Land space planning toughness identification and evaluation method

Abstract

The invention discloses a territorial space planning toughness identification and evaluation method, and belongs to the technical field of territorial space planning. The method comprises the following steps: step 1, acquiring spatial information, social economic data and ecological environment elements of an evaluation area; 2, dividing functional units as network nodes, and constructing a weighted adjacent matrix to form a complete spatial relation network structure; 3, determining an index weight in combination with an analytic hierarchy process and an entropy evaluation method; 4, calculating the toughness index of each node, and comprehensively evaluating the overall toughness level of the network in combination with network topology analysis; 5, identifying key weak nodes and links based on a network topology structure and node importance analysis; step 6, adopting a multi-objective optimization algorithm to generate candidate schemes of specific improvement measures, and utilizing a Monte Carlo simulation method to evaluate the feasibility and effect of each candidate scheme to determine an optimal implementation path; and step 7, forming an evaluation-improvement-re-evaluation loop optimization mechanism.

Description

Technical Field

[0001] The present application relates to the technical field of national land space planning, and more specifically, to a method for identifying and evaluating the resilience of national land space planning. Background Art

[0002] National land space planning is an important means to achieve regional sustainable development. Its main goals are to rationally allocate resources, optimize land use, improve ecological environment quality and enhance social and economic resilience. However, with the acceleration of urbanization and the continuous changes in the natural environment, traditional national land space planning methods have gradually exposed many limitations. In particular, planning resilience is particularly important in dealing with natural disasters, emergencies and social and economic shocks.

[0003] At present, most national land space planning evaluation systems focus on land use efficiency, ecological protection and economic development, but the resilience evaluation for dealing with emergencies, climate change and other systemic risks is relatively insufficient. Traditional planning methods often rely too much on historical data and static spatial layout, and lack the ability to predict future changes. With climate change, environmental degradation and increasing global economic uncertainty, traditional planning models have been unable to cope with complex risks and challenges.

[0004] Resilience is a systematic and dynamic characteristic, as the ability of a system to cope with external pressures and maintain functional and structural stability. In national land space planning, resilience evaluation needs to comprehensively consider multiple dimensions such as ecology, society, economy, and infrastructure to evaluate the adaptability and recovery capacity of various types of space. At present, although some studies have attempted to establish a resilience evaluation model, in practical applications, there is a lack of comprehensive, systematic and operational resilience identification and evaluation methods. These methods usually ignore the interactions between different spatial units, the dynamic changes within the system, and the impact of external shocks, resulting in insufficient accuracy and applicability of resilience evaluation results.

[0005] To sum up, how to accurately identify and evaluate planning resilience in national land space planning, especially in dealing with complex risks, natural disasters and emergencies, has become a technical problem that needs to be solved urgently. Summary of the invention

[0006] In order to overcome a series of defects in the prior art, the purpose of this application is to provide a method for identifying and evaluating the resilience of national land space planning in response to the above problems, including the following steps:

[0007] Step 1: Obtain spatial information, socio-economic data and ecological and environmental elements of the evaluation area;

[0008] Step 2: Based on administrative boundaries, natural geographical features, and land use types, functional units are divided as network nodes, and a weighted adjacency matrix is ​​constructed to quantify the intensity of material flow, energy flow, and information flow between nodes to form a complete spatial relationship network structure;

[0009] Step 3: Establish a resilience evaluation system for nodes and networks;

[0010] Step 4: Calculate the resilience index of each node and comprehensively evaluate the overall resilience level of the network;

[0011] Step 5: Identify key weak points and links, and determine the risk level based on the preset graded warning thresholds;

[0012] Step 6: Use a multi-objective optimization algorithm to generate candidate solutions for specific improvement measures, and use the Monte Carlo simulation method to evaluate the feasibility and effectiveness of each candidate solution to determine the optimal implementation path;

[0013] Step 7: Form a cyclic optimization mechanism of evaluation-improvement-re-evaluation.

[0014] Preferably, step 1 comprises the following steps:

[0015] Acquire the latest high-resolution remote sensing image data, including satellite and aerial images, and perform geometric correction and image enhancement processing;

[0016] Extract and classify land use types, vegetation coverage and water distribution elements from remote sensing images;

[0017] Collect socio-economic statistics of the assessment area and ensure the timeliness and authority of the data;

[0018] Conduct field research through GPS positioning, on-site photography and questionnaire surveys to verify the accuracy of remote sensing interpretation results and supplement ground information that is difficult to obtain through remote sensing;

[0019] Integrate remote sensing interpretation results, socio-economic statistics and field survey data to build a geographic information database of the evaluation area to achieve unified management and analysis of multi-source data.

[0020] Preferably, step 2 comprises the following steps:

[0021] The evaluation area is divided into several functional units based on the integrity of administrative boundaries, continuity of physical geographical elements, and homogeneity of land use types;

[0022] Marking attribute information for each functional unit;

[0023] By analyzing the material flow paths, energy exchange channels and information dissemination networks within the region, the association relationships between functional units can be identified;

[0024] Based on the intensity of material flow, energy flow and information flow, the association weights between adjacent units are calculated and a weighted adjacency matrix is ​​constructed;

[0025] Combine all functional units and weighted adjacency matrices to build a complete spatial relationship network structure;

[0026] The connection strength between nodes and regional flow patterns are verified to ensure that the network structure can accurately reflect the actual functional relationships and flow characteristics of the region.

[0027] Preferably, step 3 comprises the following steps:

[0028] Based on the basic framework of carrying capacity theory, starting from the three dimensions of social development, economic growth and ecological environment, we select key evaluation indicators that can objectively reflect the characteristics of node and network resilience;

[0029] Normalization method was used to convert all indicators into the same dimensional range;

[0030] Construct a judgment matrix and determine the subjective weight of each indicator through the analytic hierarchy process, and perform consistency test at the same time;

[0031] Based on the discrete degree of each indicator data, the objective weight of each indicator is calculated using the entropy method;

[0032] Combining subjective weights and objective weights, the geometric mean method is used to calculate the combined weight of each indicator;

[0033] Based on key evaluation indicators and their combined weights, a comprehensive evaluation framework for node and network resilience is constructed.

[0034] Preferably, step 4 comprises the following steps:

[0035] Realize the automatic collection, transmission and storage of key node indicators;

[0036] Regularly collect and update the operation data of each node and establish a dynamic monitoring database;

[0037] Perform weighted calculation on each indicator to achieve quantitative calculation of single node resilience index;

[0038] Calculate network topology characteristic parameters, evaluate network connectivity, clustering coefficient and modularity, and analyze the impact of network topology on overall resilience;

[0039] Realize quantitative characterization and dynamic evaluation of the overall resilience level of the network.

[0040] Preferably, step 5 comprises the following steps:

[0041] Monitor the changing trend of node resilience indicators in real time, use the moving average method to identify abnormal fluctuations that significantly deviate from the normal range, and trigger corresponding early warning signals;

[0042] Build a node importance evaluation system to identify key nodes in the network;

[0043] Evaluate the vulnerability of nodes from the dimensions of node resilience index, failure frequency, and repair time, while considering the correlation effect between nodes to identify potential weak links;

[0044] Establish a hierarchical and graded warning threshold system and make dynamic adjustments based on actual operating conditions;

[0045] Based on the early warning threshold system, combined with the degree of node abnormality and the scope of impact, the identified risks are graded and classified.

[0046] Preferably, step 6 comprises the following steps:

[0047] Utilize graph neural networks to capture dynamic relationships between nodes, build a deep learning framework that combines node attributes and topological structures, and quantitatively analyze the transmission path and diffusion effect of risk shocks;

[0048] Based on the goal of improving resilience, a multi-objective optimization model is established by comprehensively considering economic costs, social benefits and environmental impacts, and a series of candidate solutions with different optimization focuses are generated using the particle swarm algorithm;

[0049] Develop specific improvement measures covering infrastructure upgrades, land use structure adjustments, urban functional layout optimization, and resource allocation efficiency improvements to target different levels of vulnerability, and form a systematic optimization solution library;

[0050] Using the Monte Carlo simulation method, through a large number of random scenario simulation experiments, we evaluate the implementation effects and risk response capabilities of different optimization schemes under various uncertain conditions, and build a scheme evaluation index system;

[0051] Taking into account the feasibility, economy and implementation difficulty of the plan, combined with the simulation evaluation results, the hierarchical analysis method is used to comprehensively rank the candidate plans, select the optimal plan and formulate a phased implementation roadmap.

[0052] Preferably, step 7 comprises the following steps:

[0053] Establish a multi-level evaluation framework covering the evaluation index system, network topology structure and node resilience improvement effect, and design standardized evaluation processes and methods to achieve a comprehensive quantitative evaluation of node and network improvement effects;

[0054] Regularly collect and analyze node operation data and network status data, and quantitatively evaluate the actual effects of various improvement measures, including the degree of improvement in node resilience index, the enhancement of network risk prevention and control capabilities, and the improvement level of overall network performance, by combining comparative analysis before and after implementation and trend research;

[0055] Based on the effect evaluation results, the node warning threshold is adaptively adjusted and the weight of the node evaluation index is recalculated to ensure that the evaluation system can timely reflect the dynamic changes of node status and network structure;

[0056] According to the effect evaluation results, dynamically adjust the priority of existing node improvement measures and network optimization solutions, optimize resource allocation, and timely update and improve the improvement measures library in combination with newly discovered node problems and network requirements;

[0057] Build a closed-loop optimization mechanism of assessment-improvement-reassessment, and through continuous monitoring, assessment and dynamic adjustment, continuously improve the node resilience level and network risk response capabilities, and achieve continuous improvement of node and network performance.

[0058] Compared with the prior art, the beneficial effects of this application are:

[0059] This application combines remote sensing images, statistical data and field surveys, uses network topology analysis and multi-objective optimization algorithms, and establishes a comprehensive national land space planning resilience identification and evaluation method, achieving comprehensive evaluation and dynamic optimization of spatial relationship networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of a method for identifying and evaluating the resilience of national land space planning is provided as an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments.

[0062] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0063] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0064] Please refer to Figure 1, a method for identifying and evaluating the resilience of national land space planning, comprising the following steps:

[0065] Step 1: Obtain spatial information, socio-economic data and ecological and environmental elements of the evaluation area through remote sensing image interpretation, statistical data collection and field investigation;

[0066] Step 2: Based on administrative boundaries, natural geographical features, and land use types, functional units are divided as network nodes, and a weighted adjacency matrix is ​​constructed to quantify the intensity of material flow, energy flow, and information flow between nodes to form a complete spatial relationship network structure;

[0067] Step 3: Determine the indicator weights by combining the analytic hierarchy process and the entropy method, thereby establishing a node and network resilience evaluation system;

[0068] Step 4: Calculate the resilience index of each node based on the resilience evaluation system, and comprehensively evaluate the overall resilience level of the network in combination with network topology analysis;

[0069] Step 5: Monitor abnormal fluctuations in node resilience indicators, identify key weak nodes and links based on network topology and node importance analysis, and determine risk levels based on preset graded warning thresholds;

[0070] Step 6: Use a multi-objective optimization algorithm to generate candidate solutions for specific improvement measures, and use the Monte Carlo simulation method to evaluate the feasibility and effectiveness of each candidate solution to determine the optimal implementation path;

[0071] Step 7: Regularly evaluate the effectiveness of the implementation of improvement measures, and dynamically adjust the warning threshold, evaluation indicator weights, and the priority and implementation plan of improvement measures based on the evaluation results, forming a cyclic optimization mechanism of evaluation-improvement-re-evaluation.

[0072] In step 1, remote sensing image interpretation, statistical data collection and field research provide support for basic data collection, covering spatial information, socio-economic data and ecological and environmental elements. This process provides comprehensive basic data for subsequent analysis. The accuracy and timeliness of these data directly affect the scientificity and reliability of the entire evaluation system. Remote sensing images provide key features such as spatial distribution and land use types by acquiring regional information over a wide range, while statistical data and field research supplement these macro data, ensuring the multi-dimensional, precise and dynamic nature of spatial planning data.

[0073] Step 2 divides the functional units and establishes a weighted adjacency matrix, taking these units as network nodes, quantifying the intensity of material flow, energy flow and information flow between nodes, and forming a complete spatial relationship network structure. This step quantifies the relationship between different regions through network modeling methods, and then provides a more intuitive understanding framework to describe the mutual influence and linkage between these regions. This weighted adjacency matrix not only considers physical spatial relationships, but also includes factors such as information transmission and energy exchange, making the analysis of regional resilience more comprehensive, especially in the face of complex systemic risks, it helps to reveal the potential vulnerabilities and advantages between regions.

[0074] Step 3 uses the analytic hierarchy process and entropy method to determine the weights of each indicator in the evaluation system. This step solves the problem of how to fairly and scientifically assign different weights to each influencing factor through a quantitative method. The analytic hierarchy process can ensure that the weight of each factor reflects its importance in the system based on the hierarchical structure and interrelationship of the indicators; the entropy law allocates weights based on the amount of information in the data itself, avoiding the interference of human subjective factors, thereby improving the objectivity and accuracy of the evaluation system.

[0075] Step 4 calculates the resilience index of each node through the resilience evaluation system, and combines it with the network topology analysis to comprehensively evaluate the overall resilience level of the network. This step quantifies the stress resistance and recovery capacity of each node through the resilience index, making the resilience assessment from local to overall clearer and more operational. Combined with the network topology analysis, it can reveal which nodes are critical in the network, thereby optimizing the resilience management strategy and ensuring that important areas are given priority protection.

[0076] Step 5 monitors abnormal fluctuations in node resilience indicators, identifies key weak nodes based on network topology and node importance analysis, and determines risk levels using preset thresholds. This step uses dynamic monitoring and risk identification to timely identify potential risk nodes and avoid systemic failures. By dividing risk levels, detailed risk prediction reports and early warning measures can be provided to ensure that emergency management can respond quickly and intervene effectively.

[0077] Step 6 uses a multi-objective optimization algorithm to generate candidate solutions for improvement measures, and uses the Monte Carlo simulation method to evaluate the feasibility and effectiveness of each solution, and finally determines the optimal implementation path. The multi-objective optimization algorithm can comprehensively consider the trade-offs of different goals, optimize resource allocation, and maximize the benefits of improvement measures while meeting multiple needs. Monte Carlo simulation provides a possible effect estimate of each solution through a large number of random experiments, making the decision more scientific and accurate.

[0078] Step 7: Through a continuous feedback loop mechanism, the entire evaluation and improvement process can be continuously optimized. Through regular effect evaluation, problems in implementation can be discovered and adjusted in a timely manner, thereby improving the adaptability of the plan and the effectiveness of implementation.

[0079] In summary, the national land space planning resilience identification and evaluation method forms a complete evaluation-optimization closed loop through a series of precise technical means and scientific methods, from data collection to analysis, and then to the generation and implementation of improvement plans. The implementation of each step is closely linked to ensure the efficiency and systematicness of the entire process, and ultimately achieve the goal of optimizing regional planning and improving resilience. This method is not only highly practical, but can also be dynamically adjusted according to actual conditions, providing a scientific basis for the sustainable development of various cities and regions.

[0080] Furthermore, step 1 includes the following steps:

[0081] Obtain the latest high-resolution remote sensing image data, including satellite images and aerial images, and perform geometric correction and image enhancement processing. This step provides an efficient and comprehensive spatial information foundation for regional planning through remote sensing images. The combination of satellite images and aerial images can obtain ground information from different angles and resolutions to ensure comprehensive and detailed data coverage. Geometric correction ensures that the image can accurately correspond to the geographic coordinate system, which is crucial for subsequent spatial analysis. Image enhancement processing adjusts image contrast, clarity and other parameters through algorithms to make the features of the ground objects more prominent, thereby improving the availability and accuracy of remote sensing images. This stage of processing lays a solid foundation for the subsequent extraction of information such as land use and vegetation cover.

[0082] Based on the spectral characteristics and texture characteristics of the objects, combined with the human-computer interactive interpretation method, the land use type, vegetation cover and water distribution elements of the remote sensing images are extracted and classified. The spectral characteristics and texture characteristics of the objects are the basis of remote sensing image analysis. The spectral characteristics can identify different types of objects (such as water bodies, forests, cities, etc.), while the texture characteristics help to identify more subtle differences in objects (such as soil types and vegetation density). The human-computer interactive interpretation method combines manual judgment with automated analysis to make the classification process more accurate and flexible, and can overcome the limitations of traditional methods in the classification of complex objects. This method improves the accuracy of remote sensing image classification, helps to accurately extract key geographic information such as land use, vegetation cover and water distribution, and provides reliable data for subsequent spatial analysis and resilience assessment.

[0083] Collect socioeconomic statistics of the evaluation area to ensure the timeliness and authority of the data. Socioeconomic data is an important factor in assessing regional resilience, as it directly affects resource allocation, social development, and regional carrying capacity. This step improves the accuracy and comparability of the data by ensuring the timeliness and authority of the data. Socioeconomic statistics usually include indicators such as population density, economic development level, infrastructure construction, and social service capacity, which can provide support for subsequent resilience analysis. By timely updating and obtaining reliable socioeconomic data, it is ensured that the method can keep up with the current social and economic development trends, provide a true reflection, and provide a basis for formulating effective regional resilience enhancement strategies.

[0084] Field investigations are carried out through GPS positioning, on-site photography records and questionnaire surveys to verify the accuracy of remote sensing interpretation results and supplement ground information that is difficult to obtain through remote sensing. This process makes the interpretation results of remote sensing images more reliable and comprehensive through on-site verification and supplementation. Remote sensing images are often affected by factors such as weather, seasonal changes and resolution, and some ground features may be difficult to accurately identify. Therefore, field investigations can make up for the shortcomings of remote sensing data. GPS positioning provides accurate geographical location references, on-site photography records intuitively reflect ground conditions, and questionnaire surveys can obtain detailed information on land use, environmental conditions and social economy. Through the combination of these means, the results of remote sensing image interpretation can be effectively verified and corrected, and the accuracy and applicability of the data can be improved.

[0085] Integrate remote sensing interpretation results, socio-economic statistics and field research data to build a geographic information database for the evaluation area to achieve unified management and analysis of multi-source data. This step integrates data from different sources to form a multi-dimensional, highly operable database to provide support for subsequent analysis. This database contains not only spatial data, socio-economic data and environmental data, but also various attribute information related to the assessment, which can facilitate data query and analysis in different dimensions. Through unified management and efficient storage, data processing speed can be accelerated, decision-making efficiency can be improved, and a reliable platform for sustainable use and subsequent updates of data can be provided.

[0086] In general, step 1 has laid a solid foundation for the subsequent resilience identification and evaluation through the acquisition and processing of high-resolution remote sensing images, the interpretation of spectral and texture features, and the collection and field investigation of socioeconomic data. The fine operation and technical processing of each link ensure the accuracy and comprehensiveness of the data, thus providing reliable support for subsequent analysis. By effectively integrating multi-source data and building a unified geographic information database, efficient use of data can be achieved, providing accurate and reliable data support for the resilience evaluation of the entire national space planning, and ultimately improving the scientific nature and sustainability of regional planning.

[0087] Further, step 2 includes the following steps:

[0088] Based on the integrity of administrative boundaries, the continuity of natural geographical elements, and the homogeneity of land use types, the evaluation area is divided into several functional units. This step uses a reasonable division method to make each functional unit represent a regional unit with internal consistency and external relevance. Administrative boundaries provide a basic partitioning framework, while natural geographical features (such as rivers, mountains, etc.) help to divide areas with strong natural connections and ensure that each functional unit has strong connectivity within the ecosystem. At the same time, the homogeneity of land use types can also ensure that the functional units in the same area are consistent in terms of resource utilization, ecological protection, and social economy. This division method not only provides accurate spatial units for subsequent flow analysis, but also makes each functional unit more structurally consistent with the natural and social characteristics of the actual area.

[0089] Attribute information is calibrated for each functional unit, including geographic spatial coordinates, area statistics, population density distribution, land use classification and resource utilization status. This step provides detailed attribute information for each unit so that the spatial characteristics and socio-economic characteristics of each functional unit can be accurately quantified and used for analysis. This stage not only involves basic spatial data processing (such as coordinate and area calculation), but also includes the calibration of data such as population distribution, land use type, and resource utilization status. Through these attribute information, the resource carrying capacity, social service function, and ecological environment status of each functional unit can be intuitively reflected, providing important data support for the subsequent analysis of material flow, energy flow, and information flow. This process ensures the accuracy and reliability of subsequent network construction and analysis.

[0090] By analyzing the material flow paths, energy exchange channels and information dissemination networks in the region, the associations between the functional units are identified. Material flow involves the migration and exchange of resources, such as the flow of water, air, soil and other materials; energy flow involves energy exchange within the region, such as power supply, solar energy absorption, etc.; information flow involves information dissemination in social and economic activities, such as transportation and communication. By identifying these flow channels, the interaction between materials, energy and information within the region can be accurately described, and the close connection and dependency between the functional units can be further revealed. This analysis provides a theoretical basis for the subsequent construction of the weighted adjacency matrix, so that the network can more realistically reflect the actual flow characteristics of the region.

[0091] Based on the intensity of material flow, energy flow and information flow, the association weights between adjacent units are calculated, and a weighted adjacency matrix is ​​constructed. By constructing a weighted adjacency matrix, the intensity of material flow, energy flow and information flow is quantified to form structured data that can be used for calculation. The weighted adjacency matrix can clearly show the relationship between functional units, where each element in the matrix represents the flow intensity or dependence between adjacent units. This weighted method can reflect the flow of resources and information between different units in the region, helping us to evaluate the relative importance of different units and their contribution to the resilience and stability of the entire system in subsequent analysis. This process enables the correlation between functional units in the region to be quantified into specific values, which is convenient for subsequent calculations and model analysis.

[0092] All functional units and weighted adjacency matrices are combined to construct a complete spatial relationship network structure, in which nodes represent functional units and edge weights represent the intensity of flow between units. This process can better demonstrate the connection and flow characteristics between functional units in the region by constructing a complete spatial relationship network. The nodes in the network represent each functional unit, and the edge weights reflect the flow intensity of matter, energy, and information between these units. Through this network structure, a systematic analysis of the flow pattern within the region, the interaction between functional units, and the importance of different units can be carried out. The clear presentation of network relationships provides a solid foundation for subsequent topological analysis and resilience assessment, and can help decision makers intuitively understand the interdependence between functional units in each region.

[0093] The network topology analysis method is used to verify the connection strength between nodes and the regional flow pattern to ensure that the network structure can accurately reflect the actual functional relationship and flow characteristics of the region. This step verifies the constructed network structure through network topology analysis to confirm whether it can truly reflect the actual functional relationship and flow pattern of the region. The network topology analysis method can reveal the connection pattern between nodes, such as the relationship between dense regional networks and isolated nodes, and the primary and secondary context of regional flow. Through this verification, the feasibility and accuracy of the spatial relationship network in practical applications can be ensured, providing data basis for subsequent resilience assessment and network optimization.

[0094] In summary, step 2 can systematically reveal the relationship and interdependence between functional units by scientifically dividing the region and constructing a weighted adjacency matrix. The technical processing of each link provides an accurate data basis for the subsequent resilience assessment. From the division of functional units to the calibration of attributes, and then to the in-depth analysis of material flow, energy flow and information flow, the constructed spatial relationship network provides strong support for the flow characteristics and resilience of the assessment region. Through network topology analysis, the network structure can be verified and optimized to ensure that the final constructed model can accurately reflect the actual functional relationship of the region, providing a theoretical basis for the realization of scientific national land space planning and resilience improvement.

[0095] Furthermore, the association weight between adjacent units is expressed as: W ij =A ij ·((α·(F ij M / Σ (i,j)∈N F ij M )+β·(F ij E / Σ (i,j)∈N F ij E )+γ·(F ij I / Σ (i,j)∈N F ij I ))·(1-δd ij ))·C ij , where W ij A represents the association weight between unit i and unit j; ij Indicates whether unit i is adjacent to unit j, 1 if adjacent, 0 if not, that is, only the adjacent units are weighted; α, β, γ are weight coefficients, which are used to adjust the relative importance of material flow, energy flow and information flow; F ij M ,F ij E ,F ij I Respectively represent the intensity of material flow, energy flow and information flow; d ij is the distance between unit i and unit j; C ij is the association strength between unit i and unit j; N represents the set of all adjacent unit pairs; δ is the distance attenuation coefficient, which is used to describe the effect of distance on flow intensity; Σ (i,j)∈N F ij M Represents unit i and all adjacent units j

[0096] The sum of the material flows between is used to normalize the material flow intensity; Σ (i,j)∈N Fij E Represents unit i and all adjacent units j

[0097] The sum of the energy flows between is used to normalize the energy flow intensity; Σ (i,j)∈N F ij M It represents the sum of the information flow between unit i and all adjacent units j, and is used to normalize the strength of the information flow.

[0098] The association weights calculated by the above formula can accurately quantify the interaction and flow intensity between functional units in the region. In this process, multidimensional factors such as material flow, energy flow and information flow are taken into account, and distance decay effect and normalization processing are introduced to better reflect the actual connection between units in a complex geographical space and socio-economic background. The construction of the weighted adjacency matrix not only provides effective data support for subsequent network topology analysis, but also can optimize the characteristics of different regions by flexibly adjusting the weight coefficients to ensure the accuracy and reliability of the evaluation results.

[0099] Further, step 3 includes the following steps:

[0100] Based on the basic framework of carrying capacity theory, starting from the three dimensions of social development, economic growth and ecological environment, select key evaluation indicators that can objectively reflect the resilience characteristics of nodes and networks. These key evaluation indicators should comprehensively consider social, economic and environmental factors to accurately describe the carrying capacity of each node and the overall resilience of the network. For example, in the dimension of social development, indicators such as social stability and public service accessibility can be selected; in the dimension of economic growth, indicators such as industrial structure optimization and economic vitality can be selected; in the dimension of ecological environment, environmental pollution index and ecosystem resilience can be selected. This step can ensure the comprehensiveness and accuracy of the resilience assessment system in different fields by reasonably selecting evaluation indicators, and at the same time provide data basis for subsequent weight calculation and analysis.

[0101] The normalization method is used to convert all indicators into the same dimensional range, providing a basis for subsequent weight calculation and analysis. In order to ensure the consistency of the dimensions of each evaluation indicator, it is necessary to use the normalization method to convert all indicators into the same dimensional range. This is usually processed through common normalization methods such as the maximum and minimum method and the standard deviation method. Through normalization, the data of each indicator can be converted into a dimensionless value, eliminating the impact caused by dimensional differences, so that different types of indicators can be compared and weighted on the same basis. The technical effect of this step is reflected in the fact that the normalized data not only eliminates the inconsistency of dimensions, but also provides standardized numerical input for subsequent weight calculation and analysis, ensuring the operability and scientificity of the evaluation process.

[0102] Construct a judgment matrix and determine the subjective weight of each indicator through the hierarchical analysis method, and perform consistency tests to ensure the rationality of the judgment matrix. In the hierarchical analysis method, experts will compare each indicator two by two to evaluate its relative importance, thereby constructing a judgment matrix. Then, a consistency test is performed to ensure the rationality of the judgment matrix. The consistency test is usually completed by calculating the consistency ratio (CR). If the CR value is less than 0.1, the judgment matrix has good rationality. The hierarchical analysis method can scientifically convert the subjective judgment of experts into weights, and through consistency tests, ensure the compliance of the matrix and avoid evaluation errors caused by subjective bias.

[0103] Based on the degree of discreteness of each indicator data, the entropy method is used to calculate the objective weight of each indicator to fully reflect the objective characteristics of the data. After determining the subjective weight, it is also necessary to consider the objective characteristics of each indicator data. At this time, the entropy method is used to calculate the objective weight of each indicator. The entropy method determines the weight of each indicator by measuring the degree of discreteness of each indicator. The greater the discreteness, the greater the amount of information of the indicator, and its weight in the comprehensive evaluation should also be higher. The advantage of the entropy method is that it can fully reflect the objectivity of the data and does not rely on the subjective judgment of experts. The technical effect is reflected in that the entropy method reasonably allocates objective weights by quantifying the distribution characteristics of indicators, making the evaluation system more data-driven and objective.

[0104] Combining subjective weights and objective weights, the geometric mean method is used to calculate the combined weight of each indicator. The advantage of the geometric mean method is that it can balance subjective weights and objective weights, avoid over-reliance on the weight of one aspect, and ensure the fairness of the comprehensive evaluation. The calculation steps of the geometric mean method are: first normalize the subjective weight and objective weight of each indicator, and then calculate the combined weight of each indicator through geometric mean. The geometric mean method can reasonably combine expert experience and data analysis results to ensure that the evaluation model is both theoretical and has actual data support.

[0105] Based on key evaluation indicators and their combined weights, a comprehensive evaluation framework for node and network resilience is constructed. This framework comprehensively reflects the carrying capacity of each node in the region and the overall resilience level of the network by integrating the evaluation results of various dimensions. In specific applications, this framework can be used to evaluate the sustainability of regional development and the ability to respond to emergencies. By establishing a comprehensive and detailed evaluation framework, it can provide a scientific basis for decision makers, support the formulation of different policies and measures, and promote the realization of sustainable development goals in the region.

[0106] Through the above steps, the comprehensive evaluation framework of node and network resilience can comprehensively reflect the carrying capacity of each functional unit in the region and the resilience characteristics of the network through scientific indicator selection, reasonable normalization processing, effective weight calculation and comprehensive evaluation methods. The combination of subjective and objective weights, the application of hierarchical analysis method and entropy method ensure the fairness and rationality of the evaluation process and avoid the deviation that may be caused by a single evaluation method. Ultimately, this framework can not only provide an accurate evaluation tool for regional development, but also provide a basis for policymakers to promote coordination and sustainable development in all aspects of the region.

[0107] Furthermore, a judgment matrix is ​​constructed and the subjective weight of each indicator is determined by the hierarchical analysis method, and a consistency test is performed to ensure the rationality of the judgment matrix, including the following steps:

[0108] The Delphi method was used to organize experts to conduct pairwise comparison of the importance of indicators, obtain the relative importance ratios between indicators, and preliminarily construct the judgment matrix A = [a pq ], where a pq Represents the expert's subjective judgment of indicator p relative to indicator q, satisfying a pq >0, a qp =1 / a pq 、a pp =1, where a qp represents the expert's subjective judgment on indicator q relative to indicator p; a pp Indicates the comparison of the indicator with itself;

[0109] Normalize the column vectors of the judgment matrix A to form a normalized matrix B = [b pq ], the calculation formula is: b pq =a pq / (Σ k＝1 n a kq ), where a kq represents the importance of indicator k relative to indicator q, n is the order of the judgment matrix; b pq Indicates a pq Normalized value;

[0110] The row vectors of the normalized matrix B are averaged to obtain the subjective weight vector W of each indicator = [w 1 ,w 2 ,…,w n ] T , the formula is: p =(Σ q＝1 n b pq ) / n, where w p is the subjective weight of indicator p, T represents the transpose operation of the matrix;

[0111] Perform consistency check on the judgment matrix A. The specific steps include: calculating the maximum eigenvalue λ of the judgment matrix A max , the formula is: max =Σ p＝1 n ((A.W) p / w p ) / n, where (A·W) p is the product of the judgment matrix and the weight vector, which represents the comprehensive evaluation; according to λ max Calculate the consistency index CI, the formula is: CI = (λ max -n) / (n-1); calculate the consistency ratio CR, the formula is: CR=CI / RI, where RI is the random consistency index, which is obtained by looking up the table.

[0112] Through the above steps, the hierarchical analysis method can effectively determine the relative importance of each indicator from the subjective judgment of experts, and ensure the scientificity and rationality of the judgment matrix through consistency test. In the process of constructing the weight vector, the Delphi method is used to collect expert opinions, and the matrix normalization, averaging and other methods are used to ensure the fairness and objectivity of the weights of each indicator. Finally, the rationality of the judgment matrix is ​​verified through the consistency test step, which provides a reliable basis for the subsequent weight calculation and ensures that the final comprehensive evaluation results have a high degree of credibility and accuracy.

[0113] Furthermore, based on the discrete degree of each indicator data, the entropy method is used to calculate the objective weight of each indicator, including the following steps:

[0114] The data of each indicator is standardized, the formula is: rp =(x rp -min(x p )) / (max(x p )-min(x p ), where x rp is the value of the rth sample at the pth index; min(x p ) and max(x p ) are the minimum and maximum values ​​of the pth index respectively; y rp Represents the standardized value of the rth sample on the pth index;

[0115] The proportion matrix is ​​calculated based on the standardized data, and the formula is: rp =y rp / Σ r＝1 m y rp , p rp It represents the ratio of the rth sample to the pth index, and m represents the number of samples;

[0116] Calculate the entropy value, the formula is: e p =(-1 / ln(m)) Σ r＝1 m (p rp ln(p rp )), e p represents the entropy value of the pth indicator; ln(m) is the normalization factor;

[0117] According to the entropy value e p , calculate the objective weight v of each indicator p , the formula is: p =(1-e p ) / Σ p＝1 n (1-e p );

[0118] Through the above steps, the entropy method can effectively and objectively calculate the weight of each indicator according to the degree of discreteness of each indicator. Standardization eliminates the influence of different dimensions, the ratio matrix transforms the relative contribution of each sample, and the entropy calculation reflects the amount of information of each indicator. Finally, the objective weight calculated according to the entropy value can provide a scientific basis for the subsequent comprehensive evaluation.

[0119] Further, step 4 includes the following steps:

[0120] Build a real-time data collection platform based on the Internet of Things technology to realize the automatic collection, transmission and storage of key indicators of nodes. The real-time data collection platform based on the Internet of Things (IoT) technology can automatically collect key operating indicators from each node, ensuring the efficiency and accuracy of data collection. Through sensors and embedded devices, the performance indicators, environmental parameters and other information of the nodes can be monitored in real time, and these data can be transmitted to the cloud platform or local server through wireless communication technology for storage and processing. During the transmission process, the application of low-power wide area network (such as LoRa, NB-IoT) technology can ensure the transmission of data over long distances and with high efficiency without being affected by the wide distribution range of nodes and the power consumption limit of equipment. The automation of data transmission avoids the instability caused by manual data collection, and through real-time data collection, it can respond to changes in the network or nodes in a timely manner, such as load fluctuations, equipment failures, etc., which helps to improve reliability and response speed. By establishing a complete data storage system, the platform provides strong data support for subsequent analysis and decision-making, which can effectively improve operational efficiency and stability.

[0121] Formulate a standardized node status monitoring process, regularly collect and update the operating data of each node, including performance indicators, load conditions and environmental parameters, and establish a dynamic monitoring database. A standardized node status monitoring process is the key to ensuring long-term stable operation. By setting a unified data collection standard and cycle, the uniformity and comparability of node data are guaranteed, and errors caused by different data collection methods between nodes are avoided. Regularly collecting node performance indicators (such as CPU load, memory usage, network bandwidth) and environmental parameters (such as temperature, humidity, vibration, etc.) helps to promptly discover potential abnormal conditions, such as performance degradation caused by equipment overload or environmental factors. Establish a dynamic monitoring database that can save and update the status data of each node in real time, making the data timely and traceable. The dynamic database is not only the basis for data storage, but also can work with analysis tools to ensure that data can be retrieved and analyzed at any time, providing a basis for subsequent optimization and decision-making. The standardized implementation of this process makes the monitoring of the entire network more systematic and standardized, which can effectively reduce deviations caused by human factors or technical problems, and further improve the stability of nodes and networks.

[0122] Based on the established resilience evaluation system, the combined weights are used to perform weighted calculations on various indicators to achieve quantitative calculation of the resilience index of a single node. The construction of the resilience evaluation system is a comprehensive analysis of the node health status and the system's anti-interference ability. By combining the subjective weight (based on expert evaluation) with the objective weight (based on data analysis) through the combined weight method, the weighted calculation of multiple performance indicators can be achieved, and then the resilience index of a single node can be obtained. This weighted calculation can comprehensively consider the influence of multiple dimensions, such as load, performance, environmental changes and other factors, to obtain more accurate evaluation results. The resilience index provides a quantitative evaluation for each node, which can accurately reflect the node's stability and ability to cope with abnormal situations. For example, a node with a high resilience index means that it can maintain good performance and is not prone to failure when the load increases or the environment changes. Through this quantitative calculation, the health status of the node can be fed back in real time, supporting dynamic adjustment and optimization, and improving the overall stability and self-recovery capability of the system.

[0123] Using complex network theory, we calculated the network topology characteristic parameters, evaluated the network connectivity, clustering coefficient and modularity, and analyzed the impact mechanism of network topology on overall resilience. The application of complex network theory provides a powerful mathematical framework for network resilience analysis. By calculating the topological characteristics of the network, such as the connectivity, clustering coefficient and modularity of the network, we can deeply analyze the impact mechanism of network structure on overall resilience. Connectivity measures the number of connections between nodes in the network. Networks with higher node connectivity usually have stronger anti-interference ability and can better cope with node failure. The clustering coefficient reflects the closeness between nodes in the network. Networks with high clustering coefficients usually have better local robustness and can reduce the impact on the entire network when local nodes fail. Modularity evaluates the community structure in the network. Networks with higher modularity can achieve functional separation. When one module fails, the functions of other modules will not be significantly affected. Through the calculation of these topological characteristics, the structural characteristics of the network can be quantified, providing theoretical support for optimizing network design and improving network resilience. The application of complex network theory makes the structural optimization and resilience assessment of the network more systematic and scientific.

[0124] A multi-level network resilience evaluation framework is constructed to organically integrate the node resilience index with the network topology characteristics to achieve quantitative characterization and dynamic evaluation of the overall resilience level of the network. The multi-level network resilience evaluation framework organically combines the resilience index of a single node with the topological characteristics of the network to form a comprehensive evaluation system for the resilience of the entire network. By combining the node resilience index with topological characteristics (such as connectivity, clustering coefficient, etc.), not only can the health status of each node be evaluated, but also the stability and anti-interference ability of the entire network under different conditions can be evaluated. This framework can dynamically adjust the resilience evaluation results of the network according to the real-time changes of the node status and network structure, forming a flexible and real-time responsive system. For example, when the number of nodes changes or the network structure is adjusted, the overall resilience index of the network can be updated in real time to provide a decision-making basis for network managers. Through the multi-level evaluation framework, the overall resilience level of the network is quantitatively characterized, and the performance of the network in the face of external disturbances can be monitored in real time, thereby providing a scientific basis for network optimization, resource scheduling and fault recovery.

[0125] Furthermore, the calculation formula of the single node resilience index is: R i =∑ p＝1 n V p ·G ip , where W p is the combined weight of index p, which is determined by the analytic hierarchy process and entropy method. The formula is: V p =(w p ·v p ) / ∑ k＝1n (w p ·v p );G ip is the normalized value of node i on index p, and the formula is: ip =(g ip -min(g p )) / (max(g p )-min(g p )), where min(g p )、max(g p ) represent the minimum and maximum values ​​of index p in the whole network respectively; g ip Represents the original value of node i on index p;

[0126] The overall resilience level of the network is expressed as: R net (t) = ∑ i＝1 K ω i ·R i (t)+∑ l＝1 L α l ·T l (t), where R net (t) is the overall network resilience index at time t, indicating the current network resilience level; R i (t) is the resilience index of node i at time t; T l (t) is the value of the topological feature l at time t, reflecting the impact of the change in topological structure on toughness; ω i Represents the weight coefficient of node i in the overall network resilience evaluation; α l Topological features l The weight represents the influence of network topology characteristics on network resilience; K is the total number of nodes in the network; L is the total number of network topology characteristics.

[0127] In summary, by calculating the single-node resilience index and the overall resilience level of the network, the resilience evaluation of the network at different time points and under different conditions can be realized. The single-node resilience index combines the analytic hierarchy process and the entropy method to reasonably calculate the weights of each indicator and quantify the stability and anti-interference ability of the node. The overall resilience level of the network comprehensively evaluates the anti-interference ability and stability of the network under various structures and states by integrating node resilience and topological characteristics. Such an evaluation system can not only provide accurate real-time feedback, but also provide data support for network optimization, fault warning and recovery, thereby ensuring that the network can operate stably in various complex environments and emergencies.

[0128] Further, step 5 includes the following steps:

[0129] Establish an anomaly detection mechanism based on statistical analysis, monitor the changing trend of node resilience indicators in real time, use the moving average method to identify abnormal fluctuations that deviate significantly from the normal range, and trigger corresponding warning signals. The moving average method can accurately capture changes in long-term trends by smoothing time series data and reducing the impact of noise. At the same time, it can quickly trigger warning signals when fluctuations that deviate significantly from the normal range occur. In this way, abnormal situations can be discovered in time to prevent instability or failure. The introduction of the moving average method makes anomaly detection more accurate and can effectively distinguish normal fluctuations from potential anomalies. This mechanism not only improves the sensitivity to the health status of nodes, but also provides timely warnings for operation and maintenance personnel, effectively reducing possible risks.

[0130] Combining the degree centrality, betweenness centrality and eigenvector centrality indicators of nodes, comprehensively considering the functional attributes and business value of nodes, constructing a node importance evaluation system, and thus identifying key nodes in the network. By comprehensively considering the degree centrality, betweenness centrality and eigenvector centrality of nodes, the key role of nodes in the network can be comprehensively evaluated. Degree centrality reflects the number of direct connections of a node, which can reveal the basic position of a node in the network; betweenness centrality measures the importance of a node in the process of network information transmission, and can identify key nodes that are indispensable in data flow; eigenvector centrality further reveals the relative importance of a node in the entire network structure. By combining these three centrality indicators, the functional attributes and business value of each node can be accurately evaluated, and key nodes in the network can be identified. These nodes usually have high redundancy requirements, and once a failure occurs, they may have a greater impact on the entire network, so they need to be given priority monitoring and maintenance. Through a comprehensive evaluation of the importance of nodes, the optimal configuration of network resources can be achieved, and the resilience and anti-interference ability of the network can be improved.

[0131] Based on historical data analysis and expert experience, the vulnerability of nodes is evaluated from the dimensions of node resilience index, failure frequency and repair time, while considering the correlation effect between nodes to identify potential weak links. The resilience index reflects the self-recovery ability of the node when encountering interference, the failure frequency reflects the possibility of node failure, and the repair time reveals the efficiency of node recovery. By combining these indicators, the vulnerability of the node can be comprehensively evaluated. At the same time, the correlation effect between nodes cannot be ignored. The interdependence between nodes may cause the failure of a node to trigger a series of chain reactions, further weakening the overall stability. Therefore, when evaluating node vulnerability, it is also necessary to consider the correlation effect between nodes to identify potential weak links. Through this comprehensive evaluation, network management can identify high-risk nodes, take timely repair measures, avoid potential failures, and thus improve the reliability of the entire network.

[0132] Based on the importance level of nodes and historical operation experience, a hierarchical warning threshold system is established, and it is dynamically adjusted according to the actual operation situation to ensure the timeliness and accuracy of the warning. The hierarchical warning threshold system is one of the core mechanisms to ensure the stable operation of the network. Based on the importance level of nodes and historical operation experience, the warning threshold system sets different monitoring standards and warning trigger conditions for nodes of different levels. Nodes with higher importance are often set with lower warning thresholds to ensure that they can be discovered and measures can be taken as soon as abnormalities occur. For general nodes, the warning threshold is set relatively high to ensure that unnecessary operation interference will not be caused by the overly sensitive warning mechanism. This hierarchical warning mechanism can refine network management and ensure that the network can respond accurately and promptly when facing various emergencies. By adjusting the warning threshold in real time and dynamically, it can ensure that the warning mechanism is always in the optimal state, neither missing potential risks nor avoiding the impact of excessive warnings on normal operations. The ability to adjust dynamically enhances adaptability, enabling it to always maintain efficient warning capabilities under different operating environments.

[0133] Based on the early warning threshold system, combined with the degree of node abnormality and the scope of influence, the identified risks are graded and classified for management. Risk grade division and classification management are effective means to achieve precise risk control. Based on the early warning threshold system, combined with the degree of node abnormality and the scope of influence, the identified risks can be graded and classified for management. According to the node failure type, scope of influence and repair difficulty, risks can be divided into multiple levels, from high risk to low risk. High-risk nodes require immediate emergency repair measures, and may also need to adjust the network structure to avoid catastrophic consequences; medium-risk nodes can reduce risks by optimizing configuration or adjusting operation strategies; low-risk nodes can be handled within the regular operation and maintenance cycle. Classification management can ensure the rational allocation and utilization of resources, while ensuring the overall network stability, avoiding the inefficiency caused by excessive concentration of resources on high-risk nodes. In addition, classification management can also help operation and maintenance personnel prioritize the most important risks, reduce resource waste, and improve response efficiency.

[0134] In summary, through the implementation of the above steps, network resilience management can become more accurate and efficient. The anomaly detection mechanism ensures real-time monitoring of node status and timely discovery of problems; the node importance evaluation system can help managers identify key nodes and give priority attention; the node vulnerability assessment can deeply analyze the potential risks of each node and formulate response plans in advance; the hierarchical warning threshold system and risk level division ensure the reasonable allocation and efficient management of network resources, thereby enhancing the network's resilience and anti-interference capabilities in the face of emergencies. Combining these technical means, the accuracy and dynamism of network management have been greatly improved, which can effectively improve stability and security.

[0135] Furthermore, based on historical data analysis and expert experience, the vulnerability of nodes is evaluated from the dimensions of node resilience index, failure frequency, and repair time, while considering the correlation effect between nodes to identify potential weak links, including the following steps:

[0136] Collect historical performance data of nodes, including failure frequency, repair time, and past vulnerability events, and integrate the experience of domain experts to obtain qualitative assessment information of node vulnerability, so as to build a comprehensive node performance and risk data set. By collecting a large amount of historical data, we can fully understand the operating performance of each node under different conditions and the repair efficiency when a failure occurs. For example, the failure frequency can reveal the probability of a node problem, and the repair time reflects the time required for the node to resume normal operation after a failure, which is crucial for evaluating the long-term stability of the node. At the same time, combined with the experience of domain experts, it can provide qualitative support for vulnerability assessment, especially when the amount of data is insufficient or there is a lack of direct historical cases. The experience of experts can supplement the deficiencies in the data and help identify potential problems that may be overlooked in the traditional data collection process. For example, some nodes may exhibit abnormal behavior under specific operating or environmental conditions, and this information can be supplemented by the experience of experts. Finally, integrating historical data and expert experience can build a comprehensive node performance and risk data set, providing sufficient basis for subsequent evaluation work.

[0137] Based on historical data, the resilience index, failure frequency and average repair time of each node are calculated, and node recovery capability evaluation indicators are constructed to evaluate the node's rapid response and recovery potential in emergencies. The resilience index reflects the node's anti-interference ability when facing external interference, and it can quantify the stability of the node. Through historical data analysis, the resilience performance of the node under various stress conditions can be calculated, and the statistical information of the failure frequency and repair time can be further calculated. The failure frequency refers to the number of times a node fails in a specific time period, which can reflect the reliability of the node itself and the stability of long-term operation; while the repair time reveals the time from failure to recovery of the node, reflecting the node's recovery potential. Combining these indicators, a node recovery capability evaluation indicator can be constructed, which is crucial for evaluating the node's rapid response and recovery capabilities after a failure.

[0138] Analyze the complex dependencies between nodes, quantify the correlation effects of interdependence, and evaluate their potential impact on the overall network vulnerability. Nodes in a network often do not exist in isolation, and there are multiple dependencies between them. For example, the failure of a node may affect other nodes connected to it, thereby affecting the performance of the entire network. This correlation effect can amplify the vulnerability of the system in the short term, so the relationship between nodes must be analyzed and quantified. By introducing the topological analysis method in complex network theory, the dependencies between nodes can be effectively analyzed. For example, indicators such as node connectivity and clustering coefficient can be used to quantify the dependencies between nodes. For nodes with strong dependencies, once a failure occurs, it may trigger a chain reaction and affect the overall stability of the network. Therefore, quantifying these correlation effects and evaluating their potential impact on the overall network vulnerability can help identify weak links in the network and take measures to repair them in advance to prevent the spread of network failures.

[0139] Vulnerability assessment is performed by integrating the resilience index, failure frequency, repair time and correlation effect between nodes. By integrating these evaluation indicators, a comprehensive vulnerability mechanism can be established to systematically evaluate the vulnerability of each node. This model can assign a vulnerability score to each node based on different evaluation criteria. Nodes with higher scores may pose a greater risk to the entire network. Therefore, vulnerability assessment is not just a performance assessment of a single node, but also requires comprehensive consideration of the mutual influence between nodes. Through this comprehensive evaluation method, network managers can more clearly understand the vulnerable position of each node in the network, and carry out targeted maintenance and optimization, thereby avoiding the collapse of the entire network due to individual node failures.

[0140] Based on the results of vulnerability assessment, identify and locate key weak links. By sorting the vulnerability score of each node, high-risk nodes in the network can be quickly identified. These nodes are often weak links in the system and are prone to failure under excessive load or external interference. Once these nodes fail, it may cause serious chain reactions and affect the normal operation of the entire network. Therefore, identifying these key weak links is a core task to improve network resilience. By combining the results of vulnerability assessment with the importance evaluation of nodes, it is possible to further determine which nodes need to be repaired or replaced first. These high-risk nodes should receive special attention, including strengthening monitoring, optimizing configuration, adding redundancy or adopting other reliability enhancement measures to improve the overall anti-interference ability and stability of the network.

[0141] In summary, by collecting information from historical data and expert experience, combined with the node's resilience index, failure frequency, repair time, and the correlation effect between nodes, the vulnerability of each node in the network can be comprehensively evaluated. This series of evaluation methods can help network managers accurately identify weak links and take appropriate repair or improvement measures to improve the resilience and stability of the network. Especially in a complex network environment, the dependencies and mutual influences between nodes are more significant. Therefore, a comprehensive vulnerability assessment can not only help discover problems with individual nodes, but also reveal potential system risks, providing important support for optimizing network structure and improving overall network performance.

[0142] Further, step 6 includes the following steps:

[0143] Graph neural networks are used to capture the dynamic associations between nodes, build a deep learning framework that combines node attributes and topological structures, and quantify the transmission paths and diffusion effects of risk shocks. Graph neural networks (GNNs) are deep learning algorithms that can effectively process graph structure data. By capturing the dynamic associations between nodes, complex network topological structures and node attributes can be deeply analyzed. In this step, GNNs can dynamically model and predict the mutual influence between nodes in the network by combining node attributes and topological structures. Traditional graph analysis methods often only focus on static node connections, while GNNs can process time series data and update the associations between nodes and changes in network structure in real time. Therefore, GNNs are particularly suitable for analyzing how risk shocks in the network spread through the transmission paths between nodes, especially in complex networks where the interactions and influences between nodes are dynamically changing. Through graph convolution operations, GNNs can extract high-order adjacency relationships between nodes and quantify the diffusion effects of risk shocks, thereby providing more accurate predictions for resilience assessment.

[0144] Based on the goal of improving resilience, a multi-objective optimization model is established by comprehensively considering economic costs, social benefits and environmental impacts, and a series of candidate solutions with different optimization focuses are generated using the particle swarm algorithm. The multi-objective optimization model comprehensively considers multi-dimensional factors such as economic costs, social benefits and environmental impacts, and uses the particle swarm algorithm (PSO) to generate candidate solutions with different optimization focuses. The particle swarm algorithm is an intelligent optimization algorithm that simulates the foraging behavior of bird flocks. It searches and iterates multiple particles to find the optimal solution. This method performs well in multi-objective optimization problems and can find a compromise solution between multiple objectives to balance the needs of all parties. For example, improving the resilience of the network may require increased infrastructure investment, but this may bring higher economic costs; at the same time, it may also affect social benefits and environmental impacts. Through the particle swarm algorithm, the optimal solution can be found between these conflicting objectives, and then multiple candidate optimization solutions can be generated to evaluate and select the most suitable implementation path from multiple perspectives.

[0145] In response to the vulnerabilities at different levels, specific improvement measures covering infrastructure upgrades, land use structure adjustments, urban functional layout optimization, and resource allocation efficiency improvement are formulated to form a systematic optimization solution library. The key to this step is how to design feasible specific solutions based on the network vulnerability assessment results and optimization goals. For example, in terms of infrastructure upgrades, it is necessary to improve the reliability of the system by strengthening the redundant configuration of nodes, increasing backup resources, or increasing the maintenance frequency of nodes; in terms of land use structure adjustments, it is necessary to optimize the layout of the city according to the network topology and reduce the concentration of high-risk nodes; in terms of urban functional layout optimization, the chain reaction of network failures can be mitigated by reasonably allocating resources within the city and reducing the dependence between nodes. By comprehensively considering the vulnerabilities at multiple levels, the improvement measures formulated can form a complete optimization solution library, providing a variety of options for subsequent solution selection and implementation.

[0146] The Monte Carlo simulation method is used to evaluate the implementation effects and risk response capabilities of different optimization schemes under various uncertain conditions through simulation experiments of a large number of random scenarios, and a scheme evaluation index system is constructed. This step generates a large amount of random input data to simulate the performance in a variety of possible scenarios, thereby providing a series of system responses in different scenarios. This process can help evaluate the robustness of the scheme in the face of uncertainty. For example, when considering uncertainties such as environmental factors, economic fluctuations, or changes in social needs, Monte Carlo simulation can predict the execution results and potential risks of each optimization scheme under these uncertain factors. The introduction of this technology makes the evaluation process closer to the actual situation, and can analyze the adaptability of the scheme from multiple dimensions, so as to identify the optimization scheme with the greatest risk response capabilities. In addition, the simulation results can provide valuable data support for scheme optimization and help decision makers adjust the focus and implementation path of the scheme.

[0147] Taking into account the feasibility, economy and implementation difficulty of the scheme, combined with the simulation evaluation results, the candidate schemes are comprehensively ranked using the hierarchical analysis method, the optimal scheme is selected and a phased implementation roadmap is formulated. The hierarchical analysis method is a quantitative analysis method for multi-criteria decision-making. By constructing a hierarchical structure of decision-making problems, combined with expert evaluation and quantitative analysis, the weights and rankings of each candidate scheme are finally obtained. Through this method, the advantages and disadvantages of different optimization schemes can be clarified, and a scientific and reasonable ranking can be provided for decision makers. By evaluating the feasibility, economy and implementation difficulty of the scheme, the optimization scheme with the most implementation value can be further screened out, reducing the interference of subjective factors in the scheme selection. At the same time, formulating a phased implementation roadmap can ensure the gradual implementation and effective promotion of the optimization scheme, ensure that the goals of each stage can be fully achieved, and ultimately achieve a comprehensive improvement in network resilience.

[0148] In summary, by capturing the dynamic correlation between nodes through graph neural networks, establishing a solution generation mechanism based on multi-objective optimization, and combining Monte Carlo simulation for effect evaluation, it can provide comprehensive theoretical and technical support for improving network resilience. Not only do they play an important role at their independent level, but they also form a close synergy between them, thus providing decision makers with a multi-dimensional and accurate basis for decision-making. Finally, the formulation of comprehensive sorting and implementation roadmaps enables the plan to be carried out in an orderly manner and achieve the best results in a complex practical environment. Through the precise design of these steps, it can ensure that the network can maintain an efficient and stable operating state in the face of complex and changing challenges, thereby improving the overall social response capacity and sustainable development level.

[0149] Further, step 7 includes the following steps:

[0150] Establish a multi-level evaluation framework covering the evaluation index system, network topology and node resilience improvement effect, and design standardized evaluation processes and methods to achieve a comprehensive quantitative evaluation of the node and network improvement effects. The design of the hierarchical evaluation framework will ensure a comprehensive quantitative evaluation of the node and network improvement effects, thereby providing a systematic perspective to help evaluate the implementation effects of various improvement measures. First, the establishment of the evaluation index system needs to cover multiple dimensions such as node resilience index, failure frequency, repair time, and risk prevention and control capabilities to ensure that the performance of nodes in different scenarios can be comprehensively measured. Secondly, the change of network topology is also an important dimension of evaluation. By combining the topological structure of the network, the correlation and vulnerability between nodes can be analyzed, revealing potential bottlenecks and weak links in the network. The implementation of the multi-level evaluation framework can not only help monitor node performance, but also track whether the resilience and stability of the entire network have been improved after the improvement. The standardized process and methods of this framework make the evaluation work operational and consistent, and can provide a quantitative basis for the effects of network improvement measures, thereby providing scientific guidance for subsequent optimization.

[0151] Regularly collect and analyze node operation data and network status data, combine comparative analysis before and after implementation and trend research, and quantitatively evaluate the actual effects of various improvement measures, including the degree of improvement of node resilience index, the enhancement of network risk prevention and control capabilities, and the improvement level of overall network performance. By comparing the data before and after implementation, the actual effects of various improvement measures can be accurately quantified. Specifically, the degree of improvement of node resilience index is an important indicator for measuring node stability and reliability. By comparing the data before and after implementation, it can be verified whether the node's recovery ability in the face of different risk shocks has been effectively enhanced. At the same time, the enhancement of network risk prevention and control capabilities can also be obtained through this comparison. If the network can effectively prevent the occurrence of risk events and reduce the chain reaction of failures, the overall resilience of the network will be improved. In addition to data analysis at the node level, the improvement of overall network performance is also a key link in the evaluation. By comparing and analyzing the node resilience index, fault repair time, and network status changes, the comprehensive performance of the improved network can be intuitively seen. The results of regular analysis provide real-time data support for subsequent evaluation and adjustment, ensuring that the implementation effects of various measures can be fed back and optimized in a timely manner.

[0152] Based on the effect evaluation results, the node warning threshold is adaptively adjusted, and the weight of the node evaluation index is recalculated to ensure that the evaluation system can timely reflect the dynamic changes of node status and network structure. Adaptive adjustment not only improves the flexibility and accuracy of the evaluation, but also automatically adjusts the sensitivity when the node status fluctuates greatly to avoid overreaction or ignoring potential problems. In addition, the weight of the node evaluation index plays a key role in the dynamic adjustment process. According to the importance of different nodes and the actual operating status in the network, the weight of each node in the evaluation is adjusted to ensure that the evaluation results of the network are more objective and accurate. By continuously adjusting the evaluation system, it is possible to better cope with changes in node performance and fluctuations in network status in actual operations, thereby maintaining the high resilience and availability of the network in a complex environment.

[0153] According to the results of the effect evaluation, the priorities of existing node improvement measures and network optimization solutions are dynamically adjusted, resource allocation is optimized, and the improvement measures library is updated and improved in a timely manner in combination with newly discovered node problems and network requirements. This dynamic adjustment mechanism ensures that the network can fully consider new node problems and changes in network requirements during the continuous optimization process. With the continuous implementation of node resilience improvement measures, some nodes may show better stability and lower failure frequency, while some nodes may have new vulnerabilities or performance bottlenecks. Therefore, dynamically adjusting priorities and optimizing resource allocation have become necessary steps to improve the overall network resilience. By regularly evaluating the performance and requirements of nodes and adjusting the allocation of resources, it can be ensured that the improvement measures can specifically respond to new problems in the network while avoiding waste of resources. In addition, the timely discovery of new node problems and network requirements helps to quickly respond to potential risks and further improve the emergency response capabilities and stability of the network. The process of dynamically adjusting priorities ensures the flexibility and sustainability of network optimization, helps the optimization plan keep pace with the times, and meets the actual needs of different development stages.

[0154] A closed-loop optimization mechanism of evaluation-improvement-re-evaluation is established. Through continuous monitoring, evaluation and dynamic adjustment, the node resilience level and network risk response capability are continuously improved, and the node and network performance are continuously improved. This closed-loop optimization mechanism forms a circular feedback process through continuous monitoring, evaluation and dynamic adjustment to ensure that the network's resilience level and risk response capability are continuously improved. First, the evaluation link provides the necessary data support for the entire process to ensure that the weak links in the network and the effects of improvement measures are discovered in a timely manner. Secondly, the improvement link makes corresponding optimization adjustments based on the evaluation results to ensure that the actual effects of various measures can be reflected in the network. Finally, the re-evaluation link provides feedback for further improvement to help discover the shortcomings and potential risks of the solution. Through this closed-loop mechanism, the network can achieve continuous improvement in performance during the continuous optimization process. This continuous improvement and feedback mechanism not only improves the network resilience, but also enhances its ability to respond to future unknown risks. Through long-term closed-loop optimization, the long-term stability and risk resistance of the network can be maximized, thereby ensuring the sustainable development of the network in a complex environment.

[0155] In summary, the design of step 7 provides a driving force for continuous improvement in network resilience by establishing a comprehensive and flexible evaluation system, combined with multiple measures such as regular data analysis, dynamic adjustment, and closed-loop feedback. This mechanism not only ensures that the network has the ability to respond and recover quickly when facing various risks, but also can continuously adapt to new changes in long-term operation, optimize resource allocation, and improve overall system performance. Through continuous monitoring and evaluation, the network can promptly identify problems and make targeted improvements, ensuring that it always maintains an efficient and stable operating state in a complex and changing environment, and enhancing the network's self-healing ability and risk management level.

[0156] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying and evaluating the resilience of national land space planning, characterized in that: The following steps are involved: Step 1: Obtain spatial information, socio-economic data and ecological and environmental elements of the evaluation area; Step 2: Based on administrative boundaries, natural geographical features, and land use types, functional units are divided as network nodes, and a weighted adjacency matrix is ​​constructed to quantify the intensity of material flow, energy flow, and information flow between nodes to form a complete spatial relationship network structure; Step 3: Establish a resilience evaluation system for nodes and networks; Step 4: Calculate the resilience index of each node and comprehensively evaluate the overall resilience level of the network; Step 5: Identify key weak points and links, and determine the risk level based on the preset graded warning thresholds; Step 6: Use a multi-objective optimization algorithm to generate candidate solutions for specific improvement measures, and use the Monte Carlo simulation method to evaluate the feasibility and effectiveness of each candidate solution to determine the optimal implementation path; Step 7: Form a cyclic optimization mechanism of evaluation-improvement-re-evaluation.

2. According to claim 1, a method for identifying and evaluating the resilience of national land space planning is characterized in that: Step 1 includes the following steps: Acquire the latest high-resolution remote sensing image data, including satellite and aerial images, and perform geometric correction and image enhancement processing; Extract and classify land use types, vegetation coverage and water distribution elements from remote sensing images; Collect socio-economic statistics of the assessment area and ensure the timeliness and authority of the data; Conduct field research through GPS positioning, on-site photography and questionnaire surveys to verify the accuracy of remote sensing interpretation results and supplement ground information that is difficult to obtain through remote sensing; Integrate remote sensing interpretation results, socio-economic statistics and field survey data to build a geographic information database of the evaluation area to achieve unified management and analysis of multi-source data.

3. According to claim 1, a method for identifying and evaluating the resilience of national land space planning is characterized in that: Step 2 includes the following steps: The evaluation area is divided into several functional units based on the integrity of administrative boundaries, continuity of physical geographical elements, and homogeneity of land use types; Marking attribute information for each functional unit; By analyzing the material flow paths, energy exchange channels and information dissemination networks within the region, the association relationships between functional units can be identified; Based on the intensity of material flow, energy flow and information flow, the association weights between adjacent units are calculated and a weighted adjacency matrix is ​​constructed; Combine all functional units and weighted adjacency matrices to build a complete spatial relationship network structure; The connection strength between nodes and regional flow patterns are verified to ensure that the network structure can accurately reflect the actual functional relationships and flow characteristics of the region.

4. According to claim 1, a method for identifying and evaluating the resilience of national land space planning is characterized in that: Step 3 includes the following steps: Based on the basic framework of carrying capacity theory, starting from the three dimensions of social development, economic growth and ecological environment, we select key evaluation indicators that can objectively reflect the characteristics of node and network resilience; Normalization method was used to convert all indicators into the same dimensional range; Construct a judgment matrix and determine the subjective weight of each indicator through the analytic hierarchy process, and perform consistency test at the same time; Based on the discrete degree of each indicator data, the objective weight of each indicator is calculated using the entropy method; Combining subjective weights and objective weights, the geometric mean method is used to calculate the combined weight of each indicator; Based on key evaluation indicators and their combined weights, a comprehensive evaluation framework for node and network resilience is constructed.

5. According to claim 1, a method for identifying and evaluating the resilience of national land space planning is characterized in that: Step 4 includes the following steps: Realize the automatic collection, transmission and storage of key node indicators; Regularly collect and update the operation data of each node and establish a dynamic monitoring database; Perform weighted calculation on each indicator to achieve quantitative calculation of single node resilience index; Calculate network topology characteristic parameters, evaluate network connectivity, clustering coefficient and modularity, and analyze the impact of network topology on overall resilience; Realize quantitative characterization and dynamic evaluation of the overall resilience level of the network.

6. According to claim 1, a method for identifying and evaluating the resilience of national land space planning is characterized in that: Step 5 includes the following steps: Monitor the changing trend of node resilience indicators in real time, use the moving average method to identify abnormal fluctuations that significantly deviate from the normal range, and trigger corresponding early warning signals; Build a node importance evaluation system to identify key nodes in the network; Evaluate the vulnerability of nodes from the dimensions of node resilience index, failure frequency, and repair time, while considering the correlation effect between nodes to identify potential weak links; Establish a hierarchical and graded warning threshold system and make dynamic adjustments based on actual operating conditions; Based on the early warning threshold system, combined with the degree of node abnormality and the scope of impact, the identified risks are graded and classified.

7. The method for identifying and evaluating the resilience of national land space planning according to claim 1 is characterized in that: Step 6 includes the following steps: Utilize graph neural networks to capture dynamic relationships between nodes, build a deep learning framework that combines node attributes and topological structures, and quantitatively analyze the transmission path and diffusion effect of risk shocks; Based on the goal of improving resilience, a multi-objective optimization model is established by comprehensively considering economic costs, social benefits and environmental impacts, and a series of candidate solutions with different optimization focuses are generated using the particle swarm algorithm; Develop specific improvement measures covering infrastructure upgrades, land use structure adjustments, urban functional layout optimization, and resource allocation efficiency improvements to target different levels of vulnerability, and form a systematic optimization solution library; Using the Monte Carlo simulation method, through a large number of random scenario simulation experiments, we evaluate the implementation effects and risk response capabilities of different optimization schemes under various uncertain conditions, and build a scheme evaluation index system; Taking into account the feasibility, economy and implementation difficulty of the plan, combined with the simulation evaluation results, the hierarchical analysis method is used to comprehensively rank the candidate plans, select the optimal plan and formulate a phased implementation roadmap.

8. The method for identifying and evaluating the resilience of national land space planning according to claim 1 is characterized in that: Step 7 includes the following steps: Establish a multi-level evaluation framework covering the evaluation index system, network topology structure and node resilience improvement effect, and design standardized evaluation processes and methods to achieve a comprehensive quantitative evaluation of node and network improvement effects; Regularly collect and analyze node operation data and network status data, and combine comparative analysis and trend research before and after implementation to quantitatively evaluate the actual effects of various improvement measures, including the degree of improvement in node resilience index, the enhancement of network risk prevention and control capabilities, and the improvement level of overall network performance; Based on the effect evaluation results, the node warning threshold is adaptively adjusted and the weight of the node evaluation index is recalculated to ensure that the evaluation system can timely reflect the dynamic changes of node status and network structure; According to the effect evaluation results, dynamically adjust the priority of existing node improvement measures and network optimization solutions, optimize resource allocation, and timely update and improve the improvement measures library in combination with newly discovered node problems and network requirements; Build a closed-loop optimization mechanism of assessment-improvement-reassessment, and through continuous monitoring, assessment and dynamic adjustment, continuously improve the node resilience level and network risk response capabilities, and achieve continuous improvement of node and network performance.

Images