Low-altitude flight management system comprehensive capability dynamic evaluation and bottleneck identification method and system

Abstract

The invention discloses a low-altitude flight management system comprehensive capability dynamic evaluation and bottleneck identification method and system, and relates to the technical field of low-altitude flight and management systems, and the method comprises the steps: building a hierarchical index system, and obtaining the operation original data of a low-altitude flight management system, the hierarchical index system comprises a first hierarchy, a second hierarchy and a third hierarchy; performing index calculation based on the original data and the hierarchical index system to obtain index data; performing network construction and quantization processing on the hierarchical index system based on the index data to obtain a capability coupling network; based on the index data and the capability coupling network, carrying out comprehensive capability dynamic evaluation on the low-altitude flight management system to obtain a dynamic evaluation result; and based on the dynamic evaluation result and the capability coupling network, bottleneck identification processing is carried out on the low-altitude flight management system to obtain a bottleneck identification result, and the bottleneck identification result represents a node playing a role in restricting the capability of the low-altitude flight management system.

Description

Technical Field

[0001] This application relates to the technical fields of low-altitude flight and air traffic control systems, and in particular to a method and system for dynamic evaluation of the comprehensive capabilities and bottleneck identification of low-altitude flight air traffic control systems. Background Technology

[0002] The low-altitude flight management and service system is an important support platform for ensuring the safe and orderly operation of low-altitude airspace, and its comprehensive capability evaluation is an important basis for system construction, operation and maintenance and upgrading.

[0003] The relevant technologies employ a comprehensive evaluation method based on a multi-layered indicator system. This method collects system operation data, quantifies and scores each indicator, and calculates the overall system capability score using a weighted summation approach. This type of method typically includes an indicator library, a data acquisition module, and a weighted calculation module, and its evaluation process has a linear structure. However, this type of evaluation method usually assumes that the evaluation indicators are independent of each other, reflecting the differences in indicator importance only through static weights. It fails to characterize the prevalent correlations and linkages among various capability elements in the low-altitude flight management system, resulting in evaluation results that do not accurately reflect the system's true operational status. Summary of the Invention

[0004] The embodiments of this application aim to at least partially address one of the technical problems in the related art. To this end, the embodiments of this application propose a method, system, equipment, and medium for dynamic evaluation and bottleneck identification of the comprehensive capabilities of a low-altitude flight control system, thereby improving the accuracy of the evaluation results.

[0005] This application provides a method for dynamic evaluation and bottleneck identification of the comprehensive capabilities of a low-altitude flight control system, comprising: establishing a hierarchical indicator system and acquiring raw data of the operation of the low-altitude flight control system, wherein the hierarchical indicator system includes a first level, a second level, and a third level; calculating indicators based on the raw data and the hierarchical indicator system to obtain indicator data; constructing and quantifying the hierarchical indicator system based on the indicator data to obtain a capability coupling network; performing a dynamic comprehensive capability evaluation of the low-altitude flight control system based on the indicator data and the capability coupling network to obtain a dynamic evaluation result; and performing bottleneck identification processing on the low-altitude flight control system based on the dynamic evaluation result and the capability coupling network to obtain a bottleneck identification result, wherein the bottleneck identification result represents the nodes that restrict the capabilities of the low-altitude flight control system.

[0006] In some implementations, the second level includes a first capability level and a second capability level, and the third level includes a first indicator level. Based on indicator data, a network is constructed and quantified for the hierarchical indicator system to obtain a capability-coupled network. This includes: performing node-based processing on the first level, the first capability level, the second capability level, and the first indicator level to obtain capability nodes and indicator nodes; calculating contribution weights for the capability nodes corresponding to the first level, the capability nodes corresponding to the first capability level, the capability nodes corresponding to the second capability level, and the indicator nodes corresponding to the first indicator level based on indicator data to obtain contribution weights. The contribution weights represent the weights corresponding to the contribution edges, which include the edges between capability nodes corresponding to the first level and capability nodes corresponding to the first capability level, the edges between capability nodes corresponding to the first capability level and capability nodes corresponding to the second capability level, and the edges between capability nodes corresponding to the second capability level and capability nodes corresponding to the second capability level. The edges between nodes and the corresponding indicator nodes of the first indicator level are calculated. Based on indicator data and contribution weights, the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are weighted to obtain score data. Based on the score data, the deviation coefficients of the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are calculated to obtain coupling weights. The coupling weights represent the weights corresponding to the coupling edges, which include the edges between capability nodes corresponding to the first capability level and the edges between capability nodes corresponding to the second capability level. Based on the indicator data, the correlation coefficients of the indicator nodes corresponding to the first indicator level are calculated to obtain association weights. The association weights represent the weights corresponding to the association edges, which include the edges between indicator nodes corresponding to the first indicator level. Based on the capability nodes, indicator nodes, contribution weights, coupling weights, and association weights, the capability coupling network is obtained.

[0007] In some implementations, contribution weights are calculated for capability nodes corresponding to the first level, capability nodes corresponding to the first capability level, capability nodes corresponding to the second capability level, and indicator nodes corresponding to the first indicator level based on indicator data. This includes: calculating contribution weights for capability nodes corresponding to the first level and capability nodes corresponding to the first capability level based on score data to obtain a first contribution weight; calculating contribution weights for capability nodes corresponding to the first capability level and capability nodes corresponding to the second capability level based on score data to obtain a second contribution weight; and calculating contribution weights for capability nodes corresponding to the second capability level and indicator nodes corresponding to the first indicator level based on indicator data to obtain a third contribution weight.

[0008] In some implementations, the scoring data includes first scoring data corresponding to a first capability level and second scoring data corresponding to a second capability level; the deviation coefficients of the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are calculated based on the scoring data to obtain coupling weights, including: calculating the deviation coefficients of the capability nodes corresponding to the first capability level based on the first scoring data to obtain a first coupling weight; and calculating the deviation coefficients of the capability nodes corresponding to the second capability level based on the second scoring data to obtain a second coupling weight.

[0009] In some implementations, a comprehensive dynamic evaluation of the low-altitude flight management system is performed based on indicator data and a capability coupling network to obtain dynamic evaluation results. This includes: weighting the first score data corresponding to the first capability level based on a first contribution weight to obtain third score data; iteratively updating the capability nodes and indicator nodes based on indicator data, first score data, second score data, and third score data to obtain target score data and total score data corresponding to the capability nodes and indicator nodes. The iterative update calculation includes updating the score data of each node in the capability coupling network based on the score data and weights of neighboring nodes. Neighboring nodes include capability nodes and / or indicator nodes, and weights include at least one of contribution weight, coupling weight, and association weight; and obtaining dynamic evaluation results based on the target score data and total score data.

[0010] In some implementations, the capability nodes and indicator nodes are iteratively updated based on indicator data, first score data, second score data, and third score data to obtain target score data and total score data corresponding to the capability nodes and indicator nodes. This includes: iteratively updating the capability nodes and indicator nodes based on indicator data, first score data, second score data, and third score data to obtain target data after each iteration; calculating the difference between indicator data, first score data, second score data, third score data, and target data to obtain change data; and when the change data is less than a preset threshold, using the target data as target score data, and obtaining total score data based on the target score data.

[0011] In some implementations, bottleneck identification processing is performed on the low-altitude flight management system based on dynamic evaluation results and capability coupling network to obtain bottleneck identification results. This includes: determining the adjacency matrix based on capability nodes, indicator nodes, contribution weights, coupling weights, and association weights; performing feature calculation based on the adjacency matrix to obtain feature vectors; and performing bottleneck identification processing on the low-altitude flight management system based on the feature vectors and dynamic evaluation results to obtain bottleneck identification results.

[0012] In some implementations, bottleneck identification processing is performed on the low-altitude flight control system based on feature vectors and dynamic evaluation results to obtain bottleneck identification results. This includes: normalizing the feature vectors to obtain normalized feature vectors; performing bottleneck identification processing on the low-altitude flight control system based on the normalized feature vectors and target score data to obtain a bottleneck factor, wherein the bottleneck factor is obtained by: performing a difference operation based on preset values ​​and target score data to obtain a difference result; multiplying the difference result with the normalized feature vector to obtain the bottleneck factor; and obtaining the bottleneck identification result based on the bottleneck factor and preset rules.

[0013] In some implementations, the method for dynamic evaluation and bottleneck identification of the comprehensive capabilities of the low-altitude flight management and service system also includes: performing path tracing processing based on the bottleneck identification results to obtain the impact path of the low-altitude flight management and service system.

[0014] This application provides a dynamic evaluation and bottleneck identification system for the comprehensive capabilities of a low-altitude flight control system. The system includes: an acquisition module for establishing a hierarchical indicator system and acquiring raw data of the low-altitude flight control system's operation, wherein the hierarchical indicator system includes a first level, a second level, and a third level; a calculation module for calculating indicators based on the raw data and the hierarchical indicator system to obtain indicator data; a processing module for constructing and quantifying the hierarchical indicator system based on the indicator data to obtain a capability coupling network; an evaluation module for dynamically evaluating the comprehensive capabilities of the low-altitude flight control system based on the indicator data and the capability coupling network to obtain a dynamic evaluation result; and an identification module for identifying bottlenecks in the low-altitude flight control system based on the dynamic evaluation result and the capability coupling network to obtain a bottleneck identification result, wherein the bottleneck identification result represents the nodes that restrict the capabilities of the low-altitude flight control system.

[0015] An embodiment of this application provides an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by one or more processors, which are executed by one or more processors to cause the one or more processors to implement the steps of the method of any of the above embodiments.

[0016] Embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0017] The above embodiments of the method for dynamic evaluation and bottleneck identification of the comprehensive capabilities of a low-altitude flight control system include: establishing a hierarchical indicator system and acquiring raw data of the operation of the low-altitude flight control system, wherein the hierarchical indicator system includes a first level, a second level, and a third level; calculating indicators based on the raw data and the hierarchical indicator system to obtain indicator data; constructing and quantifying the hierarchical indicator system based on the indicator data to obtain a capability coupling network; performing a comprehensive dynamic evaluation of the low-altitude flight control system based on the indicator data and the capability coupling network to obtain a dynamic evaluation result; and performing bottleneck identification processing on the low-altitude flight control system based on the dynamic evaluation result and the capability coupling network to obtain a bottleneck identification result, wherein the bottleneck identification result represents the nodes that restrict the capabilities of the low-altitude flight control system. By establishing a hierarchical indicator system for the low-altitude flight control system and completing indicator calculation, network construction, and quantification based on raw data, a capability coupling network reflecting the complex relationships among various elements within the system is constructed. Through the collaborative analysis of this network model and dynamic evaluation results, real-time dynamic evaluation of the system's comprehensive capabilities is achieved. Based on this, key bottleneck nodes that restrict the overall performance of the system are accurately identified, improving the accuracy and timeliness of bottleneck identification and providing quantitative decision-making basis for system capability improvement and resource optimization. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a method for dynamic evaluation and bottleneck identification of the comprehensive capabilities of a low-altitude flight control system, provided for the implementation of this application.

[0019] Figure 2 A schematic diagram of the workflow of a dynamic evaluation and bottleneck identification system for the comprehensive capabilities of a low-altitude flight control system provided in another embodiment of this application; Figure 3 A schematic diagram of the hierarchical index system structure provided for the implementation of this application; Figure 4 This is a schematic diagram of path tracing provided for an embodiment of this application; Figure 5 A block diagram of an electronic device provided for another embodiment of this application. Detailed Implementation

[0020] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0021] The low-altitude flight management and service system is an important support platform for ensuring the safe and orderly operation of low-altitude airspace, and its comprehensive capability evaluation is an important basis for system construction, operation and maintenance and upgrading.

[0022] A comprehensive evaluation method based on a multi-layered indicator system is typically employed. This method involves collecting system operation data, quantifying and scoring each indicator, and then calculating the overall system capability score using a weighted summation approach. This type of method generally includes an indicator library, a data acquisition module, and a weighted calculation module, and its evaluation process has a linear structure. For example, relevant solutions often construct an indicator system encompassing dimensions such as operational support, command and dispatch, and monitoring and perception. Each dimension has several specific indicators, and weights are determined through expert scoring or the entropy weight method, ultimately resulting in a weighted comprehensive score.

[0023] However, such evaluation methods usually assume that the evaluation indicators are independent of each other and only reflect the differences in the importance of the indicators through static weights. They cannot depict the common correlation and linkage between various capability elements in the low-altitude flight management and service system (such as insufficient monitoring and perception capabilities directly affecting command and dispatch efficiency). As a result, the evaluation results are difficult to reflect the actual operating status of the system and cannot identify the hidden bottlenecks caused by the coupling relationship between indicators.

[0024] The evaluation method for the comprehensive capability of low-altitude flight management and service systems proposed by related technologies does not introduce a system structure model to characterize the relationship between capability elements, nor does it have an analytical mechanism that comprehensively considers the influence of indicators and the value of improvement. It is difficult to characterize the mutual influence between evaluation indicators, resulting in the evaluation results failing to truly reflect the system's operating status and making it difficult to accurately identify the capability bottlenecks that have a key restrictive effect on the overall system capability improvement under multiple indicators and multiple constraints. Specifically, this is reflected in: (1) The system capability evaluation results lack dynamic correlation: Related methods regard each indicator as an independent evaluation unit, which cannot describe the linkage effect of indicator changes on other indicators and the overall system capability. Its evaluation results have obvious static characteristics. (2) The system capability bottleneck identification method is one-sided: Related technologies usually identify system shortcomings by screening low-scoring indicators, without considering the degree of influence of indicators in the overall system capability structure and the resource cost required for improvement, which easily leads to the distortion of bottleneck judgment. (3) The evaluation model lacks system structure expression: Related methods fail to incorporate the structural relationships such as coupling, coordination, and constraints between capability elements within the system into the model, resulting in the evaluation process being disconnected from the real system operation mechanism.

[0025] In view of this, this application proposes a dynamic evaluation and bottleneck identification method for the comprehensive capabilities of a low-altitude flight management and service system. By constructing a multi-level comprehensive capability index system, introducing a capability element coupling network to model the correlation between indicators, and based on the marginal contribution of indicators to the overall system capability, a dynamic bottleneck identification method is proposed. This enables dynamic evaluation and bottleneck diagnosis of the comprehensive capabilities of the low-altitude flight management and service system, reveals the internal linkages within the system's capabilities, avoids static evaluation based solely on indicator scores, and achieves accurate identification of key capability bottlenecks. This provides a quantitative decision-making basis for system capability improvement and resource optimization.

[0026] Figure 1 This is a flowchart illustrating a method for dynamic evaluation of the comprehensive capabilities and bottleneck identification of a low-altitude flight control system, provided as an embodiment of this application.

[0027] like Figure 1 As shown, the method 100 for dynamic evaluation and bottleneck identification of the comprehensive capabilities of low-altitude flight management and service systems includes steps S110-S150.

[0028] Step S110: Establish a hierarchical indicator system and obtain the raw data of the low-altitude flight management service system operation. The hierarchical indicator system includes a first level, a second level, and a third level.

[0029] For example, the hierarchical indicator system describes the hierarchical relationship between levels, including a three-level structure, such as the first level (e.g., the target level), the second level (e.g., the capability level), and the third level (e.g., the indicator level). The capability level can include multiple capability sub-layers (e.g., two capability sub-layers). The hierarchical indicator system is stored in the indicator system management module. The raw data can be obtained from the low-altitude flight management service system (platform) or from external data sources associated with the system (e.g., meteorological information interface, airspace dynamic notification interface, third-party flight plan interface). The raw data includes platform operation logs (e.g., user access logs, command issuance logs, approval logs, alarm records, etc.), business data (e.g., core business data such as flight plans, airspace status, monitored targets, and emergency events), performance monitoring data (e.g., server resource utilization, network status, application interface response time, etc.), user interaction data (user satisfaction survey results, customer service feedback records, knowledge base query logs, etc.), and external data (e.g., meteorological information, airspace dynamic notification, third-party flight plans, etc.).

[0030] Step S120: Calculate the indicators based on the original data and the hierarchical indicator system to obtain the indicator data.

[0031] For example, based on the original data and the hierarchical indicator system, as well as the calculation formulas provided by the indicator system management module (used to manage the name, code, calculation formula, dimension, data source, and scoring threshold of the hierarchical indicator system), the indicator layer is calculated to obtain indicator data. The indicator data includes the maximum number of concurrent users, the online rate of service objects, the success rate of instruction issuance, the overall airspace utilization rate, the average time for plan approval, the controllable compliance rate of safety intervals, the false alarm rate of algorithm alarms, the target resolution, the end-to-end latency, the consistency of multi-source data fusion, the advance warning of high-impact weather, the response time of abnormal events, the time to compress rescue windows, the data accuracy, the average interval between failures, the network security protection level, the vulnerability repair time, the user satisfaction, the system functional availability rate, the abnormal response resolution rate, etc.

[0032] Step S130: Based on the indicator data, the hierarchical indicator system is constructed and quantified to obtain the capability coupling network.

[0033] For example, the elements corresponding to the first level (target layer), first capability level, second capability level, and first indicator level (e.g., the first level may include platform comprehensive capabilities, which are the elements corresponding to the first level) are processed into nodes (e.g., platform comprehensive capabilities are network nodes) to obtain the node data corresponding to the capability coupling network. Based on the indicator data, contribution weights (e.g., between the first indicator level and the second capability level, between the second capability level and the first capability level, and between the first capability level and the first level), coupling weights (between node data in the first capability level, between node data in the second capability level), and association weights (between node data in the first indicator level) are assigned to the first level, first capability level, second capability level, and first indicator level. The calculation yields the weight relationships between node data (representing the weights of contribution edges, coupling edges, and association edges; contribution edges include the edges between capability nodes corresponding to the first level and capability nodes corresponding to the first capability level, the edges between capability nodes corresponding to the first capability level and capability nodes corresponding to the second capability level, and the edges between capability nodes corresponding to the second capability level and indicator nodes corresponding to the first indicator level; coupling edges include the edges between capability nodes corresponding to the first capability level and the edges between capability nodes corresponding to the second capability level; association edges include the edges between indicator nodes corresponding to the first indicator level). Based on the weight relationships and node data, a capability coupling network is constructed (with node data as nodes, contribution edges, coupling edges, and association edges as edges, and weight relationships as the weights of the edges).

[0034] Step S140: Based on the indicator data and the capability coupling network, a comprehensive dynamic evaluation of the low-altitude flight management service system is conducted to obtain dynamic evaluation results.

[0035] For example, for the indicator layer, the indicator data is used as the initial score data. For the capability layer and the target layer, the indicator data is weighted based on the contribution weight to obtain the initial score data corresponding to the capability layer and the target layer. The node data is iteratively updated based on the initial score data to obtain the updated score data corresponding to the node data. When the difference between the updated score data and the initial score data meets the preset threshold, the iteration stops. The updated score data at this time is used as the target score data, and the total score data can be obtained based on the target score data.

[0036] Step S150: Based on the dynamic evaluation results and the capability coupling network, bottleneck identification processing is performed on the low-altitude flight control system to obtain bottleneck identification results. The bottleneck identification results represent the nodes that restrict the capability of the low-altitude flight control system.

[0037] For example, an adjacency matrix can be obtained based on the capability coupling network (node ​​data, contribution weight, coupling weight, association weight). The feature vector can be determined based on the adjacency matrix, thereby obtaining the influence degree corresponding to the node data. Bottleneck factors are calculated based on the influence degree and target score data, and compared with preset rules (including sorting all bottleneck factors from smallest to largest, and classifying them into first-level bottlenecks, second-level bottlenecks, and third-level bottlenecks according to quantile thresholds, such as upper quartile, median, lower quartile, etc. For example, for the upper quartile, the three dividing points of the four equal parts after sorting all bottleneck factors from smallest to largest are the values ​​located at the 75th position of the bottleneck factors) to obtain the bottleneck identification result. The bottleneck identification result includes the identified bottleneck node data.

[0038] As can be seen, the dynamic evaluation and bottleneck identification method for the comprehensive capabilities of the low-altitude flight control system proposed in this application establishes a hierarchical indicator system for the low-altitude flight control system. Based on raw data, it completes indicator calculation, network construction, and quantitative processing, constructing a capability coupling network that reflects the complex relationships among various elements within the system. Through the collaborative analysis of this network model and the dynamic evaluation results, real-time dynamic evaluation of the system's comprehensive capabilities is achieved. On this basis, key bottleneck nodes that restrict the overall performance of the system are accurately identified, improving the accuracy and timeliness of bottleneck identification and providing a quantitative decision-making basis for system capability improvement and resource optimization.

[0039] The embodiments of this application provide a dynamic evaluation and bottleneck identification system for the comprehensive capabilities of a low-altitude flight control and service system. The dynamic evaluation and bottleneck identification system for the comprehensive capabilities of a low-altitude flight control and service system includes: an acquisition module (data acquisition module, indicator system management module), a calculation module (data preprocessing module, indicator system management module), a processing module (capability coupling network construction module), an evaluation module (capability assessment module), and an identification module (bottleneck identification and diagnostic output module).

[0040] The acquisition module (data collection module and indicator system management module) is used to establish a hierarchical indicator system and acquire raw data from the low-altitude flight management system.

[0041] Specifically, the data acquisition module is responsible for collecting raw data from the target low-altitude flight service system (platform) and its associated external data sources. The data sources are extensive, including: (1) Platform operation logs: such as user access logs, instruction issuance logs, approval logs, alarm records, etc. (2) Business database: storing core business data such as flight plans, airspace status, monitored targets, and emergency events. (3) System performance monitoring data: including server resource utilization, network status, application interface response time, etc. (4) External data interfaces: accessing meteorological information, airspace dynamic notices, third-party flight plans, etc. (5) User interaction data: user satisfaction survey results, customer service feedback records, knowledge base query logs, etc. The established hierarchical indicator system is stored in the indicator system management module.

[0042] The calculation module (data preprocessing module and indicator system management module) is used to perform indicator calculations based on the raw data and hierarchical indicator system to obtain indicator data.

[0043] Specifically, the indicator system management module is used to store and maintain the multi-level, structured comprehensive capability indicator system (hierarchical indicator system) defined in this invention. This module manages the names, codes, calculation formulas, dimensions, data sources, scoring thresholds, and hierarchical relationships between indicators. It forms the basis for dynamic comprehensive capability evaluation of the low-altitude flight management system. Based on the raw data, the hierarchical indicator system, and the calculation formulas provided by the indicator system management module, it calculates the indicator layers to obtain indicator data. The data preprocessing module cleans the collected raw data and transforms and standardizes the indicator data calculated from the cleaned raw data, normalizing indicator data (indicator values) of different dimensions and ranges to the [0,1] interval. Positive indicators (higher values ​​are better) and negative indicators (lower values ​​are better) use different normalization functions to unify the scale.

[0044] The processing module (capability coupling network construction module) is used to construct and quantify the hierarchical indicator system based on indicator data to obtain the capability coupling network.

[0045] Specifically, the capability coupling network construction module abstracts the capability dimensions (capability elements corresponding to the target layer and capability layer) and specific indicators (indicator elements corresponding to the indicator layer) in the hierarchical indicator system into network nodes, and quantifies the mutual influence relationships between nodes based on historical data (indicator data) to construct a "capability coupling network". This network not only includes the contribution relationship of indicators to capabilities, but also depicts the complex linkage and constraint relationships between capabilities and indicators.

[0046] The evaluation module (capability assessment module) is used to conduct a comprehensive dynamic evaluation of the low-altitude flight management and service system based on indicator data and capability coupling network, and obtain dynamic evaluation results.

[0047] Specifically, the capability assessment module calculates capability scores at each level and the overall system score based on the capability coupling network and standardized data (standardized indicator data). The calculation process of this module incorporates the mutual influence between network nodes, rather than a simple linear weighting, thus obtaining dynamic assessment results that better reflect the internal linkage status and true level of the system.

[0048] The identification module (bottleneck identification and diagnosis output module) is used to perform bottleneck identification processing on the low-altitude flight control system based on dynamic evaluation results and capability coupling network, and obtain bottleneck identification results. The bottleneck identification results represent the nodes that restrict the capability of the low-altitude flight control system.

[0049] Specifically, the bottleneck identification and diagnosis output module identifies key bottlenecks that restrict the improvement of the system's overall capabilities based on the capability coupling network and dynamic evaluation results. The core includes: (1) calculating the "bottleneck factor" of each node, which comprehensively considers the node's own performance and its global influence in the network. (2) classifying bottleneck levels according to the size of the bottleneck factor. (3) tracing the influence path of the bottleneck node in the coupling network and intuitively showing how it affects other parts of the system. (4) automatically generating a structured diagnostic report containing a comprehensive score, capability analysis, bottleneck list, influence path, and improvement suggestions.

[0050] Figure 2 A schematic diagram of the workflow of a dynamic evaluation and bottleneck identification system for the comprehensive capabilities of a low-altitude flight control system provided in another embodiment of this application.

[0051] As shown in Figure 2, the workflow of the low-altitude flight management service system's comprehensive capability dynamic evaluation and bottleneck identification system includes, for example, S201-S207.

[0052] S201, Data Acquisition Module.

[0053] For example, raw data of the indicators are collected.

[0054] S202, Indicator System Management Module.

[0055] For example, maintain the indicator dictionary and hierarchical structure.

[0056] S203, Data Preprocessing Module.

[0057] For example, cleaning and standardizing data.

[0058] S204, Capability Coupled Network Building Module.

[0059] For example, modeling the relationship between metrics / capabilities.

[0060] S205, Competency Assessment Module.

[0061] For example, dynamic computing power and overall score.

[0062] S206, Bottleneck Identification and Diagnosis Output Module.

[0063] For example, bottlenecks can be identified, paths can be traced, and reports can be generated.

[0064] S207 outputs a diagnostic report and optimization suggestions.

[0065] In one example, the dynamic evaluation and bottleneck identification system for the comprehensive capabilities of the low-altitude flight management service system can be deployed on a local server or provided as a cloud service; the functional modules can be integrated into a single software system or deployed and invoked as distributed microservices.

[0066] Figure 3 A schematic diagram of the hierarchical index system structure provided for the implementation of this application.

[0067] In one example, the second level includes a first capability level and a second capability level, and the third level includes a first indicator level. Based on indicator data, a network is constructed and quantified for the hierarchical indicator system to obtain a capability-coupled network. This includes: performing node-based processing on the first level, first capability level, second capability level, and first indicator level to obtain capability nodes and indicator nodes; calculating contribution weights for the capability nodes corresponding to the first level, the capability nodes corresponding to the first capability level, the capability nodes corresponding to the second capability level, and the indicator nodes corresponding to the first indicator level based on indicator data to obtain contribution weights. The contribution weights represent the weights corresponding to the contribution edges, which include edges between capability nodes corresponding to the first level and capability nodes corresponding to the first capability level, edges between capability nodes corresponding to the first capability level and capability nodes corresponding to the second capability level, and edges between capability nodes corresponding to the second capability level. The edges between the indicator nodes corresponding to the first indicator level are calculated. Based on indicator data and contribution weights, the capability nodes corresponding to the first and second capability levels are weighted to obtain score data. Based on the score data, the deviation coefficients of the capability nodes corresponding to the first and second capability levels are calculated to obtain coupling weights. The coupling weights represent the weights corresponding to the coupling edges, which include edges between capability nodes corresponding to the first and second capability levels. Based on the indicator data, the correlation coefficients of the indicator nodes corresponding to the first indicator level are calculated to obtain association weights. The association weights represent the weights corresponding to the association edges, which include edges between indicator nodes corresponding to the first indicator level. Based on the capability nodes, indicator nodes, contribution weights, coupling weights, and association weights, a capability coupling network is obtained.

[0068] like Figure 3 As shown, specifically, the first level is the target layer (platform comprehensive capabilities), the second level is the capability layer, including the first capability level and the second capability level, and the third level is the indicator layer (first indicator level). The platform comprehensive capabilities are decomposed into six core capability dimensions, resulting in the first capability level, which includes: C1: Service access capability, evaluating the system's ability to handle the scale and stability of user and device connections. C2: Flight management capability, evaluating the system's efficiency and effectiveness in airspace planning, dynamic scheduling, safety control, and flight plan approval. C3: Flight service support capability, evaluating the quality of flight support services provided by the system, such as surveillance, intelligence, and meteorology. C4: Emergency management capability, evaluating the system's effectiveness in identifying, responding to, and supporting rescue efforts for abnormal events. C5: Platform support capability, evaluating the system's basic support level in areas such as data governance, reliability, and security. C6: User service effectiveness, evaluating the experience and support effects ultimately provided to users by the system.

[0069] The first capability level is further subdivided into the second capability level, which are as follows: C11: Scale carrying capacity, C12: Connectivity quality, C21: Airspace planning, C22: Airspace use, C23: Flight safety, C24: Flight approval efficiency, C31: Surveillance service capability, C32: Intelligence service capability, C33: Meteorological service capability, C41: Emergency response capability, C42: Rescue efficiency, C51: Data governance capability, C52: Reliability, C53: Security, C61: Service experience, and C41: Support capability.

[0070] The second capability level is further subdivided into the first indicator level, which includes: Scale carrying capacity C11 (maximum number of concurrent access devices, number of supported service object types); Connection quality C12 (service object online rate, command issuance success rate, etc.); Airspace planning C21 (overall airspace utilization rate, flexible airspace utilization rate, airspace reconstruction response time); Airspace usage C22 (platform controllable execution deviation rate, airspace dynamic scheduling response speed); Flight safety C23 (safety interval controllable compliance rate, plan conflict detection accuracy, algorithm alarm false alarm rate, effective conflict warning lead time); Flight approval efficiency C24 (average plan approval time, first-time completion rate of application process); Surveillance service capability C31 (target resolution, information update frequency, end-to-end latency, alarm latency, multi-source data fusion consistency); and Intelligence service capability C32 (intelligence...). The indicators for service accuracy and intelligence service timeliness are as follows: meteorological service capability (C33) includes meteorological service scenario coverage, meteorological service update frequency, and advance warning of high-impact weather; emergency response capability (C41) includes accurate identification rate of abnormal situations and response time of abnormal events; rescue efficiency (C42) includes rescue window compression time and proportion of rescue instructions covering key institutions; data governance capability (C51) includes data accuracy, data timeliness, data integrity, and data sharing efficiency; reliability (C52) includes mean time between failures and mean time to repair; security (C53) includes network security protection level, sensitive data encryption coverage, network attack self-healing rate, and vulnerability repair time; service experience (C61) includes user satisfaction and system function availability; and support capability (C62) includes anomaly response resolution rate and knowledge base call resolution rate.

[0071] It should be noted that the number and categories of indicators can be configured according to the actual platform business, and this application does not limit the specific set of indicators.

[0072] For example, all elements in the hierarchical indicator system can be abstracted into network nodes. One capability node corresponding to a target layer (capability node corresponding to the first level), six first capability nodes (first capability level), sixteen second capability nodes (second capability level), and forty-four first indicator nodes (first indicator level, as an example) constitute capability nodes and indicator nodes respectively. The node set is denoted as... Capability nodes include node data corresponding to the first level, the first capability level, and the second capability level; indicator nodes include node data corresponding to the first indicator level.

[0073] Define capability coupling network edges and relationships: Establish three types of edges to characterize the internal relationships of the system: (1) Contribution edge (contribution weight): connects the lower-level indicators to the upper-level capabilities (such as "multi-source data fusion consistency" → "monitoring service capability"), reflecting the composition relationship; (2) Association edge (association weight): connect the indicator nodes at the same level (such as "average time for plan approval" and "user satisfaction"), reflecting business linkage; (3) Coupling edge (coupling weight): connect different capability nodes (such as "flight management capability" and "flight service support capability"), reflecting capability synergy.

[0074] To transform capability-coupled networks from qualitative structures into computable quantitative models, it is necessary to quantify the weights (i.e., association strengths) of various edge types in the network. This application employs a combination of data-driven and knowledge-driven methods for quantification, based on the edge type.

[0075] For example, for the correlation weight, the correlation coefficient is calculated for the indicator nodes corresponding to the first indicator level based on the indicator data: the correlation edge reflects the statistical correlation or causal linkage between the specific evaluation indicators (indicator data) at the bottom layer (first indicator level), and is quantified by the correlation coefficient method based on historical indicator data.

[0076] For example, for any two index nodes C and Collect its in The time series values ​​within each evaluation period (indicator data within m evaluation periods) are denoted as follows: and ,in and They represent in Indicator node C within a cycle and C The values ​​of the indicator data are taken. The Pearson correlation coefficient between the two is calculated. As shown in formula (1): (1) in, and Sequences and value, For the evaluation period, the correlation coefficient The absolute value of the denominator indicates the strength of the linear correlation, and the sign indicates the direction of the correlation.

[0077] The calculated Pearson correlation coefficient Absolute value as an indicator of related edges Weights (association weights) As shown in formula (2): (2) Only when Only when the value exceeds a preset threshold (e.g., 0.3) will the associated edge be retained in the capability coupling network to focus on key associations.

[0078] In one example, in addition to using the Pearson correlation coefficient, grey relational analysis, mutual information, or expert scoring based on business rules can also be used to quantify the correlation weight between indicator nodes.

[0079] For example, contribution weights are calculated for capability nodes corresponding to the first level, capability nodes corresponding to the first capability level, capability nodes corresponding to the second capability level, and indicator nodes corresponding to the first indicator level based on indicator data. This includes: calculating contribution weights for capability nodes corresponding to the first level and capability nodes corresponding to the first capability level based on score data to obtain a first contribution weight; calculating contribution weights for capability nodes corresponding to the first capability level and capability nodes corresponding to the second capability level based on score data to obtain a second contribution weight; and calculating contribution weights for capability nodes corresponding to the second capability level and indicator nodes corresponding to the first indicator level based on indicator data to obtain a third contribution weight.

[0080] Specifically, contribution weights include the weights from the indicator node to the second capability node, from the second capability node to the first capability node, and from the first capability node to the corresponding capability node in the first level. These weights reflect the relative importance of the lower-level element to the higher-level capability. Contribution weights can be calculated using subjective weighting methods (such as the Analytic Hierarchy Process (AHP), objective weighting methods (such as the entropy weighting method), or a combination of subjective and objective weighting methods, thereby obtaining the fundamental structural weights (contribution weights) of the network. This application is explained using the objective empowerment method.

[0081] For example, for the third contribution weight, the indicator data corresponding to the first indicator level is used as the measurement standard. The indicator data is normalized, and information entropy and information utility value are calculated based on the normalized indicator data. For example, for the maximum number of concurrent access devices and the number of service object types supported (indicator data) corresponding to the scale carrying capacity C11 (second capability node), the information entropy and information utility value are calculated for each indicator data separately. The proportion of the information utility value of each indicator data to the sum of the information utility values ​​of all indicator data corresponding to the second capability node is used as the third contribution weight. .

[0082] Based on the third contribution weight By weighting the indicator data, we can obtain the score data corresponding to the second capability level (for example, based on the third contribution weight). By weighting the maximum number of concurrent access devices and the number of supported service object types (indicator data), the score data for scale carrying capacity C11 (second capability node) can be obtained. Using the score data of the second capability level capability node as the measurement standard, the score data is normalized, and information entropy and information utility value are calculated based on the normalized score data. For example, for the score data corresponding to scale carrying capacity C11 (second capability node) and the score data corresponding to connection quality C12 (second capability node), the information entropy and information utility value are calculated respectively. The proportion of the information utility value of each score data to the sum of the information utility values ​​of all score data corresponding to the first capability node is used as the second contribution weight. .

[0083] Based on the second contribution weight By weighting the score data corresponding to the second capability node, we can obtain the score data corresponding to the next capability level (e.g., based on the second contribution weight). The score data for service access capability C1 (first capability node) is obtained by weighting the score data for scale carrying capacity C11 and connection quality C12. Using the score data of the first capability level capability nodes as the benchmark, the score data is normalized. Information entropy and information utility values ​​are then calculated based on the normalized score data. For example, for the score data corresponding to service access capability C1 (first capability node) and the score data for flight management capability C2, flight service support capability C3, emergency management capability C4, platform support capability C5, and user service efficiency C6, information entropy and information utility values ​​are calculated respectively. The proportion of the information utility value of each score data to the sum of the information utility values ​​of all score data of the corresponding first-level capability nodes is used as the first contribution weight. .

[0084] In one example, the scoring data includes first scoring data corresponding to the first capability level and second scoring data corresponding to the second capability level. Based on the scoring data, the deviation coefficients of the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are calculated to obtain the coupling weights, including: calculating the deviation coefficients of the capability nodes corresponding to the first capability level based on the first scoring data to obtain the first coupling weights; and calculating the deviation coefficients of the capability nodes corresponding to the second capability level based on the second scoring data to obtain the second coupling weights.

[0085] Specifically, based on the score data of the first capability node (first score data) and the score data of the second capability node (second score data) obtained above, an initial score sequence (including the first score data for a period of time and the second score data for a period of time) is obtained for each first capability node and second capability node. For example, for the first capability node Ci, the score sequence for a period of time is obtained. Calculate any two first capability nodes based on the deviation coefficient. and Coupling degree of the scoring sequence (first coupling weight) As shown in formula (3): (3) in, and Don't be a capability node and The sum of squared deviations of the score sequence from its ideal state (such as a perfect score). The calculation method is shown in formula (4): (4) in, Represents capability nodes The score data in the kth evaluation period, Set its corresponding ideal score (e.g., 1). The larger the first coupling weight, the more coordinated and stronger the coupling between the two capability dimensions at the development level. This coupling value is directly used as the capability coupling edge. ( and The weights of the edges between them are calculated. Similarly, the second coupling weights of the second capability nodes are calculated.

[0086] In the above embodiments, specific weight values ​​are assigned to all edges in the capability coupling network, thereby obtaining a complete, weighted "capability coupling network" model, which lays a quantitative foundation for subsequent dynamic evaluation and bottleneck identification.

[0087] In one example, a comprehensive dynamic evaluation of the low-altitude flight management system is performed based on indicator data and a capability coupling network to obtain dynamic evaluation results. This includes: weighting the first score data corresponding to the first capability level based on a first contribution weight to obtain third score data; iteratively updating capability nodes and indicator nodes based on indicator data, first score data, second score data, and third score data. The iterative update calculation includes updating the score data of each node in the capability coupling network based on the score data and weights of neighboring nodes. Neighboring nodes include capability nodes and / or indicator nodes, and weights include at least one of contribution weight, coupling weight, and association weight, to obtain target score data and total score data corresponding to capability nodes and indicator nodes; and obtaining dynamic evaluation results based on the target score data and total score data.

[0088] Specifically, by weighting the first contribution weight and the first score data, the score data corresponding to the target layer (third score data) can be obtained. For example, the score data corresponding to service access capability C1 (first capability node), as well as the score data of flight management capability C2, flight service support capability C3, emergency management capability C4, platform support capability C5, and user service efficiency C6, and the first contribution weight. By performing weighted calculations, we can obtain the score data corresponding to the platform's overall capabilities (the third score data).

[0089] The initial scores of all nodes in the network (including capability nodes and indicator nodes) are denoted as follows: For indicator nodes, This refers to standardized indicator data; for capability nodes (capability nodes corresponding to the first level, first capability node, and second capability node). This corresponds to the first, second, or third score data. Multiple rounds of iterative calculation are performed on each node, as shown in formula (5): (5) Where: t represents the number of iterations ( ), It is any node in the network In the The score data after rounds of iterations It is a node The set of all neighboring nodes in the network (including indicator nodes and capability nodes connected to it). From neighboring nodes Pointing to node The edge weights (which may include association weights, contribution weights, or coupling weights). Neighboring nodes In the Score data after -1 round of iterations Damping factor ( This is used to balance the node's initial state with the network influence it receives. Typical values ​​are between 0.5 and 0.8, for example... This indicates that the node's final score is more likely to reflect its initial state. This indicates a greater reliance on the influence of online dissemination.

[0090] In one example, in addition to neighbor-based iterative smoothing algorithms, graph neural networks (GNNs), random walks (such as the PageRank algorithm), or other representation learning methods based on network structures can also be used to compute the dynamic state values ​​of nodes.

[0091] In the above embodiments, the evaluation process is dynamic and networked. In the traditional weighted summation model, the score of a higher-level node is a static linear combination of the scores of its subordinate nodes. This application abandons this "island-like" calculation and proposes a dynamic evaluation algorithm based on network information propagation. It considers a node's final "dynamic score" as a fusion of its own basic state and the influence information from its network neighbors. This process simulates the actual situation where performance changes in a certain link of the system can affect other links through predefined correlations. This method not only calculates the state of the nodes (indicators or capability dimensions), but more importantly, it quantifies the mutual influence between nodes through a constructed "capability coupling network" model, thereby achieving a deep diagnosis of the system's overall capabilities.

[0092] In one example, the capability nodes and indicator nodes are iteratively updated based on indicator data, first score data, second score data, and third score data to obtain the target score data and total score data corresponding to the capability nodes and indicator nodes. This includes: iteratively updating the capability nodes and indicator nodes based on indicator data, first score data, second score data, and third score data to obtain the target data after each iteration; calculating the difference between the indicator data, first score data, second score data, third score data, and target data to obtain the change data; and when the change data is less than a preset threshold, the target data is used as the target score data, and the total score data is obtained based on the target score data.

[0093] Specifically, according to formula (5), multiple rounds of iterative calculations are performed on each indicator node and capability node. When the score change (change data) of all nodes (indicator nodes and capability nodes) is less than the preset threshold (e.g., when the absolute value of the change is less than 0.001), the iteration stops. At this time, the network state tends to stabilize, and the final converged score (target data) of the indicator node and capability node is the dynamic evaluation score of that node. (Target score data), dynamic total score of system comprehensive capability (capability nodes corresponding to the first level). The total score data can be obtained by weighted aggregation of the dynamic scores (target score data) of all first-level capability nodes.

[0094] In the above embodiments, an indicator with an initial low score may have its scores reduced by iteratively lowering the scores of multiple high-influence nodes if it is strongly correlated with them; conversely, a node with an initial score that is acceptable but is on a critical influence path may have its dynamic score reduced due to being dragged down by other bottlenecks. Thus, the evaluation results have both dynamism and systematicity.

[0095] The goal of bottleneck identification is not simply to find the node with the lowest score, but to find the node that has the greatest constraint on improving the overall system capability. The bottleneck identification and diagnosis output module identifies bottlenecks based on a bottleneck factor calculated as the product of "shortcoming severity" and "network influence".

[0096] In one example, bottleneck identification is performed on the low-altitude flight management system based on dynamic evaluation results and a capability coupling network. The bottleneck identification results include: determining the adjacency matrix based on capability nodes, indicator nodes, contribution weights, coupling weights, and association weights; calculating features based on the adjacency matrix to obtain feature vectors; and performing bottleneck identification on the low-altitude flight management system based on the feature vectors and dynamic evaluation results to obtain bottleneck identification results.

[0097] Specifically, eigenvector centrality is used as the network influence of nodes (indicator nodes and capability nodes). The metric indicates that the importance of a node depends on the number and quality of its neighboring nodes, and the more important the neighboring nodes are, the more important the node is. This perfectly aligns with the characteristic in capability-coupled networks where a node's influence is amplified when it affects multiple important nodes through strong associations. Its calculation method is shown in formula (6), which involves solving the network adjacency matrix. The principal eigenvectors (eigenvectors) are obtained as follows: (6) in, It is an adjacency matrix containing all types of edge weights (contribution weights, association weights, coupling weights). (The rows and columns correspond to all capability nodes and indicator nodes. For each element in the adjacency matrix W...) , indicating from node To the node Edge weight (0 if there are no edges). yes The largest eigenvalue, It is the corresponding eigenvector (eigenvector).

[0098] In one example, bottleneck identification processing is performed on the low-altitude flight control system based on feature vectors and dynamic evaluation results to obtain bottleneck identification results. This includes: normalizing the feature vectors to obtain normalized feature vectors; performing bottleneck identification processing on the low-altitude flight control system based on the normalized feature vectors and target score data to obtain a bottleneck factor. The bottleneck factor is obtained by: performing a difference operation based on preset values ​​and target score data to obtain a difference result; multiplying the difference result with the normalized feature vector to obtain the bottleneck factor; and obtaining the bottleneck identification result based on the bottleneck factor and preset rules.

[0099] Specifically, vector The first in element That is, a node The eigenvector centrality (network influence) is usually normalized to make... For each node in the network Its bottleneck factor The definition is shown in formula (7): (7) in: Representative node The degree of "shortcomings" is dynamically scored. The lower the target score, the larger the value of this item. Representative node Global influence in a coupled network, with a default value of 1. Bottleneck factor. This approach integrates information from two dimensions: "the node is performing poorly" and "the node has a significant impact." Using a multiplicative operation means that if either the node's "weakness" or "network influence" is too low, the bottleneck factor will decrease, thus accurately identifying nodes that perform poorly but have a large impact as key bottlenecks. The higher the value, the greater the potential benefit (marginal contribution) of improving this node to the overall system capability, and this node is the key bottleneck.

[0100] Bottleneck factors calculated based on all nodes (including capability nodes and indicator nodes) The set is used to classify levels by setting quantile thresholds (preset rules). For example: Level 1 bottleneck (critical): ( The upper quartile represents the three dividing points where all bottleneck factors are sorted in ascending order and divided into four equal parts (e.g., the value at the 75th percentile of the bottleneck factor); secondary bottleneck (important): ( (This refers to the median, such as the value at the 50th percentile of the bottleneck factor); Level 3 bottleneck (potential): ( (This refers to the lower quartile, such as the value at the 25th percentile of the bottleneck factor); Non-bottleneck: The bottleneck level can be obtained by comparing the bottleneck factor with the quantile threshold (preset rule).

[0101] In one example, in addition to eigenvector centrality, network centrality metrics such as betweenness centrality and proximity centrality can also be used, or weighted calculations can be performed in combination with the difficulty (cost) of node improvement.

[0102] In one example, the method for dynamic evaluation and bottleneck identification of the comprehensive capabilities of the low-altitude flight management system also includes: performing path tracing processing based on the bottleneck identification results to obtain the impact path of the low-altitude flight management system.

[0103] Specifically, for each identified bottleneck node In capability-coupled networks, influence path tracing is performed. Graph theory algorithms (such as a modified Dijkstra's algorithm, which changes the traditional "shortest path" objective to finding the path with the maximum product of edge weights to maximize the cumulative effect of influence propagation) are used to find the path from the bottleneck node. The top-N most influential paths leading to the overall system capability node (or core capability node) are identified. An example output format is: "Bottleneck Indicator 3-1-5 Multi-source Data Fusion Consistency → (Contribution Strength 0.8) → Capability Node Flight Service Support Capability → (Coupling Strength 0.7) → Capability Node Flight Management Capability → … → System Overall Capability." This path clearly reveals the bottleneck transmission chain, greatly enhancing the interpretability of the diagnostic results.

[0104] The dynamic evaluation and bottleneck identification method for the comprehensive capability of the low-altitude flight management service system proposed in this application, compared with the related linear weighted evaluation method, brings the following significant benefits by introducing a capability coupling network and a dynamic bottleneck identification mechanism: (1) The evaluation results change from "static isolation" to "dynamic correlation": Related methods assume that the indicators are independent, and the calculation results are a static snapshot. This application, through capability coupling network modeling, enables the evaluation process to capture and quantify the complex interactions between indicators and capabilities, and the evaluation results obtained can better reflect the real and dynamic operating status of the system after the linkage of various components. (2) Bottleneck identification is deepened from "appearance screening" to "root cause diagnosis": Related methods identify shortcomings by finding low-scoring items, which may ignore those hidden bottlenecks that are still acceptable in score but occupy key positions in the network. The bottleneck factor proposed in this application can accurately locate the key nodes that have the greatest constraint on the overall capability improvement of the system, even if their absolute scores are not the lowest. This avoids the misallocation of optimized resources and realizes the transformation from "treating the symptoms" to "treating the root cause". (3) The diagnostic output is upgraded from "score listing" to "path insight": This application not only outputs scores and bottleneck lists, but also provides a visualized bottleneck impact path. This is like providing the system with a detailed "diagnostic report" and "impact map", enabling managers and technicians to clearly understand the root causes of bottlenecks and the scope of their impact, thereby formulating accurate and efficient collaborative optimization strategies. (4) The methodological framework has high scalability and universality: The methodology framework of "constructing a capability coupling network - dynamic evaluation - identifying bottlenecks based on marginal contribution" proposed in this application does not depend on a specific set of indicators. As long as the system has multi-dimensional and interrelated capability elements, this framework can be applied for analysis. Therefore, this method can be easily extended to capability evaluation and optimization scenarios of other complex integrated management and control systems (such as smart transportation and industrial internet platforms).

[0105] To facilitate understanding, this application provides the following typical application scenario to illustrate the implementation process of the method. The main work includes the following: Step 1: Data Acquisition and Network Construction (1) Data collection The system collects operational data (raw data) from the low-altitude flight management system over the past 30 days and calculates the raw values ​​(indicator data) of key indicators (such as "average time for plan approval", "consistency of multi-source data fusion", "user satisfaction", etc.) under the six capability dimensions defined in this method. The indicator values ​​are then standardized to fall within the range [0,1].

[0106] (2) Construction of capability coupling network Indicators and capabilities are abstracted as network nodes. Based on historical data (indicator data), the mutual influence between nodes is analyzed and quantified. For example, according to the correlation coefficient method (Pearson correlation coefficient) mentioned earlier, it was found that "plan approval time" is negatively correlated with "user satisfaction" (correlation weight 0.62), and "monitoring data fusion consistency" is strongly positively correlated with "security alarm accuracy" (correlation weight 0.75). These correlation strength values ​​have been thresholded to ensure that the network only includes statistically significant or business-important correlations. These correlation weights are added to the network as "edges" to form a system relationship model containing nodes and weighted edges.

[0107] Step 2: Dynamic Capability Assessment The standardized indicator data is input into the constructed "capability coupling network". Through the calculation of the network model (simulating the interaction between indicators), the dynamic scores (target score data) of each capability dimension are obtained. The evaluation found that the dynamic score of the system's "flight service support capability" (0.65) was significantly lower than its simple indicator average score (0.72). This was because the performance of several related indicators (such as data fusion consistency) was poor, which lowered the final score of this capability dimension through the network.

[0108] Step 3: Dynamic Bottleneck Identification Calculating the bottleneck factor: In the network, not only is the dynamic score (S) of a node (indicator / capability) calculated, but its network influence (I) is also calculated. For the node "Multi-source data fusion consistency", its dynamic score is low (S=0.45), but its network influence is high (I=0.85), therefore its bottleneck factor B = (1-0.45) × 0.85 = 0.47.

[0109] Identifying key bottlenecks: Comparing the bottleneck factors of all nodes, this indicator (B=0.47) was identified as a first-level key bottleneck. Another isolated indicator with a lower score (S=0.40) but weaker correlation, however, had a smaller bottleneck factor and was not identified as a key bottleneck.

[0110] Analyze the impact path: The system automatically analyzes and outputs the impact path of the bottleneck, such as... Figure 4 As shown: "Poor consistency in multi-source data fusion → Seriously impacts → Surveillance service capabilities → Consequently restricts → Flight service support capabilities → Ultimately lowers → System overall capabilities." This path clearly reveals the chain of transmission of the problem.

[0111] Step 4: Output the diagnostic report The system generates a diagnostic report, the main contents of which include: (1) the dynamic total score of comprehensive ability (total score data) and level; (2) the dynamic score of each ability dimension (target score data) and comparative analysis; (3) a list of key bottlenecks (such as "multi-source data fusion consistency" is a first-level bottleneck), with bottleneck factor values ​​attached; (4) the impact path diagram of key bottlenecks; and (5) targeted improvement suggestions (such as: prioritizing the upgrading of data fusion algorithms and calibrating heterogeneous sensors).

[0112] In the above embodiments, the relevant evaluation methods may only identify low "user satisfaction" scores and suggest a general "service optimization". This embodiment, however, delves deeper, revealing that one of the root causes of low satisfaction is the deep-seated technical bottleneck of "data fusion consistency," which amplifies its constraints on the overall system through network effects. Therefore, resources should be prioritized to address this bottleneck, thereby more efficiently improving the system's overall capabilities. This embodiment demonstrates the core advantages of this application, moving from "static scoring" to "dynamic diagnosis," and from "identifying low scores" to "locating key constraints."

[0113] The proposed method and system for dynamic evaluation and bottleneck identification of the comprehensive capabilities of low-altitude flight management and service system achieves the following: (1) abstracting the multi-level evaluation indicators and capability dimensions of the low-altitude flight management and service system into network nodes, constructing a weighted network model by quantifying the correlation strength between indicators, the coupling strength between capabilities, and the hierarchical contribution relationship, and using algorithms such as iterative updates or graph neural networks to calculate dynamic capability scores based on this network model, replacing the traditional linear static weighted summation, so that the evaluation results can reflect the complex linkage relationship within the system; (2) defining the bottleneck factor of a node as the product of its shortcoming degree and network influence, identifying the restrictive node that has the greatest marginal contribution to the overall capability improvement of the system through this factor, and classifying the bottleneck level according to the distribution of factor values. Achieving a shift from static low-score screening to dynamic key constraint identification; (3) A bottleneck impact path tracing and visualization output mechanism: After identifying the bottleneck node, based on the constructed capability coupling network, graph theory is used to trace the most important impact propagation path from the bottleneck node to the system comprehensive capability target node, and the complete impact link containing node, edge and weight information is output in a visualization manner, providing a clear and interpretable diagnostic basis for system optimization; (4) An automated diagnostic system architecture integrating the above methods: The system architecture includes modules such as data collection, indicator system management, capability network construction, dynamic evaluation and bottleneck identification, which can realize the full-process automated processing from multi-source data collection, network modeling, dynamic evaluation to bottleneck diagnosis and report generation.

[0114] The proposed method and system for dynamic evaluation and bottleneck identification of the comprehensive capabilities of low-altitude flight management and service systems include: (1) constructing a multi-level capability index system to achieve a structured and quantitative description of system capabilities; (2) introducing a capability element coupling network to model the correlation and influence intensity between indicators, overcoming the problem of neglecting indicator linkage in related methods; (3) dynamically identifying capability bottlenecks that have a key constraint on the improvement of system capabilities by analyzing the marginal contribution of capability elements to the overall system capabilities; and (4) providing quantitative diagnostic basis for system capability optimization based on bottleneck identification. Through the above technical means, a closed loop from system capability evaluation, bottleneck diagnosis to optimization decision support is realized.

[0115] Figure 5 A block diagram of an electronic device provided for another embodiment of this application.

[0116] An embodiment of this application provides an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by one or more processors, which are executed by one or more processors to cause the one or more processors to implement the steps of the method of any of the above embodiments.

[0117] As shown in the figure, for ease of understanding, an embodiment of this application illustrates a specific electronic device 500.

[0118] Electronic device 500 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0119] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0120] Multiple components in electronic device 500 are connected to input / output (I / O) interface 505. These components include: input unit 506, such as a keyboard or mouse; output unit 507, such as various types of displays or speakers; storage unit 508, such as a hard disk or optical disk; and communication unit 509, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0121] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0122] Embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0123] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0124] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0125] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0126] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0127] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0128] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0129] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

Claims

1. A method for dynamic evaluation and bottleneck identification of the comprehensive capabilities of a low-altitude flight control system, characterized in that, The method includes: A hierarchical indicator system is established and raw data on the operation of the low-altitude flight management system is obtained. The hierarchical indicator system includes a first level, a second level, and a third level. Based on the original data and the hierarchical indicator system, indicator calculations are performed to obtain indicator data; Based on the aforementioned indicator data, the hierarchical indicator system is constructed and quantified to obtain a capability coupling network. Based on the index data and the capability coupling network, a comprehensive dynamic evaluation of the low-altitude flight management system is performed to obtain dynamic evaluation results. Based on the dynamic evaluation results and the capability coupling network, bottleneck identification processing is performed on the low-altitude flight control system to obtain bottleneck identification results, wherein the bottleneck identification results represent the nodes that restrict the capabilities of the low-altitude flight control system.

2. The method according to claim 1, characterized in that, The second level includes a first capability level and a second capability level, and the third level includes a first indicator level; the step of constructing and quantifying the hierarchical indicator system based on the indicator data to obtain a capability coupling network includes: The first level, the first capability level, the second capability level, and the first indicator level are processed into nodes to obtain capability nodes and indicator nodes. Based on the indicator data, contribution weights are calculated for the capability nodes corresponding to the first level, the capability nodes corresponding to the first capability level, the capability nodes corresponding to the second capability level, and the indicator nodes corresponding to the first indicator level to obtain contribution weights. The contribution weights represent the weights corresponding to the contribution edges, which include the edges between the capability nodes corresponding to the first level and the capability nodes corresponding to the first capability level, the edges between the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level, and the edges between the capability nodes corresponding to the second capability level and the indicator nodes corresponding to the first indicator level. Based on the indicator data and the contribution weight, the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are weighted and calculated to obtain score data. Based on the score data, the deviation coefficients of the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are calculated to obtain coupling weights. The coupling weights represent the weights corresponding to the coupling edges, and the coupling edges include the edges between the capability nodes corresponding to the first capability level and the edges between the capability nodes corresponding to the second capability level. Based on the indicator data, the correlation coefficient of the indicator nodes corresponding to the first indicator level is calculated to obtain the association weight, wherein the association weight represents the weight of the association edge, and the association edge includes the edge between the indicator nodes corresponding to the first indicator level. Based on the capability nodes, the indicator nodes, the contribution weights, the coupling weights, and the association weights, a capability coupling network is obtained.

3. The method according to claim 2, characterized in that, The contribution weight calculation, based on the indicator data, for the capability nodes corresponding to the first level, the capability nodes corresponding to the first capability level, the capability nodes corresponding to the second capability level, and the indicator nodes corresponding to the first indicator level, to obtain the contribution weight, includes: Based on the score data, the contribution weights of the capability nodes corresponding to the first level and the capability nodes corresponding to the first capability level are calculated to obtain the first contribution weight. Based on the score data, the contribution weights of the capability nodes corresponding to the first capability level and the capability nodes corresponding to the second capability level are calculated to obtain the second contribution weight. Based on the indicator data, the contribution weights of the capability nodes corresponding to the second capability level and the indicator nodes corresponding to the first indicator level are calculated to obtain the third contribution weight.

4. The method according to claim 3, characterized in that, The scoring data includes first scoring data corresponding to the first capability level and second scoring data corresponding to the second capability level; the step of calculating the deviation coefficient of the capability node corresponding to the first capability level and the capability node corresponding to the second capability level based on the scoring data to obtain the coupling weight includes: Based on the first score data, the deviation coefficient of the capability node corresponding to the first capability level is calculated to obtain the first coupling weight; Based on the second score data, the deviation coefficient of the capability node corresponding to the second capability level is calculated to obtain the second coupling weight.

5. The method according to claim 4, characterized in that, The comprehensive dynamic evaluation of the low-altitude flight control system based on the indicator data and the capability coupling network yields dynamic evaluation results, including: The first score data corresponding to the first capability level is weighted and calculated based on the first contribution weight to obtain the third score data. Based on the indicator data, the first score data, the second score data, and the third score data, iterative update calculations are performed on the capability node and the indicator node to obtain the target score data and total score data corresponding to the capability node and the indicator node. The iterative update calculation includes updating the score data of each node in the capability coupling network based on the score data and weights of neighboring nodes. The neighboring nodes include the capability node and / or the indicator node, and the weights include at least one of the contribution weight, the coupling weight, and the association weight. Based on the target score data and the total score data, a dynamic evaluation result is obtained.

6. The method according to claim 5, characterized in that, The iterative update calculation of the capability node and the indicator node based on the indicator data, the first score data, the second score data, and the third score data to obtain the target score data and total score data corresponding to the capability node and the indicator node includes: Based on the indicator data, the first score data, the second score data, and the third score data, the capability node and the indicator node are iteratively updated to obtain the target data after each iteration update; The difference is calculated based on the indicator data, the first score data, the second score data, the third score data, and the target data to obtain the change data; When the changed data is less than a preset threshold, the target data is used as the target score data, and the total score data is obtained based on the target score data.

7. The method according to any one of claims 5-6, characterized in that, Based on the dynamic evaluation results and the capability coupling network, bottleneck identification processing is performed on the low-altitude flight control system to obtain bottleneck identification results, including: Based on the capability nodes, the indicator nodes, the contribution weights, the coupling weights, and the association weights, an adjacency matrix is ​​determined; Feature vectors are obtained by calculating features based on the adjacency matrix. Based on the feature vector and the dynamic evaluation result, bottleneck identification processing is performed on the low-altitude flight control system to obtain the bottleneck identification result.

8. The method according to claim 7, characterized in that, The bottleneck identification process for the low-altitude flight control system based on the feature vector and the dynamic evaluation result, to obtain the bottleneck identification result, includes: The feature vector is normalized to obtain the normalized feature vector; The low-altitude flight control system is subjected to bottleneck identification processing based on the normalized feature vector and the target score data to obtain a bottleneck factor. The bottleneck factor is obtained by performing a difference operation based on a preset value and the target score data to obtain a difference result, and then performing a product operation between the difference result and the normalized feature vector to obtain the bottleneck factor. Based on the bottleneck factor and preset rules, the bottleneck identification result is obtained.

9. The method according to any one of claims 2-6, characterized in that, The method further includes: Based on the bottleneck identification results, path tracing processing is performed to obtain the impact path of the low-altitude flight management system.

10. A dynamic evaluation and bottleneck identification system for the comprehensive capabilities of a low-altitude flight control service system, characterized in that, The system includes: The acquisition module is used to establish a hierarchical indicator system and acquire raw data of the operation of the low-altitude flight management service system. The hierarchical indicator system includes a first level, a second level, and a third level. The calculation module is used to perform indicator calculations based on the original data and the hierarchical indicator system to obtain indicator data; The processing module is used to construct and quantify the hierarchical indicator system based on the indicator data to obtain a capability coupling network. The evaluation module is used to perform a comprehensive dynamic evaluation of the low-altitude flight management system based on the indicator data and the capability coupling network, and obtain dynamic evaluation results. The identification module is used to perform bottleneck identification processing on the low-altitude flight control system based on the dynamic evaluation results and the capability coupling network, and obtain bottleneck identification results, wherein the bottleneck identification results represent the nodes that restrict the capabilities of the low-altitude flight control system.

Images