A test system configuration change impact analysis method and system

Abstract

The application provides a test system configuration change influence analysis method and system, and belongs to the technical field of aerospace test. The method is based on configuration element data to construct an aerospace test system knowledge graph. A weighted multi-hop influence propagation algorithm is used to traverse the aerospace test system knowledge graph to realize configuration element change, determine influence nodes and influence strength of the influence nodes, determine configuration change influence scores based on the influence strength, and determine system-level influence scores in combination with safety criticality determination results. Configuration change influence analysis of the aerospace test system is performed according to the system-level influence scores, a complete test system configuration change influence analysis process is formed, and change propagation links are determined in combination with the aerospace test system knowledge graph and the weighted multi-hop influence propagation algorithm, so that standardized and automatic analysis of configuration element change risks is realized.

Description

Technical Field

[0001] This invention belongs to the field of aerospace test technology, and relates to the change impact analysis technology of configuration management in aerospace test systems, specifically a method and system for analyzing the impact of configuration changes in test systems. Background Technology

[0002] Aerospace test systems refer to comprehensive test platforms and technical systems built to verify the performance, lifespan, and reliability of aircraft (including airplanes, rockets, and spacecraft) and their subsystems and components. Aerospace products are characterized by high technological integration, long development cycles, and huge testing costs. Any design flaw can lead to serious consequences. Therefore, aerospace testing is carried out throughout the entire life cycle of a product, from design and manufacturing to delivery.

[0003] Aerospace test systems consist of various configuration elements, primarily including: test specimens (objects under test), test program sets, excitation / measurement instruments, signal conditioning units, interface adapters, simulation models, calibration certificates, and test procedures. With the continuous expansion of aerospace test systems, complex coupling relationships have formed among test specimens, test programs, instruments, and calibration certificates, involving version dependencies, interface references, and function triggers. During the test cycle, any change to any configuration element—such as instrument firmware upgrades, test program version updates, interface definition revisions, or calibration certificate renewals—can affect the validity of other elements, potentially rendering test conclusions invalid or requiring re-verification. Therefore, effectively managing these configuration elements and their impact has become a core challenge in aerospace test system configuration management.

[0004] In existing technologies, configuration management of aerospace test systems mainly relies on human intervention. Since the dependencies between configuration elements are mostly scattered in unstructured documents (documents without fixed, standard data formats), there is a lack of formal descriptions that can be directly parsed and reasoned by computers. As a result, changes to configuration elements heavily rely on human experience, and the propagation links of configuration element changes are not visible. Once configuration elements undergo large-scale changes, the risks caused by these changes are difficult to standardize and automate. Summary of the Invention

[0005] As described in the background section, in existing aerospace test systems, once the configuration elements are changed, the risk caused by the change process heavily relies on human experience and the propagation link of the configuration element change is not visible, resulting in the difficulty of standardizing and automating the analysis.

[0006] To address the aforementioned technical problems, this invention proposes a method and system for analyzing the impact of changes in the configuration of an experimental system, as detailed below: A method for analyzing the impact of configuration changes on an experimental system includes the following steps: Collect configuration element data of aerospace test systems; Using the principles of knowledge graph construction, a knowledge graph for aerospace test systems is constructed based on the configuration element data. Monitor the configuration element change event stream of the aerospace test system. When a configuration element change request signal is detected, extract the change description tuple from the configuration element change request signal. Determine the change intensity coefficient based on the change description tuple and use the change intensity coefficient as the initial influence intensity of the change source node. Using a weighted multi-hop influence propagation algorithm, starting from the change source node, the knowledge graph of the aerospace test system is traversed to change the configuration elements, and the set of affected nodes affected by the configuration element changes and the influence intensity of each affected node are determined. The configuration change impact score of each affected node is calculated based on the impact intensity of each affected node, and the system-level impact score is determined based on the configuration change impact score of each affected node and the safety criticality determination result of each affected node. Analysis of the impact of configuration changes on aerospace test systems based on system-level impact scores.

[0007] Further specifying, the configuration element data includes a set of nodes, a set of relation types, and a set of attributes; The node set includes test piece nodes, test program set nodes, instrument and equipment nodes, signal parameter nodes, interface adapter nodes, calibration certificate nodes, test item nodes, and simulation model nodes; The set of relationship types includes dependency relationships, reference relationships, triggering relationships, calibration relationships, adaptation relationships, and constraint relationships; The attribute set includes version identification attributes and security-critical attributes.

[0008] Further specifying, the construction of an aerospace test system knowledge graph based on the configuration element data using knowledge graph construction principles specifically includes: By utilizing the principles of knowledge graph construction, graph nodes are constructed based on node sets, directed edges are constructed based on relation type sets, and edge weights are determined based on attribute sets, thus forming a knowledge graph for aerospace test systems composed of graph nodes, directed edges, and edge weights.

[0009] Further defined, the change description tuple includes the change source node, change type, attribute difference set, version identifier before change, version identifier after change, and / or timestamp; The change types include version updates, modifications, replacements, deletions, or expiration.

[0010] Furthermore, the change intensity coefficient is obtained by normalizing the attribute differences in the attribute difference set.

[0011] Furthermore, the attenuation factor in the weighted multi-hop influence propagation algorithm is an adaptive attenuation factor, which is determined based on the out-degree of the change source node in the aerospace test system knowledge graph; The propagation threshold in the weighted multi-hop influence propagation algorithm is adaptively determined based on Otsu's method.

[0012] Furthermore, the safety criticality determination result is determined in the following manner: Determine if the source node of the change meets any of the following conditions: ① The test subject number corresponding to the source node is a safety clause in the ATA standard or GJB standard; ② The source node of the change is in the top 5% in terms of betweenness centrality in the aerospace test system knowledge graph; If so, the safety criticality determination result is assigned a value of 1; If not, the safety criticality determination result is assigned a value of 0.

[0013] Further specifying, the analysis of the impact of configuration changes on aerospace test systems based on system-level impact scores specifically includes: Determining cluster boundary values ​​using the K-means clustering algorithm , and ; like If so, there is no impact; like Then there is a general impact; like This would have a significant impact; like This would have a significant impact; This refers to the system-level impact score.

[0014] Further specifying, the method of determining cluster boundary values ​​using the K-means clustering algorithm... , and Specifically, it includes: Obtain the set of historical change events of the aerospace test system within a set time window, calculate the system-level impact score of each set of historical change events, and obtain the score sample set; The K-means clustering algorithm was applied to the set of rating samples, with a cluster size of 4. Using Euclidean distance as the metric, four cluster centers were obtained after iterative convergence, designated as the first cluster center. Second cluster center Third cluster center and the fourth cluster center ; Cluster boundary values ​​were determined based on the four cluster centers. , and Among them, cluster boundary values First cluster center Second cluster center The average value, cluster boundary value As the second cluster center and the third cluster center The average value, cluster boundary value The third cluster center and the fourth cluster center The average value.

[0015] The experimental system configuration change impact analysis system corresponding to the above-mentioned experimental system configuration change impact analysis method includes: Acquisition module: Used to acquire configuration element data of aerospace test systems; Knowledge graph construction module: used to construct a knowledge graph of aerospace test system based on the configuration element data using the principles of knowledge graph construction; The monitoring module is used to monitor the configuration element change event stream of the aerospace test system. When a configuration element change request signal is detected, the module extracts the change description tuple from the configuration element change request signal; determines the change intensity coefficient based on the change description tuple, and uses the change intensity coefficient as the initial influence intensity of the change source node. Configuration element change module: It is used to use a weighted multi-hop influence propagation algorithm to traverse the knowledge graph of the aerospace test system to change the configuration elements, starting from the change source node, and to determine the set of affected nodes and the influence intensity of each affected node due to the configuration element change. The scoring determination module is used to calculate the configuration change impact score of each affected node based on the impact intensity of each affected node, and to determine the system-level impact score based on the configuration change impact score of each affected node and the safety criticality determination result of each affected node. And the analysis module: used to analyze the impact of configuration changes on aerospace test systems based on system-level impact scores.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1) The present invention provides a method for analyzing the impact of configuration changes on a test system. First, a knowledge graph of the aerospace test system is constructed based on configuration element data. Second, the change intensity coefficient determined by the change description tuple is used as the initial influence intensity of the change source node. A weighted multi-hop influence propagation algorithm is used to traverse the aerospace test system knowledge graph to realize configuration element changes and determine the influencing nodes and their influence intensity. Next, the configuration change impact score is determined based on the influence intensity, and a system-level impact score is determined in conjunction with the safety criticality assessment results. Finally, the configuration change impact analysis of the aerospace test system is performed based on the system-level impact score, forming a complete process for analyzing the impact of configuration changes on the test system. Furthermore, this invention, by combining the aerospace test system knowledge graph and the weighted multi-hop influence propagation algorithm, clarifies the change propagation chain, achieving standardized and automated analysis of configuration element change risks. This method effectively replaces the traditional process of relying on human experience for configuration change impact analysis, avoiding errors inherent in human analysis and improving the accuracy of the analysis.

[0017] 2) In this invention, the change intensity coefficient is determined after normalizing the attribute differences in the attribute difference set, thereby quantifying the attribute differences in the attribute difference set, that is, quantifying the impact of changes in configuration elements on other configuration elements, which facilitates the quantitative analysis and assessment of the risk of changes in configuration elements.

[0018] 3) In this invention, the adaptive attenuation factor is determined based on the out-degree of the source node in the aerospace test system knowledge graph. This transforms the fixed attenuation factor of the traditional weighted multi-hop influence propagation algorithm into a dynamic attenuation factor that is adaptively adjusted based on the source node. This disperses the influence of nodes with larger out-degrees to downstream nodes, avoiding over-propagation by nodes with larger out-degrees and ensuring that the propagation intensity matches the actual physical / logical conditions.

[0019] 4) In this invention, the propagation threshold in the weighted multi-hop influence propagation algorithm is adaptively determined based on Otsu's method to adapt to the scale of the knowledge graph of the aerospace test system and the density of the edges. The propagation termination condition is changed from being driven by human experience to being driven by data distribution. Regardless of whether the knowledge graph of the aerospace test system is sparse or dense, it can find the optimal cutoff point on its own, thus ensuring the generalization and objectivity of the weighted multi-hop influence propagation algorithm in the analysis of the impact of configuration changes in the test system.

[0020] 5) In this invention, the K-means clustering algorithm is used to determine the cluster boundary values. , and This allows for the objective division of abstract, continuously changing system-level impact scores into four risk levels, and the automatic generation of hierarchical boundaries for decision-making. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the method for analyzing the impact of configuration changes on the experimental system of the present invention; Figure 2 This is a schematic diagram of the experimental system configuration change impact analysis system of the present invention; Figure 3 This is a schematic diagram of a knowledge graph for aerospace testing systems. Detailed Implementation

[0022] The technical solution of the present invention will be further explained and described below with reference to the accompanying drawings and embodiments, but the present invention is not limited to the embodiments described below.

[0023] See Figure 1 This invention proposes a method for analyzing the impact of configuration changes on experimental systems, comprising the following steps: S1: Collect configuration element data of aerospace test systems. Configuration element data refers to objects in aerospace test systems that participate in configuration management and have version dependencies or functional coupling relationships with each other.

[0024] Configuration element data includes a set of nodes, a set of relation types, and a set of attributes; the set of nodes includes test specimen nodes, test program set nodes, instrument and equipment nodes, signal parameter nodes, interface adapter nodes, calibration certificate nodes, test item nodes, and simulation model nodes.

[0025] The attributes of the test piece node include test piece number, version number, and serial number. The attributes of the test program set node include program name, version identifier, and applicable UUT list (Units Under Test). The attributes of the instrument / equipment node include instrument model, firmware version, metrological validity period, and accuracy class. The attributes of the signal parameter node include signal name, range, accuracy, sampling rate, and modulation method. The attributes of the interface adapter node include interface ICD version, connector type, and electrical characteristics. The attributes of the calibration certificate node include certificate number, calibration date, validity period, and referenced instrument. The attributes of the test item node include test subject number (refer to ATA section or GJB standard clause), decision threshold, and referenced signal list. The attributes of the simulation model node include model name, version, and applicable operating condition range. Here, ATA section refers to the ATA 2200 specification, a globally accepted technical classification standard in the aviation industry. GJB standard refers to GJB 900A-2012 "General Requirements for Equipment Safety Operations". The set of relationship types includes dependency relationships, reference relationships, trigger relationships, calibration relationships, adaptation relationships, and constraint relationships. Dependency relationships represent version or functional dependencies between configuration elements. Reference relationships represent interface or document reference relationships between configuration elements. Trigger relationships represent functional triggering relationships between configuration elements. Calibration relationships represent calibration tracing relationships between configuration elements. Adaptation relationships represent interface adaptation relationships between configuration elements. Constraint relationships represent relationships between constraints of configuration elements.

[0026] The attribute set includes version identification attributes and security criticality attributes, that is, the version identification attributes and security criticality attributes of each configuration element.

[0027] S2: Utilizing the principles of knowledge graph construction, construct a knowledge graph for aerospace test systems based on configuration element data; See Figure 3 S2 specifically includes: utilizing the principles of knowledge graph construction, constructing graph nodes based on a node set, that is, treating each configuration element node in the node set as a graph node, representing the node set as V, and using V as the name of each configuration element node (graph node). express, An instance of a configuration element, carrying an attribute vector. and version identifier Directed edges are constructed based on the set of relation types, that is, the relationships between nodes in the node set are used as directed edges; edge weights are determined based on the attribute set, and the edge weights represent the association relationships of quantifiable parameters in the attribute set; thus forming a knowledge graph of aerospace test systems composed of graph nodes, directed edges, and edge weights.

[0028] Each directed edge is used The term "symbol" refers to a graph node. and graph nodes Existence type is Dependency relationships, It refers to a set of relation types.

[0029] Edge weights This indicates that the overlap is calculated using the Jaccard similarity coefficient, and the formula is as follows: In the formula, For graph nodes and graph nodes The edge weights between them are dimensionless. For graph nodes In attribute vector The set of all quantifiable parameters in a given context, dimensionless; For graph nodes In attribute vector The set of all quantifiable parameters in the formula is dimensionless. Representing the graph nodes and graph nodes The number of shared parameters; Representing the graph nodes and graph nodes The total number of all different parameters involved.

[0030] For different relation types, the selection rules for all quantifiable parameters in the attribute vector of a graph node are as follows: When the relationship type is a reference relationship, the set of names is retrieved from the attribute of the signal parameter node. When the relationship type is calibration relationship, the set of features representing the intersection of calibration parameters and calibrated parameters in the calibration certificate node attributes is taken; When the relationship type is a trigger relationship, the parameter set of the trigger conditions in each graph node is retrieved; When the relationship type is an adapter relationship, retrieve the set of pin attributes or signal attributes from the interface adapter node; When the relationship type is dependency, take the set of shared functional parameters in the functional dependency chain of each graph node; When the relationship type is a constraint relationship, the set of parameters involved in the constraint conditions of each graph node is taken.

[0031] The completed aerospace test system knowledge graph is stored in a graph database, supporting incremental updates. When adding or modifying configuration elements, only the local graph nodes affected by the update are changed, without needing to change all of them.

[0032] S3: Monitor the configuration element change event stream of the aerospace test system. When a configuration element change request signal is detected, extract the change description tuple from the configuration element change request signal; determine the change intensity coefficient based on the change description tuple, and use the change intensity coefficient as the initial influence strength of the change source node; wherein, the change description tuple includes the change source node, change type, attribute difference set, pre-change version identifier, post-change version identifier and / or timestamp. Change type includes version update, modification, replacement, deletion or expiration. The attribute difference set is used to reflect the specific differences of attributes before and after the change. For each attribute change (each attribute difference) in the attribute difference set, normalization is performed to obtain the change intensity coefficient. , This is used to quantify the potential impact of a change in a configuration element on other configuration elements. The greater the impact intensity, the more severe the impact of the change on other configuration elements. Its initial value is determined by the initial impact intensity corresponding to the change source node.

[0033] S4: Using a weighted multi-hop influence propagation algorithm, starting from the source node of the change, the knowledge graph of the aerospace test system is traversed to change the configuration elements, and the set of affected nodes affected by the configuration element change and the influence intensity of each affected node are determined.

[0034] In a preferred embodiment of the present invention, the attenuation factor in the weighted multi-hop influence propagation algorithm is an adaptive attenuation factor, which is determined based on the out-degree of the source node in the aerospace test system knowledge graph; the propagation threshold in the weighted multi-hop influence propagation algorithm is adaptively determined based on Otsu's method.

[0035] The specific process of traversing the knowledge graph of aerospace test systems to change configuration elements is as follows: ① Initialize the impact propagation queue Q, and define each configuration element change as a quadruple ( ), To change the graph nodes after k hops from the source node, For graph nodes The intensity of the impact, For graph nodes The number of propagation hops, To change the source node to the graph node The path of its influence.

[0036] ② Take out the quadruple from the influence propagation queue Q ( ), for graph nodes For each directed edge, make the following judgment: Blocking condition judgment: If there is a blocking rule for the relation type in the edge relation type set τ, then skip the directed edge; otherwise, continue propagation.

[0037] Topology adaptive decay calculation: computation of graph nodes The intensity of the influence at the location, the intensity of the influence hop count decay factor From graph nodes The out-degree is determined, where the out-degree refers to the number of directed edges originating from this node and pointing to other nodes. For graph nodes and graph nodes edge weights, For graph nodes The intensity of the impact.

[0038] ③ The results of each calculation After adding samples to set Ψ, or after the first traversal is completed or when the number of samples in Ψ reaches the preset minimum statistical sample size, the propagation threshold is calculated using Otsu's method. Propagation threshold The calculation formula is: In the formula, The propagation threshold is dimensionless. and These represent the proportions of samples with influence intensity less than θ and influence intensity greater than θ, respectively, and are dimensionless. and These are the sample mean values ​​for influence strength less than θ and the sample mean values ​​for influence strength greater than θ, respectively, and are dimensionless. like If so, then propagation along that directed edge will stop; If the graph node If a node is not in the set of affected nodes Φ, then the graph node will be... Add to the set of affected nodes Φ; ③ Repeat step ② above until the set Ψ in step ② is empty, and obtain the set of affected nodes Φ and the influence strength of each affected node.

[0039] S5: Calculate the configuration change impact score of each affected node based on the impact intensity of each affected node, and determine the system-level impact score based on the configuration change impact score of each affected node and the safety criticality determination result of each affected node; specifically, the system-level impact score is the product of the configuration change impact score of each affected node and the safety criticality determination result of each affected node.

[0040] Among them, the configuration change impact score Determine according to the following formula: In the formula, For graph nodes Configurational changes affect the score, dimensionless; For graph nodes The intensity of the influence (of the affected node) is dimensionless; For graph nodes The node importance weight (affecting nodes) is dimensionless; This is a change type coefficient, dimensionless.

[0041] In the above formula, The results were obtained based on the betweenness centrality of the aerospace test system knowledge graph: In the formula, For graph nodes The betweenness centrality (of the affected nodes) is dimensionless; For graph nodes To the graph node The number of shortest paths; For graph nodes To the graph node The number of shortest paths passing through graph nodes Quantity; , and All of these are affected nodes.

[0042] Change type coefficient Determined quantitatively based on the actual magnitude of changes in configuration elements: When the change type is version update, Determine according to the following formula: In the formula, For graph nodes Version distance before and after configuration element change, dimensionless, refers to the numerical span or degree of difference of the configuration element in the version number system before and after the configuration element change; The maximum version distance is dimensionless.

[0043] When the change type is modification Determine according to the following formula: In the formula, For graph nodes The ratio of parameter deviations before and after a change in configuration elements is dimensionless. It refers to the magnitude of the change in the value of a specific technical parameter (such as range, accuracy, sampling rate) before and after a change in configuration elements, and the degree of deviation from the rated or standard value of that parameter.

[0044] When the change type is replacement, the change type coefficient is equal to the number of affected interfaces or parameters divided by the total number of interfaces or parameters. When the change type is expiration, the change type coefficient is equal to the time elapsed since the expiration date divided by the calibration cycle or verification cycle; if the elapsed time is greater than the calibration cycle or verification cycle, the change type coefficient is 1. When the change type is deletion, the change type coefficient is 1.

[0045] The safety criticality determination result is determined in the following manner: Determine if the source node of the change meets any of the following conditions: ① The test subject number corresponding to the changed source node belongs to the safety clauses in the ATA standard or GJB standard; among them, the safety clauses in the ATA standard refer to ATA 26-fire protection system, ATA 27-flight control, ATA 32-landing gear and ATA 34-navigation; the safety clauses in the GJB standard refer to the criteria and standard clauses in GJB 900A-2012 "General Requirements for Equipment Safety Work" used to identify and determine "safety critical items"; ② The source node of the change is in the top 5% in terms of betweenness centrality in the aerospace test system knowledge graph; If so, the safety criticality determination result is assigned a value of 1; If not, the safety criticality determination result is assigned a value of 0.

[0046] S6: Analyze the impact of configuration changes on aerospace test systems based on system-level impact scores.

[0047] The specific aspects of analyzing the impact of configuration changes on aerospace test systems based on system-level impact scores include: Determining cluster boundary values ​​using the K-means clustering algorithm , and ; like If so, there will be no impact; the records will be automatically recorded and archived without approval or human intervention. like If so, there will be a general impact, requiring human intervention to re-analyze the impact of configuration changes; like If so, it will have a significant impact and requires re-approval and human intervention to re-analyze the impact of the configuration change; like If this happens, it will have a significant impact, triggering an emergency change review, generating a retest list, and suspending related testing activities. This refers to the system-level impact score.

[0048] The above uses the K-means clustering algorithm to determine the cluster boundary values. , and Specifically, it includes: The system acquires a set of historical change events of the aerospace test system within a set time window, calculates the system-level impact score of each set of historical change events, and obtains a score sample set. The set time window can be 30 minutes, 1 hour, 3 hours, 8 hours, or 24 hours, and the specific value can be set by the user.

[0049] The K-means clustering algorithm was applied to the rating sample set, with a cluster size of 4. Using Euclidean distance as the metric, four cluster centers were obtained after iterative convergence, designated as the first cluster center. Second cluster center Third cluster center and the fourth cluster center .

[0050] Cluster boundary values ​​were determined based on the four cluster centers. , and Among them, cluster boundary values First cluster center Second cluster center The average value, cluster boundary value As the second cluster center and the third cluster center The average value, cluster boundary value The third cluster center and the fourth cluster center The average value.

[0051] This invention discloses a method for analyzing the impact of configuration changes on a test system. First, a knowledge graph of the aerospace test system is constructed based on configuration element data. Second, the change intensity coefficient determined by the change description tuple is used as the initial influence intensity of the change source node. A weighted multi-hop influence propagation algorithm is used to traverse the aerospace test system knowledge graph to realize configuration element changes and determine the influencing nodes and their influence intensity. Next, the configuration change impact score is determined based on the influence intensity, and a system-level impact score is determined by combining the safety criticality assessment results. Finally, the configuration change impact analysis of the aerospace test system is performed based on the system-level impact score, forming a complete process for analyzing the impact of configuration changes on the test system. Furthermore, this invention, by combining the aerospace test system knowledge graph and the weighted multi-hop influence propagation algorithm, clarifies the change propagation chain, achieving standardized and automated analysis of configuration element change risks. This method effectively replaces the traditional human-based process of analyzing the impact of configuration changes, avoiding errors inherent in human analysis and improving the accuracy of the analysis.

[0052] See Figure 2 The present invention also proposes an experimental system configuration change impact analysis system, which corresponds to the above-mentioned experimental system configuration change impact analysis method, including a data acquisition module, a knowledge graph construction module, a monitoring module, a configuration element change module, a scoring determination module, and an analysis module; Acquisition module: Used to acquire configuration element data of aerospace test systems; Knowledge graph construction module: used to construct a knowledge graph of aerospace test system based on the configuration element data using the principles of knowledge graph construction; The monitoring module is used to monitor the configuration element change event stream of the aerospace test system. When a configuration element change request signal is detected, the module extracts the change description tuple from the configuration element change request signal; determines the change intensity coefficient based on the change description tuple, and uses the change intensity coefficient as the initial influence intensity of the change source node. Configuration element change module: It is used to use the weighted multi-hop influence propagation algorithm to traverse the aerospace test system knowledge graph to change configuration elements, starting from the change source node, and to determine the set of affected nodes affected by the configuration element change and the influence intensity of each affected node. The scoring determination module is used to calculate the configuration change impact score of each affected node based on the impact intensity of each affected node, and to determine the system-level impact score based on the configuration change impact score of each affected node and the safety criticality determination result of each affected node. Analysis module: Used to analyze the impact of configuration changes on aerospace test systems based on system-level impact scores.

[0053] It should be noted that the experimental system configuration change impact analysis system of the present invention is completely corresponding to the experimental system configuration change impact analysis method described above. For the relevant contents of the acquisition module, knowledge graph construction module, monitoring module, configuration element change module, scoring determination module and analysis module that are not described in detail in the experimental system configuration change impact analysis system of the present invention, please refer to the description of the experimental system configuration change impact analysis method section above.

[0054] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A method for analyzing the impact of configuration changes on an experimental system, characterized in that, Includes the following steps: Collect configuration element data of aerospace test systems; Using the principles of knowledge graph construction, a knowledge graph for aerospace test systems is constructed based on the configuration element data. Monitor the configuration element change event stream of the aerospace test system. When a configuration element change request signal is detected, extract the change description tuple from the configuration element change request signal. Determine the change intensity coefficient based on the change description tuple and use the change intensity coefficient as the initial influence intensity of the change source node. Using a weighted multi-hop influence propagation algorithm, starting from the change source node, the knowledge graph of the aerospace test system is traversed to change the configuration elements, and the set of affected nodes affected by the configuration element changes and the influence intensity of each affected node are determined. The configuration change impact score of each affected node is calculated based on the impact intensity of each affected node, and the system-level impact score is determined based on the configuration change impact score of each affected node and the safety criticality determination result of each affected node. Analysis of the impact of configuration changes on aerospace test systems based on system-level impact scores.

2. The method for analyzing the impact of changes in the configuration of an experimental system according to claim 1, characterized in that, The configuration element data includes a set of nodes, a set of relation types, and a set of attributes; The node set includes test piece nodes, test program set nodes, instrument and equipment nodes, signal parameter nodes, interface adapter nodes, calibration certificate nodes, test item nodes, and simulation model nodes; The set of relationship types includes dependency relationships, reference relationships, triggering relationships, calibration relationships, adaptation relationships, and constraint relationships; The attribute set includes version identification attributes and security-critical attributes.

3. The method for analyzing the impact of changes in the configuration of an experimental system according to claim 2, characterized in that, The construction of a knowledge graph for aerospace test systems based on the configuration element data using knowledge graph construction principles specifically includes: By utilizing the principles of knowledge graph construction, graph nodes are constructed based on node sets, directed edges are constructed based on relation type sets, and edge weights are determined based on attribute sets, thus forming a knowledge graph for aerospace test systems composed of graph nodes, directed edges, and edge weights.

4. The method for analyzing the impact of changes in the configuration of the test system according to claim 3, characterized in that, The change description tuple includes the change source node, change type, attribute difference set, version identifier before change, version identifier after change, and / or timestamp; The change types include version updates, modifications, replacements, deletions, or expiration.

5. The method for analyzing the impact of changes in the configuration of an experimental system according to claim 4, characterized in that, The change intensity coefficient is obtained by normalizing the attribute differences in the attribute difference set.

6. The method for analyzing the impact of configuration changes on experimental systems according to claim 4, characterized in that, The attenuation factor in the weighted multi-hop influence propagation algorithm is an adaptive attenuation factor, which is determined based on the out-degree of the change source node in the aerospace test system knowledge graph. The propagation threshold in the weighted multi-hop influence propagation algorithm is adaptively determined based on Otsu's method.

7. The method for analyzing the impact of changes in the configuration of an experimental system according to claim 2, characterized in that, The safety criticality determination result is determined in the following manner: Determine if the source node of the change meets any of the following conditions: ① The test subject number corresponding to the source node is a safety clause in the ATA standard or GJB standard; ② The source node of the change is in the top 5% in terms of betweenness centrality in the aerospace test system knowledge graph; If so, the safety criticality determination result is assigned a value of 1; If not, the safety criticality determination result is assigned a value of 0.

8. The method for analyzing the impact of configuration changes on experimental systems according to claim 2, characterized in that, The analysis of the impact of configuration changes on aerospace test systems based on system-level impact scores specifically includes: Determining cluster boundary values ​​using the K-means clustering algorithm , and ; like If so, there is no impact; like Then there is a general impact; like This would have a significant impact; like This would have a significant impact; This refers to the system-level impact score.

9. The method for analyzing the impact of configuration changes on experimental systems according to claim 8, characterized in that, The K-means clustering algorithm is used to determine the cluster boundary values. , and Specifically, it includes: Obtain the set of historical change events of the aerospace test system within a set time window, calculate the system-level impact score of each set of historical change events, and obtain the score sample set; The K-means clustering algorithm was applied to the set of rating samples, with a cluster size of 4. Using Euclidean distance as the metric, four cluster centers were obtained after iterative convergence, designated as the first cluster center. Second cluster center Third cluster center and the fourth cluster center ; Cluster boundary values ​​were determined based on the four cluster centers. , and Among them, cluster boundary values First cluster center Second cluster center The average value, cluster boundary value As the second cluster center and the third cluster center The average value, cluster boundary value The third cluster center and the fourth cluster center The average value.

10. A test system configuration change impact analysis system corresponding to the test system configuration change impact analysis method according to claim 1, characterized in that, include: Acquisition module: Used to acquire configuration element data of aerospace test systems; Knowledge graph construction module: used to construct a knowledge graph of aerospace test system based on the configuration element data using the principles of knowledge graph construction; The monitoring module is used to monitor the configuration element change event stream of the aerospace test system. When a configuration element change request signal is detected, the module extracts the change description tuple from the configuration element change request signal; determines the change intensity coefficient based on the change description tuple, and uses the change intensity coefficient as the initial influence intensity of the change source node. Configuration element change module: It is used to use a weighted multi-hop influence propagation algorithm to traverse the knowledge graph of the aerospace test system to change the configuration elements, starting from the change source node, and to determine the set of affected nodes affected by the configuration element change and the influence intensity of each affected node. The scoring determination module is used to calculate the configuration change impact score of each affected node based on the impact intensity of each affected node, and to determine the system-level impact score based on the configuration change impact score of each affected node and the safety criticality determination result of each affected node. And the analysis module: used to analyze the impact of configuration changes on aerospace test systems based on system-level impact scores.

Images