it has a good effect in controlling the spread of pollutants to the surrounding areas; the third optimization strategy Global planning compliance value for This indicates a low degree of deviation from the city's long-term planning goals. Substituting these values ​​into the utility function, the calculation is performed as follows: The final overall strategy utility value obtained. This high numerical value quantitatively supports the scientific nature and coordination of the strategy combination, indicating that the generated optimized strategy combination has a satisfactory overall effect in taking into account emergency response, medium-term buffer and long-term development goals.

[0071] Optionally, the step of generating a first optimization strategy for the core conflict zone using the emergency-short-term optimization model and inheriting the dynamic contingency plan includes:

[0072] The objective function of the emergency-short-term optimization model is simplified to the objective value of the conflict impact, and non-core constraints are removed;

[0073] The simplified model is solved to generate a set of candidate strategies;

[0074] The feasibility of the candidate strategy set is verified, and the target optimization candidate strategy set is selected from the verified candidate strategy set as the first optimization strategy.

[0075] Specifically, the objective function of the emergency short-term optimization model needs to be simplified to the target value of conflict impact, and non-core constraints need to be removed. The emergency short-term optimization model is specifically designed for core conflict areas, and its primary goal is to quickly control the conflict. Therefore, the original multi-objective function of the model, such as considering long-term economic benefits or social equity, is simplified to a single-objective function maximizing the conflict deceleration rate. The conflict deceleration rate is the rate of change of conflict intensity over time, representing the decrease in conflict intensity per unit time. This value is calculated by inheriting the strategy atoms and related parameters from the dynamic contingency plan. Simultaneously, considering the urgency of the core conflict area, non-core constraints are removed, retaining only core safety and resource constraints such as maximum response time or emergency resource budget. This significantly improves the model's solution efficiency and response speed without affecting the emergency response effect.

[0076] The simplified model is solved using a fast optimization algorithm, such as a heuristic algorithm or simplified linear programming, to quickly calculate all action plans that meet the conditions while retaining the core constraints, thus generating a candidate strategy set.

[0077] The candidate strategy set is subjected to feasibility verification to ensure that each candidate strategy can be implemented immediately in terms of resources, time, and regulations. The target optimization candidate strategy set is then selected as the first optimization strategy from the verified candidate strategy set. The final selected first optimization strategy is the optimal combination among all feasible solutions that achieves the fastest conflict reduction speed.

[0078] For example, a sudden, high-risk hazardous waste dumping pollution conflict occurred in a core conflict area. The system has inherited a dynamic contingency plan containing "immediate cleanup" and "site containment" strategy atoms. The system simplifies the short-term emergency optimization model by setting the objective function to maximize the amount of hazardous waste removed per unit time, removing non-core constraints and retaining only core constraints such as the maximum emergency response time of forty-eight hours. Subsequently, the simplified model is solved, generating three candidate strategy sets: strategy Strategy for removing 80% of hazardous waste within 48 hours Strategy for removing 75% of hazardous waste within 36 hours The plan was to remove 85% of the hazardous waste within 30 hours, but this exceeded the emergency budget. A feasibility study was then conducted. Due to the strategy... This exceeded the emergency resource budget constraint and was deemed infeasible. Strategy and All verifications were successful. Finally, the system... and Select the strategy with the higher degree of objective optimization. A relatively higher cleanup rate was achieved in a shorter time; therefore, the system selected a strategy. Optimize the candidate policy set for the target and output it as the first optimization policy.

[0079] Optionally, the integration of the first optimization strategy, the second optimization strategy, and the third optimization strategy includes:

[0080] Calculate the conflict index among the first optimization strategy, the second optimization strategy, and the third optimization strategy, and aggregate them into a total conflict index;

[0081] The total conflict index is compared with a preset tolerance level. When the total conflict index exceeds the tolerance level, a coordination optimization process is initiated to adjust the strategy parameters and generate a coordinated strategy combination.

[0082] The coordinated strategy combination is output as the optimized strategy combination.

[0083] Specifically, it is necessary to calculate the conflict index among the first, second, and third optimization strategies and aggregate them into a total conflict index. The conflict index refers to the degree to which different optimization strategies mutually hinder each other and negatively impact resource requirements, spatial boundaries, or temporal execution. For any two strategies... and Conflict indicators between Its calculation formula is set as follows:

[0084] ,

[0085] In the formula, and Each represents a strategy atom. It is the resource overlap, and its value comes from the strategy. and The analysis of the overlap of key resources such as required funds and land, with values ​​ranging from zero to one, represents the intensity of resource competition. It is a spatial-temporal inconsistency, the value of which is obtained through spatial analysis or temporal scheduling analysis, quantifying the degree of contradiction in the spatial layout or execution timing of the strategy, with a value range from zero to one. and These are the corresponding weighting coefficients, used to balance the importance of resource competition and spatial-temporal contradictions, and Adding the conflict metrics of all strategy pairs together yields the total conflict metric:

[0086] ,

[0087] The total conflict index is then compared with a preset tolerance level. This refers to the maximum degree of inter-strategy conflict that the system can tolerate; it is an empirical threshold. When the total conflict index... Exceeding the tolerance level When this happens, the system initiates a coordination optimization process to adjust strategy parameters. The coordination optimization process is a multi-objective optimization solution process. Its goal is to minimize the total conflict index. By fine-tuning strategy parameters, such as adjusting the execution time window, changing land use boundaries, or reallocating resources, a coordinated strategy combination is generated.

[0088] The coordinated strategy combination is output as the optimized strategy combination. This method effectively avoids conflicts and internal friction that may arise during the integration of zoning optimization strategies by quantifying and controlling the inherent contradictions between strategies. It ensures that the final output optimized strategy combination has a high degree of internal consistency and synergy, thereby improving the overall effectiveness and stability of territorial spatial governance.

[0089] For example, if the system has already generated a first optimization strategy Ecological restoration of the core area and the second optimization strategy Strategies for the construction of tourism supporting facilities in the outer areas and the third optimization strategy To ensure industrial upgrading in areas not affected. (Calculation strategy) and Conflict indicators between At that time, ecological restoration was discovered With tourism development There are land use type conflicts on specific plots of land, and competition exists for the same scarce aggregate resource. A weighting coefficient is set. and Analysis revealed resource overlap. The determination is as This indicates high resource competition and spatial-temporal inconsistency. The determination is as This indicates a spatial contradiction. Calculation. Other conflict indicators were calculated as follows: for , for Then the total conflict index The system's preset tolerance. for .because Exceed The system initiates a coordination and optimization process. This process adjusts the strategy parameters: the second optimization strategy... The tourism facility land area will be set back fifty meters to avoid the ecological restoration boundary, and and The strategy involves staggered supply of aggregate resources to meet peak demand. and Conflict indicators Down to New total conflict index for .because Less than The system outputs this coordinated strategy combination as the optimized strategy combination.

[0090] S5. Execute the combination of optimization strategies, monitor the execution process to generate feedback data, and adjust the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the feedback data;

[0091] Optionally, adjusting the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the feedback data includes:

[0092] Extract the conflict reduction rate and measure implementation parameters from the feedback data;

[0093] Obtain the desired deceleration rate threshold, compare the conflict deceleration rate with the desired deceleration rate threshold, and adjust the target weights in the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model according to the comparison result;

[0094] Based on the implementation parameters of the measures, the constraints in the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model are relaxed or tightened to obtain the adjusted optimization model parameters.

[0095] Specifically, two key parameters need to be precisely extracted from the feedback data generated during the optimization strategy execution process: the conflict reduction rate and the measure implementation rate. Conflict reduction rate This refers to the rate of decrease in conflict intensity per unit of time. Its dimension is conflict intensity divided by time, and it can be obtained by analyzing the rate of change in land use type, ecological value, or economic loss in the conflict-affected area. (Measures implementation parameters) It is a dimensionless indicator that quantifies the matching degree between the progress of strategy implementation, resource consumption and expected goals. Its value is usually obtained by monitoring the funding arrival rate, project completion rate or policy implementation rate.

[0096] The system obtains the desired deceleration rate threshold. This threshold is a preset minimum acceptable standard used to guide the effectiveness of conflict management. The actual monitored conflict reduction rate will be used as the benchmark. With the expected deceleration rate threshold Compare. If Less than This indicates that the current optimization strategy's objective weight allocation is insufficient to quickly resolve conflicts, and the system will increase the objective weight of the conflict reduction objective in the emergency short-term optimization model, medium-term optimization model, and long-term optimization model. Weight adjustment amount It can be calculated using the formula:

[0097] ,

[0098] in It is a dimensionless adjustment coefficient used to control the magnitude of weight changes.

[0099] Based on the implementation parameters of the measures The constraints in the optimization model are dynamically adjusted. A lower value indicates that policy execution is limited by resources or time, and the system will relax the relevant constraints in the model; if If the conflict reduction is high but not ideal, it may be necessary to tighten the constraints to improve resource utilization efficiency, thereby obtaining the adjusted optimization model parameters. The conflict reduction deceleration rate feedback adjustment curve is shown below. Figure 3 As shown.

[0100] For example, in a conflict scenario where urban green space is encroached upon by commercial development, after implementing the optimization strategy for a period of time, the system extracts the following parameter from the feedback data: conflict reduction rate. for Hectares / month represents the area of ​​green space that can be restored each month. The system's preset expected deceleration rate threshold. for hectares / month. Meanwhile, the monitored measures implementation parameters... for This indicates that the actual execution efficiency or resource availability rate of the strategy is only 60%. First, a system comparison... and ,Discover This means the actual reduction rate is lower than the expected value, requiring an increase in the target weight. An adjustment coefficient is used. Calculate the adjustment amount of the target weight. If the original weight of the objective "green space restoration" in the mid-term optimization model is The adjusted new target weights Will increase , become This increases the model's priority for green space restoration in the next iteration. Secondly, due to the implementation parameter... Only The system determined that low execution was the main reason for the ineffective strategy implementation, likely stemming from resource constraints such as funding or personnel. Therefore, the system relaxed the "annual funding cap" constraint in the mid-term optimization model, for example, by increasing its cap by 20%, allowing for the mobilization of more resources to accelerate project progress. Ultimately, the system output included an increase in the weight of the "green space restoration" objective. The optimized model parameters are adjusted with the "annual capital investment ceiling" constraint relaxed.

[0101] S6. Update the knowledge graph of territorial spatial conflicts using the feedback data.

[0102] Optionally, updating the territorial spatial conflict knowledge graph includes:

[0103] Based on the feedback data and the adjusted optimization model parameters, the effectiveness score of the dynamic contingency plan is calculated.

[0104] Obtain an effectiveness threshold, compare the effectiveness score with the effectiveness threshold, and when the effectiveness score exceeds the effectiveness threshold, solidify the strategy atom combination in the dynamic plan into a new conflict evolution mode;

[0105] The new conflict evolution model is added to the territorial spatial conflict knowledge graph to update the conflict evolution rules.

[0106] Specifically, based on the feedback data and the adjusted optimization model parameters, the effectiveness score of the dynamic contingency plan is calculated. The validity score It is a comprehensive evaluation indicator designed to quantify the actual effectiveness of the dynamic contingency plan in resolving specific conflicts. Its calculation formula is set as follows:

[0107] ,

[0108] In the formula, It is the normalized value of the conflict reduction rate, reflecting the direct effect of governance. Its value is calculated from the conflict reduction rate extracted from the feedback data, and its range is from 0 to 1. This is a parameter for the degree of implementation of measures, which reflects the operability of the strategy and the degree of resource matching. Its value is also obtained from the feedback data, and the value ranges from 0 to 1. It is a model adjustment factor used to penalize excessive reliance on model parameter adjustments. Its value is calculated based on the degree of deviation between the adjusted optimized model parameters and the initial parameters, and is expressed as a dimensionless penalty term. These are weighting coefficients used to balance the importance of the three indicators, and they satisfy... .

[0109] After completing the score calculation, the system obtains the validity threshold. This threshold is a preset minimum standard for judging whether a strategy combination is worth fixing, and the effectiveness score is calculated accordingly. With the validity threshold Compare the results. When the validity score is... Exceeding the validity threshold If the dynamic contingency plan is deemed efficient and robust, then the strategy atom combination in the dynamic contingency plan is solidified into a new conflict evolution mode. For example, "strategy atom A + strategy atom B" is identified as the standard procedure for handling a certain type of conflict.

[0110] This proven successful experience for a specific conflict type was then formally added to the Territorial Spatial Conflict Knowledge Graph as a knowledge unit. The Territorial Spatial Conflict Knowledge Graph is a knowledge base storing strategy atoms and conflict evolution rules, serving as the knowledge foundation for the system to formulate solutions. Updating conflict evolution rules means that nodes or relationships related to that specific conflict type in the knowledge graph are modified, enhanced, or expanded.

[0111] For example, in response to a conflict arising from the encroachment of tourist facilities on an ecological reserve in a certain area, the system generates a dynamic contingency plan comprising two atomic strategies: "restricting tourist flow" and "ecological restoration." After monitoring the implementation of these strategies, the normalized value of the conflict mitigation rate is extracted from the feedback data. for ; Measure implementation parameters for Because the adjustment range of the optimized model parameters is small, the calculated model adjustment factor is... for Set weighting coefficients. The effectiveness score of this dynamic contingency plan is calculated as follows: The system's preset validity threshold for Due to the calculated validity score Greater than the validity threshold The system determined that the strategy atomic combination was successful and efficient. Therefore, the strategy atomic combination of "limiting tourist flow + ecological restoration" was solidified into a new conflict evolution model and added to the territorial spatial conflict knowledge graph as a conflict evolution rule for handling the conflict of "ecological protection area being encroached by tourism facilities".

[0112] Based on the same inventive concept, such as Figure 4 As shown, the present invention also provides a land spatial planning conflict detection and intelligent auxiliary correction system, the system comprising:

[0113] The conflict situation analysis module is used to acquire land space monitoring data, analyze the land space monitoring data, and generate conflict situation parameters.

[0114] The dynamic contingency plan generation module is used to acquire a territorial spatial conflict knowledge graph containing strategy atoms and conflict evolution rules, query the territorial spatial conflict knowledge graph based on the conflict situation parameters, and dynamically assemble the strategy atoms to generate a dynamic contingency plan.

[0115] The partitioning module is used to divide the core conflict zone, the peripheral impact zone, and the unaffected zone based on the impact range in the conflict situation parameters.

[0116] The optimization strategy combination generation module is used to generate optimization strategy combinations by activating emergency-short-term optimization models, medium-term optimization models, and long-term optimization models for the core conflict zone, the peripheral impact zone, and the unaffected zone, respectively, based on the dynamic contingency plan.

[0117] The strategy execution and model tuning module is used to execute the combination of optimization strategies, monitor the execution process to generate feedback data, and adjust the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model and long-term optimization model based on the feedback data.

[0118] The feedback learning and knowledge graph evolution module is used to update the territorial spatial conflict knowledge graph using the feedback data.

[0119] It should be noted that the electrical connections between the various units described above do not necessarily represent direct or indirect connections. Any indirect connection method can be applied to the embodiments of the present invention as long as it achieves the purpose of the present invention. The above descriptions are merely exemplary embodiments of the present invention and should not be construed as limiting the scope of the present invention.

[0120] All equivalent changes and modifications made in accordance with the teachings of this invention are still within the scope of this invention. Those skilled in the art will readily conceive of other embodiments of this invention upon considering the specification and the disclosure of practical truth. This application is intended to cover any variations, uses, or adaptations of this invention that follow the general principles of this invention and include common knowledge or conventional techniques in the art not described herein.

Claims

1. A method for detecting and intelligently assisting in the correction of conflicts in land spatial planning, characterized in that, The method includes: Acquire land space monitoring data and analyze the land space monitoring data to generate conflict situation parameters, including: extracting spatial distribution features and time series features from the land space monitoring data to form multi-source data features; calculating conflict probability scores based on the multi-source data features; and determining conflict events and encapsulating relevant features based on the conflict probability scores and preset probability thresholds to form conflict situation parameters. A knowledge graph of territorial spatial conflicts containing strategy atoms and conflict evolution rules is obtained. Based on the conflict situation parameters, the knowledge graph is queried, and the strategy atoms are dynamically assembled to generate a dynamic contingency plan. This includes: parsing the conflict situation parameters to obtain conflict types, and matching conflict evolution rules from the territorial spatial conflict knowledge graph based on the conflict types; selecting multiple strategy atoms from a preset strategy atom library according to the conflict evolution rules, and calculating the compatibility weights between each pair of strategy atoms; combining the multiple strategy atoms based on the compatibility weights, and arranging them to generate a dynamic contingency plan. Based on the influence range in the conflict situation parameters, the core conflict zone, the outer influence zone, and the unaffected zone are divided. Based on the dynamic contingency plan, emergency-short-term optimization model, medium-term optimization model and long-term optimization model are activated for the core conflict zone, the peripheral impact zone and the unaffected zone, respectively, to generate a combination of optimization strategies. The optimization strategy combination is executed, and feedback data is generated during the execution process. The optimization model parameters of the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model are adjusted based on the feedback data. The feedback data is used to update the knowledge graph of territorial spatial conflicts.

2. The method for detecting and intelligently assisting in the correction of conflicts in land spatial planning according to claim 1, characterized in that, The combination of generation optimization strategies includes: For the core conflict zone, the first optimization strategy is generated by using the emergency-short-term optimization model and inheriting the dynamic contingency plan; For the aforementioned peripheral influence zone, a second optimization strategy for buffering the influence is generated using the mid-term optimization model and taking the first optimization strategy as input. For the unaffected region, the long-term optimization model is used, and a third optimization strategy is generated to ensure the global objective, constrained by the first and second optimization strategies. The first optimization strategy, the second optimization strategy, and the third optimization strategy are integrated to form an optimization strategy combination.

3. The method for detecting and intelligently assisting in the correction of conflicts in land spatial planning according to claim 1, characterized in that, The adjustment of the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model based on the feedback data includes: Extract the conflict reduction rate and measure implementation parameters from the feedback data; Obtain the desired deceleration rate threshold, compare the conflict deceleration rate with the desired deceleration rate threshold, and adjust the target weights in the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model according to the comparison result; Based on the implementation parameters of the measures, the constraints in the emergency-short-term optimization model, medium-term optimization model, and long-term optimization model are relaxed or tightened to obtain the adjusted optimization model parameters.

4. The method for detecting and intelligently assisting in the correction of conflicts in land spatial planning according to claim 1, characterized in that, The update of the territorial spatial conflict knowledge graph includes: Based on the feedback data and the adjusted optimization model parameters, the effectiveness score of the dynamic contingency plan is calculated. Obtain an effectiveness threshold, compare the effectiveness score with the effectiveness threshold, and when the effectiveness score exceeds the effectiveness threshold, solidify the strategy atom combination in the dynamic plan into a new conflict evolution mode; The new conflict evolution model is added to the territorial spatial conflict knowledge graph to update the conflict evolution rules.

5. The method for detecting and intelligently assisting in the correction of conflicts in land spatial planning according to claim 2, characterized in that, The first optimization strategy generated for the core conflict zone, using the emergency-short-term optimization model and inheriting the dynamic contingency plan, includes: The objective function of the emergency-short-term optimization model is simplified to the objective value of the conflict impact, and non-core constraints are removed; The simplified model is solved to generate a set of candidate strategies; The feasibility of the candidate strategy set is verified, and the target optimization candidate strategy set is selected from the verified candidate strategy set as the first optimization strategy.

6. The method for detecting and intelligently assisting in the correction of conflicts in land spatial planning according to claim 2, characterized in that, The integration of the first optimization strategy, the second optimization strategy, and the third optimization strategy includes: Calculate the conflict index among the first optimization strategy, the second optimization strategy, and the third optimization strategy, and aggregate them into a total conflict index; The total conflict index is compared with a preset tolerance level. When the total conflict index exceeds the tolerance level, a coordination optimization process is initiated to adjust the strategy parameters and generate a coordinated strategy combination. The coordinated strategy combination is output as the optimized strategy combination.

7. The method for detecting and intelligently assisting in the correction of conflicts in land spatial planning according to claim 1, characterized in that, The extraction of spatial distribution features and time series features from the land space monitoring data to form multi-source data features includes: Spatial grid division and spatial autocorrelation analysis were performed on the aforementioned land space monitoring data to obtain spatial clustering and heterogeneity indicators; The land space monitoring data were subjected to time series decomposition and trend analysis to obtain periodic and trend change indicators; The spatial clustering and heterogeneity indicators are combined with the periodic and trend change indicators through feature fusion and standardization to form multi-source data features.

8. A land spatial planning conflict detection and intelligent auxiliary correction system, applied to the land spatial planning conflict detection and intelligent auxiliary correction method as described in any one of claims 1-7, characterized in that, The system includes: The conflict situation analysis module is used to acquire land space monitoring data and analyze the land space monitoring data to generate conflict situation parameters. This includes: extracting spatial distribution features and time series features from the land space monitoring data to form multi-source data features; calculating a conflict probability score based on the multi-source data features; and determining conflict events and encapsulating relevant features based on the conflict probability score and a preset probability threshold to form conflict situation parameters. A dynamic contingency plan generation module is used to acquire a territorial spatial conflict knowledge graph containing strategy atoms and conflict evolution rules, query the territorial spatial conflict knowledge graph based on the conflict situation parameters, and dynamically assemble the strategy atoms to generate a dynamic contingency plan. This includes: parsing the conflict situation parameters to obtain the conflict type, and matching the conflict type with the conflict evolution rules from the territorial spatial conflict knowledge graph; selecting multiple strategy atoms from a preset strategy atom library according to the conflict evolution rules, and calculating the compatibility weights between each pair of strategy atoms; combining the multiple strategy atoms based on the compatibility weights, and arranging them to generate a dynamic contingency plan. The partitioning module is used to divide the core conflict zone, the peripheral impact zone, and the unaffected zone based on the impact range in the conflict situation parameters. The optimization strategy combination generation module is used to generate optimization strategy combinations by activating emergency-short-term optimization models, medium-term optimization models, and long-term optimization models for the core conflict zone, the peripheral impact zone, and the unaffected zone, respectively, based on the dynamic contingency plan. The strategy execution and model tuning module is used to execute the combination of optimization strategies, monitor the execution process to generate feedback data, and adjust the optimization model parameters of the emergency-short-term optimization model, medium-term optimization model and long-term optimization model based on the feedback data. The feedback learning and knowledge graph evolution module is used to update the territorial spatial conflict knowledge graph using the feedback data.

Images