An airborne hydrogen system safety analysis method, system, device and medium

Abstract

The application discloses an airborne hydrogen system safety analysis method, system, equipment and medium, and relates to the technical field of aviation safety, and comprises the following steps: a function resonance confrontation hierarchical model is established; the function resonance influence of a subcomponent is obtained by dividing the overall system through function resonance analysis method of the function feature description of the airborne hydrogen system, and a dynamic interaction network of the function module of the subcomponent is drawn; the dynamic interaction network is simplified by dimension reduction, and a reachable matrix is obtained; the obtained data is subjected to confrontation layer extraction based on the reachable matrix, and overall division of the resonance model is obtained; hierarchical importance and structural importance are obtained through topological structure data analysis, and safety analysis of the airborne hydrogen system is carried out. The application establishes the function resonance confrontation hierarchical model, and can solve a series of technical problems such as that the analysis method of the prior art airborne hydrogen system is greatly influenced by the subjectivity of researchers, and hierarchical blurring occurs in the internal hierarchical division.

Description

A method, system, equipment and medium for safety analysis of airborne hydrogen systems Technical Field

[0001] This invention relates to the field of aviation safety technology, and in particular to a method, system, equipment and medium for safety analysis of airborne hydrogen systems. Background Technology

[0002] In the civil aviation sector, safety analysis is an essential part of ensuring the safe and stable operation of systems. The security of these systems faces challenges such as leaks, sensor failures, and insufficient redundancy.

[0003] The dynamic functional interactions of a system are prone to triggering cascading resonance failures, and existing models cannot effectively quantify the priority of their impact. Furthermore, traditional security analysis methods (such as the ISM model) rely on static hierarchical divisions, making it difficult to capture the dynamic resonance risks of complex systems. Moreover, they lack multi-perspective adversarial verification, which can easily lead to one-sided analysis results.

[0004] In existing technologies, the coupled use of functional resonance analysis and interpretive structural models can, to some extent, solve the shortcomings of traditional security analysis methods that rely on expert judgment, are highly subjective, and ignore the differences in the strength of relationships between elements. Furthermore, "dynamic functional analysis" can compensate for the traditional methods' neglect of dynamic functional interaction and cascading resonance failure. However, due to the monotony of layer extraction, hierarchical division can lead to hierarchical ambiguity errors within the hierarchical division, resulting in unreliable security analysis results. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of the prior art by providing a method, system, equipment, and medium for safety analysis of airborne hydrogen systems, thereby solving the problems in the prior art.

[0006] The present invention specifically provides the following technical solution:

[0007] A method for safety analysis of an airborne hydrogen system includes:

[0008] This paper analyzes the interrelationships of sub-components in an airborne hydrogen system from the perspective of functional resonance. By hierarchical division under an adversarial network structure, a functional resonance diagram corresponding to the interrelationships of sub-components is derived. Through topological data analysis of the functional resonance diagram, the influence of sub-components on the whole, key risk nodes, and optimized paths for protective measures are obtained, and a hierarchical model of functional resonance adversarial measures is established. The key risk nodes are the links in the overall process that play a decisive role in safety.

[0009] The airborne hydrogen system is divided by system functional feature description, the functions of each sub-component are extracted, and the functional relationships are coupled to obtain the dynamic interaction network of sub-component functions; the system functional feature description is a conditional feature that represents the system functions during startup, operation, and execution.

[0010] Based on the directed graph of ordered triples, the dynamic interaction network is dimensionality reduced to obtain the adjacency matrix generated by the relationship between sub-components. The adjacency matrix is ​​then repeatedly processed by the transition matrix to obtain the reachability matrix of the connection paths between sub-components.

[0011] Adversarial layer extraction is performed on the reachability matrix to obtain the overall partitioning of the functional resonance adversarial hierarchy model. Topological data analysis is then performed on the overall partitioning results to obtain the hierarchy importance and structural importance. The safety analysis of the airborne hydrogen system is then conducted based on the hierarchy importance and structural importance.

[0012] Preferably, the process of dividing the airborne hydrogen system through system functional feature description, extracting the functions of each sub-component, and coupling functional relationships to obtain a dynamic interaction network of sub-component functions specifically involves:

[0013] The airborne hydrogen system is divided into sub-components including electronic / electrical systems, hydrogen storage systems, booster pump pressurization systems, hydrogen supply pipeline systems, hydrogen internal combustion engine combustion systems, cooling systems, and sensing systems.

[0014] The functions of sub-components are defined by the input, output, time, prerequisite, resource, and control dimensions in the system functional characteristic description, and the relationships between functions are obtained to obtain the dynamic interaction network of sub-component functions.

[0015] Preferably, the step of using a directed graph of ordered triples as a basis to reduce the dimensionality of the dynamic interaction network and obtain the adjacency matrix generated by the relationships between sub-components is as follows:

[0016] Input and output analysis is performed on dynamic interaction networks, where the input of the dynamic interaction network is the thing that initiates the function or the thing that the function will process or transform, and the output, time, premise, resources, and control of the dynamic interaction network are all points to the starting point of the connection when the function is running.

[0017] Through the data in the matrix This represents the relationship between node i and node j in a directed graph; where each node represents the function of a sub-component.

[0018] Based on the aforementioned relationships, the interrelationships of the sub-components of the airborne hydrogen system are digitized from the directed graph, and an adjacency matrix is ​​constructed using the matrix data generated from the digitized interrelationships of the sub-components.

[0019] Preferably, the relationship includes: =0 indicates that node i has no direct influence on node j, that is, there is no path of length 1 from node i to node j, and there is no influence between the functions; =1 indicates that node i has a direct influence on node j, meaning there exists a path of length 1 that allows node i to directly reach node j, indicating a relationship between their functions.

[0020] Preferably, the adversarial layer removal of the reachability matrix specifically includes:

[0021] The factors influencing a function are hierarchically divided using the reachability matrix, and the reachability set and antecedent set of each sub-component are obtained. The common set is determined based on the reachability set and the antecedent set. The antecedent set refers to the set of all upstream factors that cause a functional component to fail, the reachability set refers to the set of all cascading failure states that can be triggered by the system coupling relationship starting from the failure of a component, and the common set refers to the intersection of the reachability set and the antecedent set.

[0022] The first layer of extraction is performed on the antecedent set, reachable set, and common set through adversarial layer extraction. After the first layer extraction is completed, the rows and columns of the adjacent Boolean matrices of the extracted sub-components are deleted to obtain the updated matrix. The extraction is then iterated for each layer to complete the adversarial layer extraction.

[0023] Preferably, the anti-delamination method specifically includes:

[0024] The methods for combating layer extraction include: result-priority UP-type extraction and DOWN-type topology layer extraction; wherein, result-priority UP-type extraction means that the sub-component is the highest priority sub-component during layer extraction, and DOWN-type topology layer extraction is performed according to the principle of cause priority, and the sub-component is the highest priority sub-component during layer extraction.

[0025] Preferably, the step of performing topological structure data analysis on the overall partitioning results to obtain hierarchical importance and structural importance specifically involves:

[0026] The overall system is divided into n layers by adversarial extraction, with each layer represented by L.

[0027] The specific expression for obtaining the weights of each level in the overall system is as follows:

[0028] ;

[0029] Among them, the weights M of each level are... L Indicates the importance of the hierarchy;

[0030] Functions that point to the function to be computed are defined as in-degree functions, and functions that the function to which the function to be computed points are defined as out-degree functions. Based on the weights of each level, the structural importance of functional modules at each level is obtained, as expressed in the following expression:

[0031] ;

[0032] in, Indicates the structural importance of function p. This represents the weight of the layer where the function p to be calculated resides. This indicates the weight of the level at which function q, which points to function p, resides. This represents the weight of the level at which function k, which is pointed to by function p. O is the out-degree coefficient, and I is the in-degree coefficient.

[0033] This invention provides an airborne hydrogen system safety analysis system, comprising:

[0034] The model building module is used to analyze the mutual influence relationships of sub-components of the airborne hydrogen system from the perspective of functional resonance. It derives the functional resonance diagram corresponding to the mutual influence relationships of sub-components through hierarchical division under the adversarial network structure. Through topological data analysis of the functional resonance diagram, it obtains the influence relationship of sub-components on the whole, key risk nodes, and optimized paths of protective measures, and establishes a functional resonance adversarial hierarchical model. The key risk nodes are the links in the overall process that play a decisive role in safety.

[0035] The system partitioning module is used to partition the airborne hydrogen system through system functional feature description, extract the functions of each sub-component, and couple the functional relationships to obtain the dynamic interaction network of sub-component functions; the system functional feature description is a conditional feature that represents the system functions during startup, operation, and execution.

[0036] The image matrix processing module is used to reduce the dimensionality of the dynamic interactive network based on the directed graph of ordered triples, obtain the adjacency matrix generated by the relationship between sub-components, and repeatedly perform transition matrix processing on the adjacency matrix to obtain the reachability matrix of the connection path between sub-components.

[0037] The layer extraction analysis module is used to perform adversarial layer extraction on the reachability matrix, obtain the overall partition of the functional resonance adversarial hierarchy model, and perform topological structure data analysis on the overall partition results to obtain the hierarchy importance and structural importance. The safety analysis of the airborne hydrogen system is then performed based on the hierarchy importance and structural importance.

[0038] The present invention provides a computer device, including a memory and a processor. The memory stores a program, and when the program is executed by the processor, the processor performs the steps of the above-described airborne hydrogen system safety analysis method.

[0039] The present invention provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described airborne hydrogen system safety analysis method.

[0040] Compared with the prior art, the present invention has the following significant advantages:

[0041] This invention dissects the interrelationships of sub-components in an airborne hydrogen system from the perspective of functional resonance. It derives a functional resonance diagram through hierarchical partitioning under an adversarial network structure, performs topological structure calculations to analyze the impact of sub-components on the overall system, identify key risk nodes, and optimize protective measures. This establishes a functional resonance adversarial hierarchical model, addressing a series of technical problems in existing airborne hydrogen system analysis methods, such as significant subjective influence from researchers and hierarchical ambiguity within the hierarchical partitioning. Furthermore, by describing the system's functional characteristics through functional resonance analysis, the overall system is partitioned to obtain the functional resonance impact of sub-components. Adversarial layer extraction is performed on the reachability matrix to obtain the overall partitioning of the functional resonance adversarial hierarchical model. This achieves the effect of dynamically identifying key risk nodes in the system, optimizing protective measures, and reducing the probability of system resonance failure through functional resonance analysis and adversarial hierarchical partitioning. It improves the reliability of safety analysis and provides theoretical support for the safety design of complex aerospace systems. Attached Figure Description

[0042] Figure 1 is an overall flowchart of the method proposed in this invention;

[0043] Figure 2 is a functional diagram of the airborne hydrogen system FRAM of the present invention;

[0044] Figure 3 is a simplified directed graph of the FRAM function diagram of the airborne hydrogen system of the present invention;

[0045] Figure 4 is a hierarchical diagram of the airborne hydrogen system FRAM-AISM of the present invention;

[0046] Figure 5 is a functional diagram of the FRAM-AISM airborne hydrogen system under the protection of the present invention;

[0047] Figure 6 is a hierarchical division diagram of the FRAM-AISM results priority-UP type for the airborne hydrogen system under the protection of the present invention;

[0048] Figure 7 is a hierarchical diagram of the FRAM-AISM cause-priority-DOWN type for the airborne hydrogen system under the protection of the present invention.

[0049] Figure 8 is a FRAM-AISM hierarchy diagram of the airborne hydrogen system under the protection of the present invention;

[0050] Figure 9 is a flowchart of a safety analysis method for an airborne hydrogen system proposed in this invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0052] As shown in Figures 1 and 9, this embodiment of the invention provides a safety analysis method for airborne hydrogen systems based on resonance theory and adversarial interpretation structures, including the following steps:

[0053] Step S1: Analyze the mutual influence relationship of sub-components of the airborne hydrogen system from the perspective of functional resonance, and derive the functional resonance diagram corresponding to the mutual influence relationship of sub-components through hierarchical division under the adversarial network structure. Analyze the functional resonance diagram through topological data to obtain the influence relationship of sub-components on the whole, key risk nodes, and optimized paths of protective measures, and establish a functional resonance adversarial hierarchical model; key risk nodes are the links in the overall process that play a decisive role in safety.

[0054] Functional resonance diagrams are diagrams that describe the functions of a large system from a functional perspective, aiming to complete the overall function. They summarize the relationships between multiple interconnected subsystems that complete their respective functions and the mutual influence of these subsystems.

[0055] Optimization path refers to the method of increasing the overall system security by adding protective measures in areas where functional resonance has a significant impact when the obtained results are unacceptable.

[0056] Step S2: Divide the airborne hydrogen system by system functional feature description, extract the functions of each sub-component, and couple the functional relationships to obtain the dynamic interaction network of sub-component functions; system functional feature description is a conditional feature that represents the system functions during startup, operation, and execution.

[0057] The airborne hydrogen system is divided into sub-components including electronic / electrical systems, hydrogen storage systems, booster pump pressurization systems, hydrogen supply pipeline systems, hydrogen internal combustion engine combustion systems, cooling systems, and sensing systems.

[0058] The functional resonance analysis method describes the functional characteristics of an airborne hydrogen system, including input, output, time, prerequisites, resources, and control. It focuses on the resonant relationships between each sub-component and the overall function, divides the system into functional parts, and extracts the specific functions of different systems. That is, it defines the functions of sub-components through the input, output, time, prerequisites, resources, and control dimensions in the system's functional characteristic description, fills in the functional characteristics of the airborne hydrogen system, couples functional relationships, obtains the relationships between functions, and obtains the dynamic interaction network of sub-component functions.

[0059] Step S3: Based on the directed graph of ordered triples, the dynamic interaction network is dimensionality reduced to obtain the adjacency matrix generated by the relationship between sub-components. The adjacency matrix is ​​then repeatedly processed by the transition matrix to obtain the reachability matrix of the connection paths between sub-components.

[0060] Input-output analysis is performed on the dynamic interaction network of the resonant system. The inputs to the dynamic interaction network are those pointed to by directed arrows, representing the things that initiate the function or the things the function will process or transform. The outputs, time, prerequisites, resources, and control of the dynamic interaction network are all the starting points of the connections during function execution. This analysis is conducted using data from the matrix. This represents the relationship between nodes i and j in a directed graph; where nodes represent the functions of sub-components; based on the relationship, the mutual relationships of the subsystems in the directed graph are simplified, i.e., digitized. The numbers in the matrix represent the one-way relationship between column elements and row elements, and the adjacency matrix is ​​constructed using the matrix data generated from the digitized mutual relationships of the sub-components.

[0061] Relationships include: =0 indicates that node i has no direct influence on node j, that is, there is no path of length 1 from node i to node j, and there is no influence between the functions; =1 indicates that node i has a direct influence on node j, meaning there exists a path of length 1 that allows node i to directly reach node j, indicating a relationship between their functions.

[0062] Step S4: Perform adversarial layer extraction on the reachability matrix to obtain the overall partitioning of the functional resonance adversarial hierarchical model, and perform topological structure data analysis on the overall partitioning results to obtain the hierarchical importance and structural importance. Then, conduct airborne hydrogen system safety analysis based on the hierarchical importance and structural importance.

[0063] Methods to combat layer extraction include: result-priority UP extraction and DOWN-type topology layer extraction; among them, result-priority UP extraction means that the sub-component is the highest priority sub-component during layer extraction, i.e., R( )= T( DOWN-type topology hierarchy extraction follows a cause-first principle, where the sub-component has the highest priority during hierarchy extraction, i.e., T( )= Q( ).

[0064] Adversarial layer extraction is performed on the reachability matrix, specifically as follows:

[0065] The factors influencing a function are hierarchically divided using the reachability matrix. The reachability set and antecedent set of each sub-component are obtained, and the common set is determined based on the reachability set and antecedent set. The antecedent set refers to the set of all upstream factors that cause a functional component to fail. The reachability set refers to the set of all cascading failure states that can be triggered by the system coupling relationship starting from the failure of a component. The common set refers to the intersection of the reachability set and the antecedent set.

[0066] By performing first-level extraction on the antecedent set, reachable set, and common set through adversarial layer extraction, the extracted sub-components are then deleted. The row and column of the adjacent Boolean matrix are used to obtain the updated matrix. Model reasoning is performed again to obtain the secondary factors again. The adversarial layer extraction is iteratively processed to complete the extraction of UP / DOWM type adversarial layers. That is, the extraction of each layer is performed iteratively to complete the extraction of adversarial layers.

[0067] The hierarchy importance and structural importance are obtained through topological data analysis, specifically:

[0068] The overall system is divided into n layers by adversarial extraction, with each layer represented by L.

[0069] The specific expression for obtaining the weights of each level in the overall system is as follows:

[0070] ;

[0071] Among them, the weights M of each level are... L Indicates the importance of the hierarchy.

[0072] Functions that point to the function to be computed are defined as in-degree functions, and functions that the function to which the function to be computed points are defined as out-degree functions. Based on the weights of each level, the structural importance of functional modules at each level is obtained, as expressed in the following expression:

[0073] ;

[0074] in, Indicates the structural importance of function p. This represents the weight of the layer where the function p to be calculated resides. This indicates the weight of the level at which function q, which points to function p, resides. This represents the weight of the level at which function k, which is pointed to by function p. O is the out-degree coefficient, and I is the in-degree coefficient.

[0075] Example:

[0076] To verify the feasibility of the method proposed in this invention, an airborne hydrogen system is used as an example to conduct a safety analysis. The system components include: F1: electronic / electrical system, F2: hydrogen storage system, F3: booster pump pressurization system, F4: hydrogen supply pipeline system, F5: hydrogen internal combustion engine combustion system, F6: pilot control system, F7: cooling system, and F8: sensing system. The airborne hydrogen system safety analysis method based on resonance theory and antagonistic interpretation structure provided in this invention includes the following steps performed in sequence:

[0077] Step 1: Classify airborne hydrogen systems from a functional perspective:

[0078] The core function of an airborne hydrogen system is to achieve the safe storage and efficient allocation of hydrogen. Its primary task is to continuously and stably deliver hydrogen fuel to the aircraft engines under complex flight conditions (covering different altitudes, attitudes, and environmental conditions) while meeting the required pressure and flow parameters. The system also possesses several synergistic functions: providing thermal management for airborne equipment through hydrogen fuel cycling, adjusting aircraft attitude using hydrogen fuel trim technology, and dynamically controlling the aircraft's center of gravity through intelligent scheduling of fuel tanks to ensure it remains within the optimal aerodynamic configuration range.

[0079] A large system completes a portion of the overall task, and this large system is further decomposed into different smaller systems. These smaller systems perform specific tasks, and the tasks of the large system are completed through the interaction between the smaller systems. This invention does not divide systems simply based on mechanical relationships, but rather on their ability to complete tasks. In this example, the large system is the airborne hydrogen system, whose core function is the safe storage and efficient allocation of hydrogen. To accomplish this task, the large system is divided into the following eight smaller systems, as described in step 2 of the specification.

[0080] Step 2: Based on the functional division of the airborne hydrogen system, identify the relationships between functions and construct an FRAM functional model:

[0081] The electronic / electrical system (F1) receives input from pilot flight commands and data collected by various sensors. Through automatic control algorithms, it outputs electrical signals to various components of the system, enabling the onboard hydrogen system to perform its various functions. The hydrogen storage system (F2) refuels from ground-based hydrogen tanks and supplies the stored hydrogen to the booster pump pressurization system (F3) during aircraft operation. The sensing system (F8) acquires hydrogen storage information and provides it to the electronic / electrical system (F1) via electrical signals, which are then displayed in real-time to the pilot (F6) on the instrument panel. Upon receiving the electrical signal from the electronic / electrical system (F1), the booster pump pressurization system (F3) activates the booster pump to pressurize the hydrogen supply pipeline (F4). The hydrogen enters the combustion chamber of the hydrogen internal combustion engine (F5) through the hydrogen supply pipeline system (F4) for combustion. The sensing system (F8) detects the flow rate at both ends of the pipeline and provides it to the electronic / electrical system (F1) in real-time. The hydrogen internal combustion engine (F5), protected by the cooling system (F7), provides power for aircraft flight. To maintain a reasonable temperature in the hydrogen internal combustion engine combustion system (F5), a sensing system (F8) is added to the cooling system (F7), which can output signals to protect the internal combustion engine. The established FRAM model is shown in Figure 2.

[0082] Step 3: Simplify the established FRAM model using a directed graph:

[0083] Based on the functional relationship network between the systems, the FRAM network diagram, i.e., the functional connection situation in it, is transformed into the mutual influence relationship between the elements of the system. The system is then simplified into a directed graph, and the result is shown in Figure 3.

[0084] The mutual influence relationship refers to the "coupling of functional relationships to obtain a dynamic interaction network of sub-component functions" in step S2 of the specification. Each small system completes the function of the larger system through interaction. However, this interaction may be accomplished through different forms and characteristic descriptions between different systems. For example, in this case, the input source of the electronic / electrical system (F1) is the pilot's command, which is manual input from the pilot. The sensing system receives electrical signals directly from the hydrogen storage system. The hydrogen storage system outputs hydrogen gas, a resource, to the booster pump system. These different forms and characteristic descriptions of interaction all belong to functional connection situations. When the hydrogen storage system runs out of hydrogen, it cannot supply hydrogen gas to the booster pump, causing the booster pump to malfunction. This situation is called the "mutual influence relationship between elements." Because this influence relationship is also linked through different forms and characteristic descriptions, there are inconveniences in hierarchical division and calculation during subsequent analysis, so a directed graph simplification is used.

[0085] Step 4: AISM Reachability Matrix Establishment:

[0086] Based on the pointing relationship of the directed graph, the link from the pointing end to the pointed end is defined. The relationship between nodes is determined by whether there is an output effect, and the adjacency matrix A can be established. Under the improved Warshall algorithm, the self-multiplication matrix is ​​obtained and the transition matrix is ​​repeatedly processed to obtain the reachability matrix R.

[0087] ;

[0088] Step 5: AISM Extraction and Hierarchical Partitioning:

[0089] After calculating the reachability matrix R, it can be used to hierarchically divide the influencing factors. First, identify each system. The reachable set R( ) and the antecedent set Q( Then, based on these two, their intersection is determined, as shown in Table 1.

[0090] The influencing factors here are actually the small systems mentioned in step 3. Due to the concept of functional resonance, changes in each small system directly or indirectly affect other systems. For other systems, the changed small system becomes an influencing factor. Furthermore, because the changed or affected small system cannot complete all functions or errors occur in the interaction between functions, the larger system also cannot fully perform its functions. In this case, the larger system is also affected from a functional perspective; therefore, the changed or affected small system is an influencing factor for the larger system. Since every small system can change or be affected, every small system is an influencing factor for the larger system.

[0091] Table 1. Reachable set, cause set, and their intersection

[0092]

[0093] After obtaining the reachable sets, antecedent sets, and intersections of each factor, the UP / DOWN hierarchical division can be obtained according to the AISM adversarial extraction rules, as shown in Table 2. The UP type hierarchical differentiation is {F5}≺{F1, F3, F4, F6, F7, F8}≺{F2}. The DOWN type hierarchical differentiation is {F5}≺{F1, F3, F4, F6, F7, F8}≺{F2}. The UP and DOWN hierarchical differentiations are exactly the same, as shown in Figure 4, which yields the FRAM analysis diagram.

[0094] Table 2 Hierarchical decomposition extraction results

[0095]

[0096] Step 6: Add protection analysis:

[0097] Based on the basic FRAM airborne hydrogen system, safety protection measures are introduced for the entire system, conforming to the protection and suppression principles of FRAM. The specific protection measures are based on the concepts of "human-machine-environment" and "intrinsic safety." A leak prevention system (F9) is added to the hydrogen storage system (F2) to reduce the risk of hydrogen leakage from the storage system (F2) and to effectively suppress hydrogen leakage in the event of a leak. In accordance with relevant airworthiness regulations, a redundant design is added to the pump pressurization system (F3) – a booster pump redundancy design system (F10). The sensing system (F8) is monitored and protected (F11).

[0098] The protected FRAM model is shown in Figure 5. The security analysis protection practice here implements the concept of "optimization path of protection measures" in step S1 of the aforementioned specification. Protection measures refer to introducing new subsystems into a large system to reduce the impact of functional resonance between systems and improve system security. The purpose of adding the protection system here is mainly twofold: 1. To demonstrate the usefulness of the invention. After analyzing the unprotected system, a comparative analysis is performed on the system with added protection, showing that this system can indeed improve system security. (However, for some other cases where the desired security standard can be achieved without adding protection measures, protection measures may not be necessary.) 2. In this example, the hierarchical decomposition extraction result obtained from the large system without added protection measures does not show the existence of a resonance root layer, and therefore cannot demonstrate the advantages of AISM in this method compared to the traditional ISM method.

[0099] Step 7: Repeat steps 2 to 5 above for the protected system to obtain the result priority-UP type network diagram of the FRAM-AISM model of the airborne hydrogen system under protection (Figure 6) and the cause priority-DOWN type network diagram of the FRAM-AISM model of the airborne hydrogen system under protection (Figure 7).

[0100] Step 8: FRMA-AISM under protection:

[0101] Regarding the adversarial extraction results, the UP and DOWN extraction results for the first three layers are consistent, with each layer being directly independent and non-interfering with each other. Systems F10 and F11 are at different layers in the UP and DOWN types. From the perspective of the AISM method, all systems in the third to fourth layers have equal effects and belong to the resonance root layer in this case. The specific FRMA-AISM hierarchical division network diagram under protection is shown in Figure 8.

[0102] Step 9: Functional Structure Quantitative Analysis:

[0103] For the structural quantitative analysis of the airborne hydrogen system, the weights of the levels from the system resonance surface to the system resonance root are calculated according to the hierarchical division of the FRAM-AISM model. Substituting n=3 into (1) can calculate the weight of each level. The results are shown in Table 3.

[0104] Table 3. Importance of FRAM-AISM Functional Hierarchy

[0105]

[0106] Based on the calculation of the functional hierarchy importance of FRAM-AISM, and according to the introduced concept of node in-degree and out-degree, the calculation of subsystem in-degree and out-degree is added to each system.

[0107] Substituting the calculated results of the functional hierarchy importance into (2), the structural importance R of each level of the system can be calculated. i The calculation results are shown in Table 4.

[0108] Table 4. Calculation of the structural importance of the FRAM-AISM system.

[0109]

[0110] The weights of each level can be calculated by substituting n=4 into (1) based on the FRAM-AISM hierarchical network diagram under protection. The results are shown in Table 5.

[0111] Table 5. Importance of FRAM-AISM Functional Levels under Protection

[0112]

[0113] Based on the calculation of the functional hierarchy importance of FRAM-AISM, and according to the introduced concept of node in-degree and out-degree, the in-degree and out-degree calculation of each subsystem is added to the calculation of each system. Substituting the calculation results of the functional hierarchy importance into (2), the structural importance of each level of the system can be calculated. The calculation results are shown in Table 6.

[0114] Table 6. Calculation of Structural Importance of FRAM-AISM System under Protection

[0115]

[0116] This invention proposes an airborne hydrogen system safety analysis system, comprising: a model building module, a system partitioning module, an image matrix processing module, a layer extraction analysis module, and a safety analysis module.

[0117] The model building module is used to analyze the mutual influence relationships of sub-components of the airborne hydrogen system from the perspective of functional resonance. It derives the functional resonance diagram corresponding to the mutual influence relationships of sub-components through hierarchical division under the adversarial network structure. Through topological data analysis of the functional resonance diagram, it obtains the influence relationship of sub-components on the whole, key risk nodes, and optimized paths of protective measures, and establishes a hierarchical model of functional resonance adversarial measures. Key risk nodes are the links in the overall process that play a decisive role in safety. The system partitioning module is used to partition the airborne hydrogen system through the description of system functional characteristics, extract the functions of each sub-component, and couple the functional relationships to obtain the dynamic interaction network of sub-component functions. The system functional characteristics description represents the conditional characteristics that define the system functions during startup, operation, and execution. The image matrix processing module is used to reduce the dimensionality of the dynamic interaction network based on the directed graph of ordered triples, obtain the adjacency matrix generated by the interrelationships of sub-components, and repeatedly perform transition matrix processing on the adjacency matrix to obtain the reachability matrix of the connection paths between sub-components. The layer extraction analysis module is used to perform adversarial layer extraction on the reachability matrix, obtain the overall partition of the functional resonance adversarial hierarchical model, and perform topological structure data analysis on the overall partition results to obtain the hierarchical importance and structural importance. The safety analysis of the airborne hydrogen system is then performed based on the hierarchical importance and structural importance.

[0118] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a program, and when the program is executed by the processor, the processor performs the steps of an airborne hydrogen system safety analysis method.

[0119] According to the disclosed embodiments, the computer device can communicate with one or more external devices (e.g., keyboard, pointing device, Bluetooth communication, etc.) or with any device that enables the computing device to communicate with one or more other computing devices (e.g., router, demodulator, etc.).

[0120] The above description, in conjunction with specific preferred embodiments, provides a more detailed explanation of the present invention. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such deductions or substitutions should be considered to fall within the scope of protection of the present invention.

Claims

1. A method for safety analysis of an airborne hydrogen system, characterized in that, include: This paper analyzes the inter-component relationships of an airborne hydrogen system from the perspective of functional resonance. A functional resonance diagram corresponding to these relationships is derived through hierarchical partitioning under an adversarial network structure. Topological data analysis of the functional resonance diagram reveals the impact of sub-components on the overall system, key risk nodes, and optimized paths for protective measures, establishing a hierarchical model of functional resonance adversarial measures. Key risk nodes are links in the overall process that play a decisive role in safety. Optimized paths refer to methods that increase overall system safety by adding protective measures where the functional resonance impact is significant, when the obtained results are unacceptable. The airborne hydrogen system is partitioned using system functional characteristic descriptions, extracting the functions of each sub-component and coupling their functional relationships to obtain a dynamic interaction network of sub-component functions. The system functional characteristic descriptions represent the system's... The system defines the conditional characteristics of the function during startup, operation, and execution; based on a directed graph of ordered triples, the dynamic interaction network is dimensionality reduced to obtain the adjacency matrix generated by the relationships between sub-components, and the adjacency matrix is ​​repeatedly processed by the transition matrix to obtain the reachability matrix of the connection paths between sub-components; adversarial layer extraction is performed on the reachability matrix to obtain the overall partition of the functional resonance adversarial hierarchical model, and topological data analysis is performed on the overall partition results to obtain the hierarchical importance and structural importance, and the airborne hydrogen system safety is analyzed through hierarchical importance and structural importance; the topological data analysis of the overall partition results to obtain hierarchical importance and structural importance specifically involves: dividing the overall system into n layers through adversarial extraction, with each layer represented by L; obtaining the weight of each layer of the overall system, specifically expressed as: Among them, the weight M of each level L This represents the importance of the hierarchy; functions that point to the function to be calculated are defined as in-degree functions, and functions that the function to which the function to be calculated points are defined as out-degree functions. Based on the weight of each hierarchy, the structural importance of each functional module is obtained, and the specific expression is as follows: ;in, Indicates the structural importance of function p. This represents the weight of the layer where the function p to be calculated resides. This indicates the weight of the level at which function q, which points to function p, resides. This represents the weight of the level at which function k, which is pointed to by function p. O is the out-degree coefficient, and I is the in-degree coefficient.

2. The airborne hydrogen system safety analysis method as described in claim 1, characterized in that, The process involves dividing the airborne hydrogen system using system functional feature descriptions, extracting the functions of each sub-component, and coupling functional relationships to obtain a dynamic interaction network of sub-component functions. Specifically, the airborne hydrogen system is divided into sub-components including an electronic / electrical system, a hydrogen storage system, a booster pump pressurization system, a hydrogen supply pipeline system, a hydrogen internal combustion engine combustion system, a cooling system, and a sensing system. The functions of the sub-components are defined using the input, output, time, prerequisite, resource, and control dimensions in the system functional feature descriptions, and the relationships between functions are obtained to obtain a dynamic interaction network of sub-component functions.

3. The airborne hydrogen system safety analysis method as described in claim 2, characterized in that, The method uses a directed graph of ordered triples as a basis to reduce the dimensionality of the dynamic interaction network and obtain the adjacency matrix generated by the relationships between sub-components. Specifically, it performs input and output analysis on the dynamic interaction network, where the input of the dynamic interaction network is the thing that initiates the function or the thing that the function will process or transform, and the output, time, prerequisites, resources, and control of the dynamic interaction network are all pointing to the starting point of the connection when the function is running; through the data in the matrix... This represents the relationship between node i and node j in a directed graph; The nodes represent the functions of the sub-components; based on the relationships, the interrelationships of the sub-components of the airborne hydrogen system are digitized from the directed graph, and an adjacency matrix is ​​constructed using the matrix data generated from the digitized interrelationships of the sub-components.

4. The airborne hydrogen system safety analysis method as described in claim 3, characterized in that, The relationship includes: =0 indicates that node i has no direct influence on node j, that is, there is no path of length 1 from node i to node j, and there is no influence between the functions; =1 indicates that node i has a direct influence on node j, meaning there exists a path of length 1 that allows node i to directly reach node j, indicating a relationship between their functions.

5. The airborne hydrogen system safety analysis method as described in claim 1, characterized in that, The adversarial layer extraction of the reachability matrix specifically involves: hierarchically dividing the influencing factors of a function using the reachability matrix, obtaining the reachability set and antecedent set for each sub-component, and determining a common set based on the reachability set and antecedent set; wherein, the antecedent set refers to the set of all upstream factors that lead to the failure of a certain functional component, the reachability set refers to the set of all cascading failure states that can be triggered by the system coupling relationship starting from the failure of a certain component, and the common set refers to the intersection of the reachability set and the antecedent set; the antecedent set, reachability set, and common set are extracted at the first layer through adversarial layer extraction, and after the first layer extraction is completed, the rows and columns of the adjacency matrix of the extracted sub-component are deleted to obtain the updated matrix, and the extraction is iteratively performed at each layer to complete the extraction of the adversarial layer.

6. The airborne hydrogen system safety analysis method as described in claim 5, characterized in that, The anti-layer extraction method specifically includes: result-priority UP extraction and DOWN topology layer extraction; wherein, result-priority UP extraction means that the sub-component is the highest priority sub-component during layer extraction, and DOWN topology layer extraction is performed according to the principle of cause priority, and the sub-component is the highest priority sub-component during layer extraction.

7. A safety analysis system for an airborne hydrogen system, characterized in that, include: The model building module analyzes the inter-component relationships of the airborne hydrogen system from the perspective of functional resonance. It derives a functional resonance diagram corresponding to these inter-component relationships through hierarchical partitioning under an adversarial network structure. Topological data analysis of the functional resonance diagram yields the impact relationships of sub-components on the overall system, key risk nodes, and optimized paths for protective measures, establishing a hierarchical model of functional resonance adversarial measures. Key risk nodes are links in the overall process that play a decisive role in safety. Optimized paths refer to methods that increase overall system safety by adding protective measures where the functional resonance impact is significant, when the obtained results are unacceptable. The system partitioning module divides the airborne hydrogen system based on system functional feature descriptions, extracts the functions of each sub-component, and couples functional relationships to obtain a dynamic interaction network of sub-component functions. The system functional feature descriptions represent the system's functional characteristics. The system includes: a conditional feature defined during startup, operation, and execution; an image matrix processing module for dimensionality reduction of a dynamic interactive network based on a directed graph of ordered triples, obtaining an adjacency matrix generated by the relationships between sub-components, and repeatedly performing transition matrix processing on the adjacency matrix to obtain a reachability matrix of connection paths between sub-components; a layer extraction analysis module for adversarial layer extraction of the reachability matrix, obtaining the overall partitioning of the functional resonance adversarial hierarchical model, and performing topological data analysis on the overall partitioning results to obtain hierarchical importance and structural importance, and performing airborne hydrogen system safety analysis based on hierarchical importance and structural importance; specifically, the topological data analysis of the overall partitioning results to obtain hierarchical importance and structural importance involves: dividing the overall system into n layers through adversarial extraction, with each layer represented by L; and obtaining the weight of each layer of the overall system, specifically expressed as: Among them, the weight M of each level L This represents the importance of the hierarchy; functions that point to the function to be calculated are defined as in-degree functions, and functions that the function to which the function to be calculated points are defined as out-degree functions. Based on the weight of each hierarchy, the structural importance of each functional module is obtained, and the specific expression is as follows: ;in, Indicates the structural importance of function p. This represents the weight of the layer where the function p to be calculated resides. This indicates the weight of the level at which function q, which points to function p, resides. This represents the weight of the level at which function k, which is pointed to by function p. O is the out-degree coefficient, and I is the in-degree coefficient.

8. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores a program that, when executed by the processor, causes the processor to perform the steps of the airborne hydrogen system safety analysis method as described in any one of claims 1 to 6.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the airborne hydrogen system safety analysis method according to any one of claims 1 to 6.

Images