Intelligent rcma analysis method and device based on mbse-llm

Abstract

The application relates to the technical field of the cross of artificial intelligence and reliability system engineering, and provides an intelligent RCMA analysis method and device based on MBSE-LLM. The method constructs a logic model of an analysis object by using an MBSE method, determines a component type of each component in the logic model by using LLM based on component information and FMEA results, determines a fault influence consequence type of an important type component by using LLM, determines a preventive maintenance work type based on a fault cause and the fault influence consequence type, and finally obtains RCMA results of each important type component by combining the fault influence consequence type and the preventive maintenance work type. The method is based on a large model and performs maintenance analysis in a reliability-centered manner, realizes intelligent division of a fault influence type, accurate matching of maintenance work and automatic generation of RCMA results, greatly improves analysis efficiency and accuracy, and provides a scientific and objective decision basis for equipment maintenance strategy making.

Description

Technical Field

[0001] This application relates to the interdisciplinary field of artificial intelligence and reliability systems engineering, and in particular to an intelligent RCMA analysis method and apparatus based on MBSE-LLM. Background Technology

[0002] Reliability-Centered Maintenance Analysis (RCMA) is a process that uses logical decision-making to determine preventative maintenance requirements for equipment, based on the principle of maintaining inherent reliability and safety with minimal maintenance resource consumption. Traditional RCMA analysis methods rely heavily on manual analysis by designers or reliability analysts, which is time-consuming and costly. Furthermore, the quality of traditional RCMA analysis is heavily dependent on the designer's experience, leading to inaccuracies and inconsistencies in terminology. Moreover, manual analysis struggles to comprehensively cover potential failure modes in complex systems, posing a risk of omissions. This is particularly problematic in the rapidly iterating environment of modern equipment development, where traditional methods are no longer sufficient to meet the demands for timeliness and comprehensiveness.

[0003] Currently, Large Language Models (LLMs), with their powerful text understanding and generation capabilities, are widely used in professional and technical fields to assist engineers in analysis, design, and development activities. Therefore, there is an urgent need for an intelligent RCMA analysis method that can leverage LLM to deeply integrate systems engineering knowledge and accurately understand the RCMA logical decision-making process. Summary of the Invention

[0004] In view of this, embodiments of this application provide an intelligent RCMA analysis method and apparatus based on MBSE-LLM to solve the problems of low efficiency and poor accuracy in RCMA analysis in the prior art.

[0005] A first aspect of this application provides an intelligent RCMA analysis method based on MBSE-LLM, comprising:

[0006] A logical model of the analysis object is constructed using the Model-Based Systems Engineering (MBSE) approach; the logical model includes at least hierarchical entity components of the analysis object.

[0007] Obtain component information and Failure Mode and Effect Analysis (FMEA) results for each component in the logical model; the component information should at least include the functional description of the component, and the FMEA results should at least include the failure mode, failure cause, failure effect, and severity level of the component;

[0008] Traverse all components in the logical model and use the Large Language Model (LLM) to determine the component type based on the functional description of each component and the FMEA results; the component type includes important types and non-important types.

[0009] Iterate through all important component types, and for each important component type, iterate through all failure modes of this component; for each failure mode, use LLM to determine the type of failure impact consequences by combining the failure impact and severity level corresponding to this failure mode.

[0010] For each failure mode, iterate through its corresponding failure causes and use LLM to determine the type of preventive maintenance work based on the failure cause and the type of failure impact consequences.

[0011] By combining the failure impact consequence type and preventive maintenance work type of each failure mode, the RCMA result of this failure mode is obtained, and then the RCMA result of each important type of component is obtained.

[0012] A second aspect of this application provides an intelligent RCMA analysis device based on MBSE-LLM, comprising:

[0013] The building module is configured to construct a logical model of the analysis object using the Model-Based Systems Engineering (MBSE) approach; the logical model includes at least hierarchical entity components of the analysis object.

[0014] The acquisition module is configured to acquire component information and Failure Mode and Effects Analysis (FMEA) results for each component in the logical model. The component information includes at least a functional description of the component, and the FMEA results include at least the failure mode, failure cause, failure effect, and severity level of the component.

[0015] The classification module is configured to traverse all components in the logical model and use the Large Language Model (LLM) to determine the component type based on the functional description of each component and the FMEA results; the component type includes important types and non-important types.

[0016] The analysis module is configured to traverse all important type components, traverse all failure modes of each important type component, and for each failure mode, use LLM to determine the failure impact consequence type by combining the failure impact and severity level corresponding to the failure mode.

[0017] The analysis module is also configured to iterate through the corresponding causes of each failure mode and use LLM to determine the type of preventive maintenance work based on the cause and the type of failure impact consequences.

[0018] The combination module is configured to combine the failure impact consequence type and preventive maintenance operation type of each failure mode to obtain the RCMA result of this failure mode, and then obtain the RCMA result of each important type of component.

[0019] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0021] The beneficial effects of the embodiments in this application compared with the prior art are:

[0022] This application utilizes the MBSE method to construct a logical model of the analysis object. For each component in the logical model, LLM is used to determine the component type based on its component information and FMEA results. For important component types, LLM first determines the failure impact consequence type, then determines the preventive maintenance work type based on the failure cause and failure impact consequence type, and finally combines the failure impact consequence type and preventive maintenance work type to obtain the RCMA results for each important component type. This method performs maintenance analysis based on a large model with reliability as the core, realizing intelligent classification of failure impact types, accurate matching of maintenance work, and automated generation of RCMA analysis, significantly improving the efficiency and accuracy of RCMA analysis and providing a scientific and objective decision-making basis for equipment maintenance strategy formulation. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating an intelligent RCMA analysis method based on MBSE-LLM provided in an embodiment of this application.

[0025] Figure 2 This is a schematic diagram illustrating the process of generating structured judgment criteria for component nodes provided in the embodiments of this application.

[0026] Figure 3 This is a schematic diagram illustrating the process and basis for determining the type of fault impact results provided in the embodiments of this application.

[0027] Figure 4 This is a schematic diagram illustrating the process and basis for determining the type of preventive maintenance work provided in the embodiments of this application.

[0028] Figure 5 This is a schematic diagram of various RCMA tables generated using the ground mobile platform-control box as the analysis object.

[0029] Figure 6 This is a flowchart illustrating another intelligent RCMA analysis method based on MBSE-LLM provided in this application embodiment.

[0030] Figure 7 This is a schematic diagram of an intelligent RCMA analysis device based on MBSE-LLM provided in an embodiment of this application.

[0031] Figure 8 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0032] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0033] The following will describe in detail, with reference to the accompanying drawings, an intelligent RCMA analysis method and apparatus based on MBSE-LLM according to embodiments of this application.

[0034] As mentioned above, traditional RCMA analysis methods are usually performed manually by designers or analysts, and their efficiency and effectiveness heavily rely on human experience.

[0035] In view of this, this application provides an intelligent RCMA analysis method based on MBSE-LLM. First, the Failure Mode and Effect Analysis (FMEA) results of relevant components are input, and important functional components are identified based on the large model, generating a list of important functional components. Then, the FMEA results of all important functional components are traversed, and decisions and analyses are performed item by item based on the large model to determine the type of failure impact consequences and select applicable and effective preventive maintenance work types. At the same time, preventive maintenance work for non-important functional components is determined based on the large model, thereby obtaining the RCMA analysis results of relevant components. This achieves intelligent classification of failure impact types, accurate matching of maintenance work, and automated generation of RCMA analysis, greatly improving the efficiency and accuracy of RCMA analysis and providing a scientific and objective decision-making basis for equipment maintenance strategy formulation.

[0036] Figure 1 This is a flowchart illustrating an intelligent RCMA analysis method based on MBSE-LLM provided in an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0037] In step S101, the logical model of the analysis object is constructed using the MBSE method.

[0038] The logical model includes at least hierarchical entity components of the analysis objects.

[0039] In step S102, the component information and FMEA results of each component in the logical model are obtained.

[0040] The component information should include at least a functional description of the component, and the FMEA results should include at least the failure mode, failure cause, failure effect, and severity level of the component.

[0041] In step S103, all components in the logical model are traversed, and the component type is determined using LLM based on the functional description of each component and the FMEA results.

[0042] The component types include important types and non-important types.

[0043] In step S104, all important type components are traversed, and for each important type component, all failure modes of the component are traversed; for each failure mode, the failure impact consequence type is determined by combining the failure impact and severity level corresponding to the failure mode using LLM.

[0044] In step S105, for each fault mode, its corresponding fault causes are traversed, and the preventive maintenance work type is determined based on the fault cause and the type of fault impact consequence using LLM.

[0045] In step S106, the failure impact consequence type and preventive maintenance work type of each failure mode are combined to obtain the RCMA result of this failure mode, and then the RCMA result of each important type of component is obtained.

[0046] In some embodiments of this application, the method may be executed by a server or by a terminal device with certain processing capabilities.

[0047] In some embodiments of this application, a logical model of the analysis object can first be constructed using the model-based systems engineering (MBSE) method. Then, component information and Failure Mode and Effects Analysis (FMEA) results for each component in the logical model are obtained.

[0048] The component information should include at least a functional description of the component, and the FMEA results should include at least the component's failure mode, failure cause, failure impact, and severity level. The failure impact can include local impact, higher-level impact, and final impact.

[0049] In some embodiments of this application, all components in the logical model can be traversed, and the component types can be determined using LLM based on the functional descriptions of each component and FMEA results. Corresponding lists can then be generated for different component types. For example, if the component types include important and unimportant types, separate lists of important and unimportant components can be generated.

[0050] In some embodiments of this application, all important component types can be traversed, and for each important component type, all failure modes of that component can be traversed. For each failure mode, LLM is used to determine the failure impact consequence type by combining the failure impact and severity level corresponding to that failure mode. Furthermore, for each failure mode, its corresponding failure cause is also traversed, and LLM is used to determine the preventive maintenance work type based on the failure cause and the failure impact consequence type.

[0051] In some embodiments of this application, the failure impact consequence type and preventive maintenance work type of each failure mode can be combined to obtain the RCMA result of the failure mode, and then the RCMA result of each important type of component can be obtained.

[0052] According to the technical solution provided in the embodiments of this application, a logical model of the analysis object is constructed using the MBSE method. For each component in the logical model, the component type is determined using LLM based on its component information and FMEA results. For important component types, LLM first determines the failure impact consequence type, then determines the preventive maintenance work type based on the failure cause and failure impact consequence type. Finally, the failure impact consequence type and preventive maintenance work type are combined to obtain the RCMA results for each important component type. This method performs maintenance analysis based on a large model with reliability as the core, achieving intelligent classification of failure impact types, precise matching of maintenance work, and automated generation of RCMA analysis, significantly improving the efficiency and accuracy of RCMA analysis and providing a scientific and objective decision-making basis for equipment maintenance strategy formulation.

[0053] In some embodiments of this application, constructing the logical model of the analysis object using the MBSE method can be done by assuming the overall analysis object system is as follows: The MBSE method is used to divide the analyzed object into multiple logical levels based on its actual hardware architecture, denoted as... , of which each layer It includes a set of components that make up this layer. .in this way, It can represent the first The first in the layer Each component, also known as a physical unit, For the first The total number of components in a layer.

[0054] Within each defined hierarchy, identify and create all entity component nodes that constitute that hierarchy, and then perform component... Define its attribute tuple ;in, A unique identifier for the component. For component name, This is for component functionality. All node components can be stored in the structured data format JSON.

[0055] After defining the component nodes at each level, you can further establish connections between components. Connections can be formed by port definitions and connection lines: Let the port set be... Each port Must be with one and only one component Associated, with a unique port identifier Port type and port direction Let the set of connection relations be... Each connection A directed edge from one output port to one input port is defined as: Among them, the connection types include three categories: matter, information, and energy, namely... Text attributes This provides a supplementary explanation of the connection relationship, highlighting that the connection lines have directionality.

[0056] in, Indicates Boolean type, Indicates integer. Indicates floating-point type, Represents a character type; Indicates the input port type. Indicates the output port type; It is a positive integer; express A unique identifier, express input port, express The output port, express Connection type; To represent information, Indicates energy. It refers to matter.

[0057] The logical model constructed through the above steps is represented as a set of nodes. and edge set The directed graph formed The data is stored in the structured data format JSON to obtain the logical model of the object being analyzed.

[0058] It can traverse all component nodes in the logical model, obtain the functional description and FMEA results of each component node, and then call LLM to identify the important functional components that have a key impact on the security, task completion and economics of the system being analyzed based on the functional description and FMEA results of each component node.

[0059] In other words, in some embodiments of this application, the component information may also include the component code and component name of the component. In this case, determining the component type using LLM based on the functional descriptions of each component and the FMEA results may include: combining the functional description of the target component with the fault impact and severity level from its FMEA results to obtain a fault description of the target component; wherein, the target component is any component in the logical model; using the fault description as a prompt to guide LLM in determining the component type of the target component; constructing lists of important and unimportant component types based on the determined component types; wherein each component list includes the component code, component name, functional description, and fault mode of the component of that type.

[0060] In some implementations, the functional description "function" of a component node in the logical model and the failure mode list "failure_mode_list" under that component node can be used as the failure description of the component. The "failure_mode_list" can include information such as the failure mode name "name", the failure impact, and the severity level "severity_level".

[0061] Then, based on the fault description, the LLM is guided to analyze whether the component is important according to the fault impact and severity level, and output structured judgment criteria to ensure that the judgment process is transparent and traceable.

[0062] Figure 2 This is a schematic diagram illustrating the process of generating structured judgment criteria for component nodes provided in the embodiments of this application. For example... Figure 2 As shown, you can first input data in JSON format, including system component information and FMEA data. The system component information can include component name, component code, and functional description, while the FMEA data can include failure modes, failure causes, failure effects, and severity levels. Then, the data is parsed, component nodes (items) are traversed, and based on the functional description (function), the failure mode list "failure_mode_list", and the severity level "severity_level", a large model is invoked for multi-step analysis to identify important functional components.

[0063] The analysis process may include: comprehensively assessing the safety, mission, and economic impacts based on severity levels; and generating structured decision criteria. An example structured decision criterion could be that a component's failure mode "F01" has a severity level of 1 and involves equipment safety, therefore it is determined to be an important functional component. After the analysis is complete, the importance determination result for the component can be output, including yes or no (yes / no) and the decision criteria.

[0064] Finally, different types of components are automatically categorized and structured JSON lists are generated for RCMA input processing.

[0065] In some embodiments of this application, using LLM to determine the type of failure impact consequences by combining the failure impact and severity level corresponding to the failure mode may include: generating a failure description of the failure mode; the failure description includes the component name corresponding to the failure mode, the functional description of the component corresponding to the failure mode, the failure mode, the failure impact and severity level; and guiding LLM to perform preset logic judgment based on the failure description to determine the type of failure impact consequences of the failure mode.

[0066] The preset logic judgment includes at least one of the following: whether the discovery of the functional failure corresponding to this failure mode is obvious to the operator of the equipment under normal use; whether the functional failure corresponding to this failure mode or the secondary loss caused by the functional failure has a direct harmful impact on the safety of use; whether the functional failure corresponding to this failure mode has a direct harmful impact on the completion of the task; whether the combination of the hidden functional failure corresponding to this failure mode or other related or backup functional failures of this component has a harmful impact on the safety of use; or whether the combination of the hidden functional failure corresponding to this failure mode or other related or backup functional failures of other components has a harmful impact on the completion of the task.

[0067] Backup function failure refers to a functional failure in which the backup system used for emergency response also fails when a system malfunctions, resulting in the system being unable to operate normally.

[0068] In other words, it is possible to iterate through the failure mode node "mode" of each important type component in the list of important type components and determine the corresponding failure impact consequence type. In some examples, the failure description, including component name "name", function description "function", failure mode "failure_mode", local effects "local_effects", higher-level effects "system_effects", end effects "end_effects", and severity level "severity_level", can be entered into the LLM. The LLM will then execute preset logic to determine the failure impact consequence type of this failure mode.

[0069] When determining whether the discovery of a functional fault corresponding to this fault mode is readily apparent to operators performing normal operations, guiding prompts can be constructed based on the fault description above, combined with the definitions of "normal duties" and "discovery." This guides the LLM to determine whether the occurrence of the functional fault is readily apparent to the operator; that is, whether the operator performing "normal duties" can promptly or immediately "discover" the fault, and outputs structured judgment criteria. For example, {"Reason": "In most cases, fuse overload blowing is only a passive measure to protect the circuit and cannot actively send a signal. Although the circuit loses power after the fuse blows, which may ultimately manifest as a control box not responding, directly leading to functional failure, the operator cannot directly discover the fuse blowing during operation, so it is not readily apparent." "Judgment": "No"}.

[0070] When determining whether a functional failure or secondary loss caused by a functional failure has a direct harmful impact on operational safety, the impact of the failure can be determined based on the above failure description, combined with concepts such as "direct harmful impact" and "secondary damage". Guiding prompts can be constructed by combining judgment criteria such as "whether it involves personnel safety or equipment damage", "whether it is unavoidable" and "whether it has an immediate impact", to guide the LLM in determining whether a functional failure or secondary loss caused by a functional failure has a direct harmful impact on operational safety, and output structured judgment criteria.

[0071] When determining whether a functional failure corresponding to this failure mode has a direct harmful impact on task completion, based on the above failure description, guiding prompts can be constructed by combining judgment criteria such as "whether it has a direct harmful impact on task completion", "whether it involves the core functions of task execution" and "whether it is unavoidable" to guide LLM in determining whether a functional failure has a direct harmful impact on task completion.

[0072] When determining whether the combined effects of a concealed function failure or other related or backup function failures of this component corresponding to this failure mode pose a harmful impact on operational safety, the impact of the failure can be determined based on the above failure description, combined with concepts such as "direct harmful impact," "secondary damage," and "multiple failures." Furthermore, guiding prompts can be constructed by combining judgment criteria such as "whether it involves personnel safety or equipment damage," "whether it is unavoidable," and "whether it has an immediate impact," to guide the LLM in determining whether the combined effects of a concealed function failure or other related or backup function failures of this component pose a harmful impact on operational safety.

[0073] When determining whether the combination of hidden function failures or other component-related or backup function failures corresponding to this failure mode has a detrimental effect on task completion, based on the above failure description, guiding prompts can be constructed by combining judgment criteria such as "whether it has a direct detrimental effect on task completion", "whether it involves the core functions of task execution" and "whether it is unavoidable" to guide LLM in determining whether the combination of hidden function failures or other component-related or backup function failures has a detrimental effect on task completion.

[0074] The above logical judgment results can be output in a structured manner, including the fault impact type judgment result i_type and all logical judgment criteria, to ensure that the judgment process is transparent and traceable.

[0075] In some embodiments of this application, the determined types of failure impact consequences include at least safety impact types, mission-related impact types, and economic impact types. Specifically, safety impact types may include overt safety impact types and covert safety impact types; mission-related impact types may include overt mission-related impact types and covert mission-related impact types; and economic impact types may include overt economic impact types and covert economic impact types.

[0076] In some embodiments of this application, the type of preventive maintenance work can be further determined based on the type of failure impact consequences. Determining the type of preventive maintenance work based on failure causes and failure impact consequence types using LLM can include: determining a decision path based on the failure impact consequence type of the target failure mode; wherein, the target failure mode is any failure mode; safety impact type failure impact consequences, task impact type failure impact consequences, and economic impact type failure impact consequences correspond to different decision paths; using LLM to perform applicability analysis on each preventive maintenance work in the decision path to obtain applicability judgment criteria; using LLM to perform effectiveness analysis on each preventive maintenance work in the decision path to obtain effectiveness judgment criteria; combining the applicability judgment criteria, effectiveness judgment criteria, and the definitions of each preventive maintenance work in the decision path to obtain comprehensive prompt words; and using LLM to determine the type of preventive maintenance work for the target failure mode based on the comprehensive prompt words.

[0077] In other words, when using LLM to determine preventive maintenance work types based on failure causes and failure consequences, one can iterate through the failure cause nodes "cause" under the failure mode "mode" and determine the applicable and effective preventive maintenance work types based on the failure causes and FMEA results. These preventive maintenance work types can include maintenance, operator monitoring, usage inspection, functional testing, scheduled disassembly and repair, and scheduled scrapping.

[0078] For each cause of a failure, the decision-making path for preventive maintenance work is first determined based on the type of consequences of the failure.

[0079] For safety-related impact types, the decision path requires a comprehensive analysis of all preventative maintenance issues. It involves selecting the most effective solution from all applicable and effective preventative measures (e.g., maintenance, operator monitoring, functional testing, scheduled repairs, scheduled scrapping) and considering the existence of integrated measures. If no applicable measures exist, a mandatory design change is necessary to ensure safety. When analyzing safety-related impact types using the decision path, regardless of whether an intermediate answer is "yes," the analysis must continue to the last node of the decision path.

[0080] For mission-related impact types, the decision-making path requires starting with the problem (Is maintenance applicable and effective?), followed by analyzing whether subsequent preventative work (operator monitoring, functional testing, scheduled disassembly and repair, scheduled scrapping) is applicable and effective. If any subsequent preventative maintenance work is answered "yes," the decision-making process terminates and that work is selected. At the same time, the existence of comprehensive work is considered. It is not necessary to exhaust all questions. If all are answered "no," then there is no applicable preventative work. Depending on the equipment type, the mission impact must be weighed, or for civilian products, the mission losses and design costs must be weighed to decide whether to change the design.

[0081] For economic impact types, the decision-making path is basically the same as for task-related types. The difference is that design changes only need to weigh the failure losses against the cost of redesigning (without considering task or safety impacts), without considering comprehensive work. However, for economic impact types, when analyzing according to the decision-making path, the decision-making process can be terminated when the answer to any intermediate node is "yes," without having to exhaust all questions.

[0082] For the types of hidden failure impacts, the decision-making paths for safety, mission, and economy are the same as those for obvious failures. The only difference is that the preventive maintenance work to be analyzed, "operator monitoring," is replaced with "usage inspection."

[0083] Then, an applicability analysis is performed on each preventive maintenance task in the decision-making path. During the analysis, cue words can be constructed based on the applicability definitions of each preventive maintenance task. For example, when determining the applicability conditions for preventive maintenance task "operator monitoring," it is necessary to determine whether component functional degradation is detectable and whether there are definable potential fault states. Using the above method, LLM is employed to determine whether the applicability conditions for various preventive maintenance tasks are met, and structured judgment criteria are output.

[0084] Next, an effectiveness analysis is performed on each preventative maintenance activity in the decision-making path. During the analysis, cue words can be constructed based on the effectiveness definition of each preventative maintenance activity. For example, when determining the effectiveness condition of maintenance activity "operator monitoring," it is necessary to determine whether the maintenance activity can reduce the probability of failure or multiple failures to an acceptable level. If the failure impact is economic, it is necessary to determine whether the activity can effectively reduce economic losses to an acceptable level. Using the above method, LLM is employed to determine whether the effectiveness conditions of various preventative maintenance activities are met, and structured judgment criteria are output.

[0085] Finally, the applicability and validity criteria can be input into the LLM as the initial prompts. Guided prompts can then be constructed based on the definitions of various preventative maintenance tasks and the overall work definition, and input into the LLM as well, so that the LLM can determine the preventative maintenance task type for the failure mode. The preventative maintenance task type can include a single preventative maintenance task type or a combination of multiple preventative maintenance task types.

[0086] After determining the type of preventive maintenance work, if it is determined that the preventive maintenance work type is not applicable to this failure mode, a design change suggestion can be generated; at least based on the design change suggestion, the applicability judgment criteria and the effectiveness judgment criteria can be updated; based on the updated applicability judgment criteria and the updated effectiveness judgment criteria, the comprehensive prompt words can be updated, and the preventive maintenance work type of the target failure mode can be determined again using LLM based on the updated comprehensive prompt words.

[0087] The identified preventive maintenance work type can be output in a structured manner, including the preventive maintenance work type selection result (type) and all logical decision criteria, to ensure that the decision-making process is transparent and traceable. Furthermore, both the structured fault impact consequence type and the preventive maintenance work type can be stored in JSON format under the "RCMA Decision" attribute of the "cause" node in each component.

[0088] In some embodiments of this application, RCMA results for non-critical type components can also be determined.

[0089] In some implementations, all non-critical components can be traversed, and for each non-critical component, the failure impact consequence type and preventive maintenance work type of the component can be determined using LLM based on the FMEA results of that component, thereby obtaining the RCMA results for each non-critical component. For each non-critical component, the structured preventive maintenance work type in JSON format can be stored in the "Non-critical Component Maintenance Work" attribute of the component node "item". For example, the "Non-critical Component Maintenance Work" attribute of the component node "item" can store {"Selection": "Yes", "Work Type": "Functional Testing", "Work Description": "Periodically test the overload protection function and output voltage stability of the fuse to ensure that it works normally within the specified range", "Work Timing": "Periodic Inspection in Use", "Maintenance Level": "Basic Level"}.

[0090] Furthermore, the RCMA results of all important component types and all non-important component types can be combined to obtain the RCMA results of the analyzed object.

[0091] Taking the ground mobile platform-control box-drive rod circuit as the analysis object, and inputting its system component information and FMEA results, the fault impact type is first determined using the fault mode node of "signal transmission interruption or instability". The structured judgment process and basis are shown in Figure 3. Based on the selected fault impact type, the maintenance work type is determined using the fault cause node of "loose connection or poor contact in the push rod drive circuit". The structured judgment process and basis are shown in Figure 4.

[0092] In some embodiments of this application, combining the failure impact consequence type and preventive maintenance work type of each failure mode to obtain the RCMA result of this failure mode may include: traversing the failure cause nodes of each important type of component, extracting the information stored in its RCMA decision attributes, and converting the extracted information into an important type of component RCMA analysis table; merging the items in the important type of component RCMA analysis table that correspond to the same preventive maintenance work type to obtain an RCMA analysis classification work table.

[0093] The RCMA results for each non-critical component type can be determined as follows: traverse each non-critical component type, extract the information stored in its non-critical component maintenance work attributes, and convert the extracted information into a non-critical component RCMA analysis table.

[0094] The RCMA results of the analysis object can be obtained by combining the RCMA analysis classification worksheet and the RCMA analysis table of non-important components.

[0095] In other words, for critical component types, the system can automatically read JSON files containing the "RCMA Decision" attribute from the list of critical component types, iterate through each fault cause node "cause," and extract key information from the node's "RCMA Decision" attribute (such as extracting yes / no results for multiple judgment items from "Logical Decision Response," including "Whether there is a significant safety impact" and "Whether there is detectable fault degradation," etc.; extracting key decision information and decision basis from fields such as "Fault Consequences" and "Preventive Maintenance Work Types"). This generates tables such as the RCMA analysis system and equipment analysis table, the fault consequence determination and maintenance analysis table, the usage and inspection analysis table, the operator monitoring or functional testing analysis table, the scheduled disassembly / repair analysis table, and the comprehensive work analysis table. The specific execution is as follows:

[0096] First, based on the key information in the "RCMA Decision" attribute under the fault cause node "cause," including all logical decision processes and bases, it is converted into a complete analysis table containing multiple judgment items (i.e., the system and equipment analysis table for RCMA analysis). Then, according to the key information "Work Type" (maintenance, usage inspection, etc.) in the "RCMA Decision" attribute, the RCMA results are categorized into different work tables, generating a fault consequence determination and maintenance analysis table, a usage inspection analysis table, an operator monitoring or functional testing analysis table, and a scheduled disassembly / scrapping analysis table. Finally, the data from all maintenance work types are merged and transformed into a comprehensive work analysis table.

[0097] On the other hand, it can also read the JSON file of the list of non-critical components, traverse all component nodes, extract the content of the component node attribute "non-critical component maintenance work", and generate a table of non-critical functional component maintenance work.

[0098] Figure 5 This is a schematic diagram of various RCMA tables generated using the ground mobile platform-control box as the analysis object. Table 1 shows the RCMA results for the "Use Check" task of the "2-way fuses" as an example.

[0099] Table 1. RCMA Results of "Usage Inspection" for "2-Way Fuse"

[0100]

[0101] Figure 6 This is a flowchart illustrating another intelligent RCMA analysis method based on MBSE-LLM provided in this application embodiment. Figure 6 As shown, the important functional components can be determined first, including reading the component information and FMEA results in the corresponding logical model of the system. Both the component information and the FMEA results can be JSON format data. The important types of components are identified by traversing each component node in the logical model, and lists of important types of components and non-important types are generated respectively.

[0102] Then, RCMA analysis is performed, including traversing the list of important types of components to determine the types of failure impact consequences, further traversing the failure cause nodes to determine the types of preventive maintenance work, in order to generate a JSON file of the "RCMA decision" attribute of important types of components; and traversing the list of non-important types of components to generate a JSON file of RCMA maintenance work.

[0103] Finally, RCMA analysis result tables are generated, including RCMA tables for critical types of components generated using a JSON file with the "RCMA Decision" attribute, and RCMA tables for non-critical types of components generated using a JSON file with RCMA maintenance work.

[0104] The technical solution provided in this application first reads the failure modes and FMEA results of relevant components, and then identifies important functional components based on LLM to generate a list of important functional components. Next, it iterates through the FMEA results of all important functional components, performs decision-making and analysis item by item based on LLM, determines the type of failure impact consequences, and selects applicable and effective preventive maintenance work types. Simultaneously, it determines preventive maintenance work for non-critical functional components based on LLM. Finally, it generates RCMA results for the system and equipment based on the RCMA analysis results. This achieves automatic application logic decision-making to determine applicable and effective preventive maintenance work for equipment, solving the problem of low efficiency in manual analysis and providing technical support for improving the efficiency of preventive maintenance analysis.

[0105] The equipment involved in the embodiments of this application may include aviation equipment, aerospace equipment, weaponry equipment, marine equipment, electronic equipment, nuclear equipment, new energy vehicle equipment, high-speed rail equipment, or other equipment.

[0106] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0107] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0108] Figure 7 This is a schematic diagram of an intelligent RCMA analysis device based on MBSE-LLM provided in an embodiment of this application. Figure 7 As shown, the device includes:

[0109] Module 701 is configured to construct a logical model of the analysis object using the Model-Based Systems Engineering (MBSE) approach; the logical model includes at least hierarchical entity components of the analysis object.

[0110] The acquisition module 702 is configured to acquire component information and Failure Mode and Effects Analysis (FMEA) results for each component in the logical model; the component information includes at least a functional description of the component, and the FMEA results include at least the failure mode, failure cause, failure effect, and severity level of the component.

[0111] The classification module 703 is configured to traverse all components in the logical model and use the Large Language Model (LLM) to determine the component type based on the functional description of each component and the FMEA results; the component type includes important types and non-important types.

[0112] Analysis module 704 is configured to traverse all important type components, traverse all failure modes of each important type component, and for each failure mode, use LLM to determine the failure impact consequence type by combining the failure impact and severity level corresponding to this failure mode.

[0113] Analysis module 704 is also configured to, for each failure mode, traverse its corresponding failure causes and use LLM to determine the type of preventive maintenance work based on the failure cause and the type of failure impact consequences.

[0114] The combination module 705 is configured to combine the failure impact consequence type and preventive maintenance work type of each failure mode to obtain the RCMA result of this failure mode, and then obtain the RCMA result of each important type of component.

[0115] According to the technical solution provided in the embodiments of this application, a logical model of the analysis object is constructed using the MBSE method. For each component in the logical model, the component type is determined using LLM based on its component information and FMEA results. For important component types, LLM first determines the failure impact consequence type, then determines the preventive maintenance work type based on the failure cause and failure impact consequence type. Finally, the failure impact consequence type and preventive maintenance work type are combined to obtain the RCMA results for each important component type. This method performs maintenance analysis based on a large model with reliability as the core, achieving intelligent classification of failure impact types, precise matching of maintenance work, and automated generation of RCMA analysis, significantly improving the efficiency and accuracy of RCMA analysis and providing a scientific and objective decision-making basis for equipment maintenance strategy formulation.

[0116] In some implementations, component information also includes the component code and component name of the component; the component type is determined using LLM based on the functional description of each component and FMEA results, including: combining the functional description of the target component with the fault impact and severity level in its FMEA results to obtain the fault description of the target component; wherein, the target component is any component in the logical model; the fault description is used as a prompt to guide LLM to determine the component type of the target component; based on the determined component types, lists of important types of components and lists of non-important types of components are constructed respectively; wherein, each component list includes the component code, component name, functional description, and fault mode of the component of this type.

[0117] In some implementations, LLM is used to determine the type of failure impact consequences by combining the failure impact and severity level corresponding to the failure mode. This includes: generating a failure description of the failure mode; the failure description includes the component name corresponding to the failure mode, the functional description of the component corresponding to the failure mode, the failure mode, the failure impact, and the severity level; and guiding LLM to perform preset logic judgments based on the failure description to determine the type of failure impact consequences of the failure mode. The preset logic judgments include at least one of the following: whether the discovery of the functional failure corresponding to the failure mode is obvious to operators using the equipment normally; whether the functional failure corresponding to the failure mode or the secondary losses caused by the functional failure have a direct harmful impact on operational safety; whether the functional failure corresponding to the failure mode has a direct harmful impact on task completion; whether the combination of the hidden functional failure corresponding to the failure mode or other related or backup functional failures of the component has a harmful impact on operational safety; or whether the combination of the hidden functional failure corresponding to the failure mode or other related or backup functional failures of other components has a harmful impact on task completion.

[0118] In some implementations, the failure impact consequence types include at least safety impact types, task-related impact types, and economic impact types. LLM is used to determine the type of preventative maintenance work based on the failure cause and failure impact consequence types, including: determining a decision path based on the failure impact consequence type of the target failure mode; wherein, the target failure mode is any failure mode; the failure impact consequences of safety impact type, task-related impact type, and economic impact type correspond to different decision paths; using LLM to perform applicability analysis on each preventative maintenance work in the decision path to obtain applicability judgment criteria; using LLM to perform effectiveness analysis on each preventative maintenance work in the decision path to obtain effectiveness judgment criteria; combining the applicability judgment criteria, effectiveness judgment criteria, and the definitions of each preventative maintenance work in the decision path to obtain comprehensive prompt words; and using LLM to determine the type of preventative maintenance work for the target failure mode based on the comprehensive prompt words.

[0119] In some implementations, after determining the type of preventive maintenance work, the method further includes: generating a design change recommendation in response to determining that the type of preventive maintenance work is not applicable to the target failure mode; updating the applicability and validity criteria based at least on the design change recommendation; updating the comprehensive prompt based on the updated applicability and validity criteria; and using LLM to re-determine the type of preventive maintenance work for the target failure mode based on the updated comprehensive prompt.

[0120] In some implementations, after determining the component type, the process further includes: traversing all non-critical components, using LLM to determine the failure impact consequence type and preventive maintenance work type of each non-critical component based on the FMEA results of that component, thereby obtaining the RCMA results of each non-critical component; and combining the RCMA results of all critical components and all non-critical components to obtain the RCMA results of the analysis object.

[0121] In some implementations, for each critical type component, the failure impact consequence type and preventive maintenance work type corresponding to the failure cause are stored in the RCMA decision attribute of each failure cause node of the component; and for each non-critical type component, the preventive maintenance work type of the component is stored in the non-critical component maintenance work attribute of the component node; combining the failure impact consequence type and preventive maintenance work type of each failure mode to obtain the RCMA result of the failure mode includes: traversing the failure cause nodes of each critical type component, extracting the information stored in its RCMA decision attribute, and converting the extracted information into an RCMA analysis table for critical type components. Merge items with the same preventive maintenance work type in the RCMA analysis tables of important component types to obtain an RCMA analysis classification work table; the RCMA results of each non-important component type are determined as follows: traverse each non-important component type, extract the information stored in its non-important component maintenance work attributes, and convert the extracted information into a non-important component RCMA analysis table; combine the RCMA results of all important component types and all non-important component types to obtain the RCMA results of the analysis object, including: combining the RCMA analysis classification work table and the non-important component RCMA analysis table to obtain the RCMA results of the analysis object.

[0122] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0123] Figure 8 This is a schematic diagram of an electronic device provided in an embodiment of this application. Figure 8 As shown, the electronic device 8 of this embodiment includes a processor 801, a memory 802, and a computer program 803 stored in the memory 802 and executable on the processor 801. When the processor 801 executes the computer program 803, it implements the steps in the various method embodiments described above. Alternatively, when the processor 801 executes the computer program 803, it implements the functions of each module / unit in the various device embodiments described above.

[0124] Electronic device 8 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 8 may include, but is not limited to, processor 801 and memory 802. Those skilled in the art will understand that... Figure 8 This is merely an example of electronic device 8 and does not constitute a limitation on electronic device 8. It may include more or fewer components than shown, or different components.

[0125] The processor 801 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0126] The memory 802 can be an internal storage unit of the electronic device 8, such as a hard disk or RAM of the electronic device 8. The memory 802 can also be an external storage device of the electronic device 8, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the electronic device 8. The memory 802 can also include both internal and external storage units of the electronic device 8. The memory 802 is used to store computer programs and other programs and data required by the electronic device.

[0127] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0128] If integrated modules / units are implemented as software functional units and sold or used as independent components, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0129] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A smart RCMA analysis method based on MBSE-LLM, used to determine applicable and effective preventive maintenance work for equipment; characterized in that, The method includes: The logical model of the analysis object is constructed using the Model-Based Systems Engineering (MBSE) method. Specifically, the logical model for constructing the analysis object is as follows: Using the MBSE method, the system of the analysis object is analyzed based on its actual hardware architecture. Divided into multiple logical levels ; among them, each layer It includes a set of components that make up this layer. , Indicates the first The first in the layer Each component For the first The total number of components in the layer; Within each defined level, identify and create all entity component nodes that constitute that level, and determine the components. attribute tuples ;in, A unique identifier for the component. For component name, For component functionality; all components are stored in the structured data format JSON; Establish connections between components; these connections consist of port definitions and connection lines; wherein, the port set is... Each port Must be with one and only one component Associated, with a unique port identifier Port type and port direction The set of connection relationships is Each connection It is a directed edge from one output port to one input port. Connection types include matter, information, and energy; text attributes. This provides a supplementary explanation of the connection relationship, noting that the connection lines have directionality. Indicates Boolean type, Indicates integer. Indicates floating-point type, Represents a character type; Indicates the input port type. Indicates the output port type; It is a positive integer; express A unique identifier, express input port, express The output port, express Connection type; Determine the set of nodes and edge set The directed graph formed The data is stored in the structured data format JSON to obtain the logical model of the analyzed object. Obtain the component information and Failure Mode and Effects Analysis (FMEA) results for each component in the logical model; the component information includes at least a functional description of the component, and the FMEA results include at least the failure mode, failure cause, failure effect, and severity level of the component; Traverse all components in the logical model and use the Large Language Model (LLM) to determine the component type based on the functional description of each component and the FMEA results; wherein, the component type includes important types and non-important types; Iterate through all important component types, and for each important component type, iterate through all failure modes of this component; for each failure mode, use the LLM to determine the failure impact consequence type by combining the failure impact and severity level corresponding to this failure mode. For each failure mode, iterate through its corresponding failure causes and use the LLM to determine the type of preventive maintenance work based on the failure causes and the type of failure impact consequences; By combining the failure impact consequence type and preventive maintenance work type of each failure mode, the RCMA result of this failure mode is obtained, and then the RCMA result of each important type of component is obtained.

2. The method according to claim 1, characterized in that, The component information also includes the component code and component name of this component; Component types are determined using LLM based on the functional descriptions of each component and FMEA results, including: The fault description of the target component is obtained by combining the functional description of the target component with the fault impact and severity level in its FMEA results; wherein, the target component is any component in the logical model; The fault description is used as a prompt word to guide the LLM in determining the component type of the target component; Based on the determined component types, construct lists of important and unimportant components respectively; each component list includes the component code, component name, functional description, and failure mode of the component type.

3. The method according to claim 1, characterized in that, The failure impact consequence type is determined by combining the LLM with the failure impact and severity level corresponding to this failure mode, including: The failure description of the cost failure mode includes the component name corresponding to the failure mode, the functional description of the component corresponding to the failure mode, the failure mode, the failure impact and the severity level. Based on the fault description, the LLM is guided to perform a preset logic judgment to determine the type of fault impact consequences of this fault mode; The preset logic judgment includes at least one of the following: whether the discovery of the functional failure corresponding to this failure mode is obvious to the operator of the equipment under normal use; whether the functional failure corresponding to this failure mode or the secondary loss caused by the functional failure has a direct harmful impact on the safety of use; whether the functional failure corresponding to this failure mode has a direct harmful impact on the completion of the task; whether the combination of the hidden functional failure corresponding to this failure mode or other related or backup functional failures of this component has a harmful impact on the safety of use; or whether the combination of the hidden functional failure corresponding to this failure mode or other related or backup functional failures of other components has a harmful impact on the completion of the task.

4. The method according to claim 1, characterized in that, The types of consequences of the failure include at least safety impact types, mission-related impact types, and economic impact types; The LLM is used to determine the type of preventive maintenance work based on the cause of the failure and the type of the consequences of the failure, including: Decision paths are determined based on the failure impact consequences type of the target failure mode; where the target failure mode is any failure mode; the failure impact consequences of safety impact type, the failure impact consequences of task impact type, and the failure impact consequences of economic impact type correspond to different decision paths; The LLM is used to perform an applicability analysis on each preventive maintenance task in the decision path to obtain the applicability judgment criteria; The effectiveness of each preventive maintenance task in the decision-making path is analyzed using the LLM to obtain the basis for effectiveness judgment; The applicability judgment criteria, the validity judgment criteria, and the definitions of each preventive maintenance task in the decision path are combined to obtain a comprehensive prompt word. The LLM is then used to determine the preventive maintenance task type for the target failure mode based on the comprehensive prompt word.

5. The method according to claim 4, characterized in that, After determining the type of preventative maintenance work, the method further includes: In response to determining that the preventive maintenance work type is not applicable to the target failure mode, a design change suggestion is generated; The applicability determination criteria and the validity determination criteria shall be updated at least based on the proposed design changes; The comprehensive prompt words are updated based on the updated applicability and validity criteria, and the preventive maintenance work type for the target failure mode is determined again using the LLM based on the updated comprehensive prompt words.

6. The method according to claim 1, characterized in that, After determining the component type, the method further includes: Traverse all non-critical components, and for each non-critical component, use the LLM based on the FMEA results of this component to determine the failure impact consequence type and preventive maintenance work type of this component, and then obtain the RCMA results of each non-critical component. The RCMA results of the analyzed object are obtained by combining the RCMA results of all important types of components and all non-important types of components.

7. The method according to claim 6, characterized in that, For each critical type of component, the failure impact consequence type and preventive maintenance work type corresponding to the failure cause are stored in the RCMA decision attribute of each failure cause node of this component; and for each non-critical type of component, the preventive maintenance work type of this component is stored in the non-critical component maintenance work attribute of this component node. By combining the failure impact consequence types and preventive maintenance activity types of each failure mode, the RCMA results for this failure mode are obtained, including: Traverse the fault cause nodes of each important type of component, extract the information stored in its RCMA decision attributes, and convert the extracted information into an RCMA analysis table for important type components; Merge the items in the RCMA analysis table of the important component types that correspond to the same preventive maintenance work type to obtain the RCMA analysis classification work table; The RCMA results for each non-critical type of component are determined as follows: Traverse each non-critical component type, extract the information stored in the maintenance work attributes of the non-critical component, and convert the extracted information into an RCMA analysis table for non-critical components; The RCMA results of the analyzed object are obtained by combining the RCMA results of all significant component types and all non-significant component types, including: By combining the RCMA analysis classification worksheet and the non-significant type component RCMA analysis table, the RCMA results of the analyzed object are obtained.

8. An intelligent RCMA analysis device based on MBSE-LLM, used to determine applicable and effective preventive maintenance work for equipment; characterized in that, The device includes: The building module is configured to construct a logical model of the analysis object using the Model-Based Systems Engineering (MBSE) approach. Specifically, the logical model for constructing the analysis object is as follows: Using the MBSE method, the system of the analysis object is analyzed based on its actual hardware architecture. Divided into multiple logical levels ; among them, each layer It includes a set of components that make up this layer. , Indicates the first The first in the layer Each component For the first The total number of components in the layer; Within each defined level, identify and create all entity component nodes that constitute that level, and determine the components. attribute tuples ;in, A unique identifier for the component. For component name, For component functionality; all components are stored in the structured data format JSON; Establish connections between components; these connections consist of port definitions and connection lines; wherein, the port set is... Each port Must be with one and only one component Associated, with a unique port identifier Port type and port direction The set of connection relationships is Each connection It is a directed edge from one output port to one input port. Connection types include matter, information, and energy; text attributes. This provides a supplementary explanation of the connection relationship, noting that the connection lines have directionality. Indicates Boolean type, Indicates integer. Indicates floating-point type, Represents a character type; Indicates the input port type. Indicates the output port type; It is a positive integer; express A unique identifier, express input port, express The output port, express Connection type; Determine the set of nodes and edge set The directed graph formed The data is stored in the structured data format JSON to obtain the logical model of the analyzed object. The acquisition module is configured to acquire component information and Failure Mode and Effects Analysis (FMEA) results for each component in the logical model; the component information includes at least a functional description of the component, and the FMEA results include at least the failure mode, failure cause, failure effect, and severity level of the component. The classification module is configured to traverse all components in the logical model and determine the component type using the Large Language Model (LLM) based on the functional description of each component and the FMEA results; wherein, the component type includes important types and non-important types; The analysis module is configured to traverse all important type components, traverse all failure modes of each important type component, and for each failure mode, use the LLM to determine the failure impact consequence type by combining the failure impact and severity level corresponding to the failure mode. The analysis module is also configured to, for each fault mode, traverse its corresponding fault causes and use the LLM to determine the type of preventive maintenance work based on the fault causes and the type of fault impact consequences. The combination module is configured to combine the failure impact consequence type and preventive maintenance work type of each failure mode to obtain the RCMA result of this failure mode, and then obtain the RCMA result of each important type of component.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images