A hydrogen fuel cell reliability evaluation method and system under complex working conditions
Patent Information
- Application Number
- CN202611257688.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-19
- Publication Date
- 2026-10-20
AI Technical Summary
[0004]本发明的目的是解决现有评价方法多基于单一故障模式或部件特性，缺乏系统性、多因素综合建模手段的技术问题，而提供一种氢燃料电池可靠性评价方法及系统
1、本发明提供的复杂工况下氢燃料电池可靠性评价方法及系统，通过采用FMEA方法获取氢燃料电池各故障模式、失效原因和关键零部件的折中排序，再以故障模式、失效原因、关键零部件数据的影响关系矩阵，以及折中排序构建氢燃料电池的可靠性分析模型，通过可靠性分析模型与VIKOR算法结合获取氢燃料电池的可靠性评分；且可靠性分析模型涵盖了故障模式与失效原因、关键零部件之间、失效原因之间、关键零部件之间、故障模式之间的关联关系，相较于现有针对单一故障模式或部件特性的评价方法，可以实现对不同故障模式、失效原因、关键零部件进行评价，使评价结果更加全面、真实地反映氢燃料电池在复杂工况下的整体可靠性状态。
Smart Images
Figure FT_1 
Figure SMS_104 
Figure SMS_119
Abstract
Description
Technical Field
[0001] This invention relates to the field of hydrogen fuel cell data analysis technology, specifically to a method and system for evaluating the reliability of hydrogen fuel cells under complex operating conditions. Background Technology
[0002] With the rapid development of the hydrogen energy industry, hydrogen fuel cells, as a new generation of clean energy power systems, have broad application prospects due to their high power density and fast response. However, they suffer from poor operational stability and high failure rate under complex operating conditions such as high temperature, high pressure, and rapid charging and discharging, posing challenges to system safety.
[0003] Currently, most evaluation methods are based on single failure modes or component characteristics, lacking systematic and multi-factor comprehensive modeling methods. Summary of the Invention
[0004] The purpose of this invention is to address the technical problem that existing evaluation methods are mostly based on single failure modes or component characteristics and lack systematic and multi-factor comprehensive modeling methods, and to provide a hydrogen fuel cell reliability evaluation method and system.
[0005] To solve the above-mentioned technical problems, the technical solution provided by the present invention is as follows: A method for evaluating the reliability of hydrogen fuel cells under complex operating conditions includes the following steps: S1. Collect key indicator data of hydrogen fuel cells under typical operating conditions, and establish a matrix of the relationship between failure modes, failure causes and key components within the key indicator data. S2. Use the FMEA method to obtain the trade-off ranking of each failure mode, failure cause, and key components; S3. A reliability analysis model for hydrogen fuel cells is constructed by combining the impact relationship matrix of failure modes, failure causes, and key component data, as well as a compromise ranking. S4. Set the reliability threshold for the reliability index data within the key index data, and establish a set of drivers based on the measured values of the reliability index data and the reliability threshold. The total reliability fluctuation of hydrogen fuel cells caused by the set of driving factors is calculated by a reliability analysis model, and the VIKOR algorithm is used to map the total reliability fluctuation and the compromise ranking to the interval (0,1) as the reliability score of hydrogen fuel cells. S5. Map the compromise ranking, total reliability fluctuation, reliability score of hydrogen fuel cell, and influence relationship matrix into a visual chart and output it as a file to complete the reliability evaluation of hydrogen fuel cell.
[0006] Furthermore, in step S1, the key indicator data includes typical failure modes, failure causes and frequencies, reliability indicator data, key component data of hydrogen fuel cells, and data on the influence relationships between hydrogen fuel cell components, failure modes and failure causes.
[0007] Furthermore, step S1 specifically includes: S11. Collect data on hydrogen fuel cells under typical operating conditions. A typical failure mode, One cause of failure Key components, failure causes, and frequency of key component failures Frequency of failure modes caused by component failure and reliability index data And store them in a two-dimensional relational table format; in: Indicates the i-th cause of failure; , These represent the i-th and j-th key components, respectively. This represents the j-th fault mode; S12. Based on historical failure data, determine the correlations between failure causes, failure modes, and key components. Establish corresponding tables using the HoR cross-correlation matrix natural language value-to-numerical conversion table and the HoR autocorrelation matrix scoring criteria. Failure cause influence relationship matrix Failure Mode and Effect Matrix and The key component influence relationship matrix.
[0008] Furthermore, step S2 specifically includes: S21. Establish a natural description language terminology set and corresponding quantitative fuzzy numbers for each risk factor in the FMEA method, as the quantitative evaluation criteria for each risk factor. Risk factors include severity (S), incidence (O), and difficulty of detection (D); S22. Based on the quantitative evaluation criteria corresponding to each risk factor, score the failure modes, failure causes, and risk factors of key components of hydrogen fuel cells. The scores corresponding to each risk factor are summarized as the original dataset of risk factor scores for hydrogen fuel cell failure modes, failure causes, and key components. S23. Using rough set theory, the score values in the original dataset are transformed into interval numbers, and the interval numbers are fused to obtain the lower bound of the interval numbers for each failure cause, failure mode, and key component under various risk factors. Upper bound of the sum of interval numbers ,in: ; ; In the formula, For the kth failure cause, failure mode, or critical component; for The number of objects included in the lower approximation; This is the score for the kth failure cause, failure mode, or critical component. for The lower approximate set; for The number of objects included in the upper approximation; for The upper approximate set; Calculate the number of fusion intervals [ , ],in: ; ; In the formula, To be the lower bound of the number of fusion intervals, for The corresponding lower bound of the interval number; To be the upper bound of the number of fusion intervals, for The corresponding upper bound of the interval number; S24. The optimal values of each failure mode, failure cause, and key component in each quantitative evaluation criterion are taken as the ideal solution. The worst values of each failure mode, failure cause, and key component in each quantitative evaluation criterion are taken as the negative ideal solution. ; S25. Calculate the distance between the roughness number and the ideal solution and negative ideal solution for each failure mode, failure cause, and key component under each risk factor in the evaluation interval. : ; In the formula, [ , [Is the] number The failure mode is related to the first Rough score of each risk factor; S26, Combining the number of fusion intervals [ , Calculate the weights of each failure mode, failure cause, and key component. ; And by analyzing the weights of each failure mode, failure cause, and key components. Calculate the corresponding group utility maximization ranking Individual Regret Minimization Sort ; ; Ranking based on the maximum group utility of each failure mode, failure cause, and critical components. Individual Regret Minimization Sort Calculate the corresponding compromise sort : ; In the formula, , , ; The weight of the "majority criterion" strategy or the maximum group utility weight; When, it means that a decision is made based on the opinion of the majority; When, it means that a decision is made based on the level of agreement; When this is the case, it indicates that a decision will be made based on the circumstances of refusal. Compromise sorting Rank the failure modes, failure causes, and severity of critical components. Furthermore, in step S3, the reliability analysis model is: RM = <V,R,W>; In the formula, V is the set of all nodes in the reliability analysis model, and V = { }, It serves as the central node of the reliability analysis model; These are nodes representing reliability indices, failure causes, key components, and failure modes in the reliability analysis model. R is the set of all relation matrices in the reliability analysis model, and R = { }; The weight matrix of the nodes causing the failure; This is a weight matrix representing the correlation between failure causes; This is a weighted matrix representing the correlation between failure causes and key components; The weight matrix for key component nodes; This is a weight matrix representing the relationships between key components. This is a weighted matrix representing the correlation between key components and failure modes. The weight matrix for the fault mode nodes; This is a weight matrix representing the correlation between failure modes; This is the weight matrix representing the correlation between failure modes and the model's central node; The method for determining each matrix in the set R of all relation matrices in the reliability analysis model is as follows: Based on key performance indicator data for hydrogen fuel cells, the presence and frequency of failures of critical components caused by failure factors are used as the basis for determining the correlation matrix and weight matrix between failure factors and critical components in the reliability analysis model; the presence and frequency of failure modes caused by critical component failures are used as the basis for determining the correlation matrix and weight matrix between the critical component layer and failure modes in the reliability analysis model; the correlation matrix includes the correlation matrix between reliability indicators and failure factors. Correlation matrix between failure causes and failure modes Correlation matrix between key components and failure modes And the matrix of relationships between key components and systems. The weight matrix includes the weights between reliability metrics and failure causes. Weight matrix of the relationship between failure causes and key components Weight matrix of correlation between key components and failure modes And the matrix of relationships between key components and systems. ; Based on the key performance data of hydrogen fuel cells, as well as the correlation matrix between failure causes, the correlation data between key components, and the correlation data between failure modes, a correlation matrix between failure causes was determined. ; Correlation matrix between failure modes Relationship matrix between key components ; The weight matrix of the failure cause node is determined by a compromise ranking based on failure cause, critical components, and failure mode severity. Weight matrix of key component nodes Weight matrix of fault mode nodes .
[0009] Furthermore, step S4 specifically includes: S41. Based on historical operating data, experiments, and expert prior knowledge, set the normal range of reliability index data as the reliability threshold. Calculate the distance between the measured value of the reliability index data and the reliability threshold. If the distance is greater than the safe distance, then the reliability index data is... Add to the set of motivations ; ; In the formula, The distance between the measured value of the reliability index data and the reliability threshold is generally calculated using a membership function; This is the safe distance at which the indicator takes effect without causing failure. S42, Traversing the set of motivations Each element in A breadth-first search strategy is used to find all nodes in the reliability analysis model that reach the central node. The pathways, all of which constitute the cut set of the reliability analysis model; S43. Calculate the set of drivers based on the cut set of the reliability analysis model. Total fluctuation in system reliability caused by propagation along all pathways: The VIKOR algorithm is used to rank the total reliability fluctuations and trade-offs. Mapped to (0, 1), it serves as a score for the reliability of the hydrogen fuel cell.
[0010] A reliability evaluation system for hydrogen fuel cells under complex operating conditions includes a data acquisition module, a model building module, a reliability assessment module, and a graphical risk display module connected sequentially via a data bus or API. The model building module combines various risk factors in the FMEA method to obtain a trade-off ranking of each failure mode, failure cause, and key component, and combines key indicator parameters and influence relationship matrices collected by the data acquisition module to build a reliability analysis model; the reliability assessment module combines the reliability analysis model and trade-off ranking to output a reliability score for the hydrogen fuel cell. Finally, the graphical risk display module outputs visual charts in a visual manner, including trade-off ranking, total reliability fluctuation, hydrogen fuel cell reliability scores, and impact relationship matrix mappings.
[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The reliability evaluation method and system for hydrogen fuel cells under complex operating conditions provided by this invention obtains a compromise ranking of various failure modes, failure causes, and key components of the hydrogen fuel cell using the FMEA method. Then, a reliability analysis model for the hydrogen fuel cell is constructed using the influence relationship matrix of failure modes, failure causes, and key component data, as well as the compromise ranking. The reliability score of the hydrogen fuel cell is obtained by combining the reliability analysis model with the VIKOR algorithm. Moreover, the reliability analysis model covers the correlation between failure modes and failure causes, key components, failure causes, key components, and failure modes. Compared with existing evaluation methods that target a single failure mode or component characteristic, it can evaluate different failure modes, failure causes, and key components, making the evaluation results more comprehensive and realistic in reflecting the overall reliability status of the hydrogen fuel cell under complex operating conditions.
[0012] 2. The present invention provides a method and system for evaluating the reliability of hydrogen fuel cells under complex operating conditions. The present invention introduces a set of driving factors and a cut set. By setting a reliability threshold and calculating the distance between the measured value of the reliability index data and the reliability threshold, the set of driving factors is adjusted. Then, a breadth-first search strategy is used to traverse all paths to the central node in the reliability analysis model and form a cut set. On this basis, the total fluctuation of system reliability caused by the propagation of the set of driving factors along all paths is calculated. This allows the present invention to capture the chain risk of "one component failure triggering multiple failure modes" in hydrogen fuel cells under complex operating conditions, which is something that traditional single failure mode evaluation methods cannot achieve.
[0013] 3. The present invention provides a method and system for evaluating the reliability of hydrogen fuel cells under complex operating conditions. This invention utilizes the VIKOR algorithm to uniformly map the total reliability fluctuation and compromise ranking to the (0, 1) interval, obtaining a reliability score for the hydrogen fuel cell. This makes the reliability evaluation results of different fuel cell systems and under different operating conditions directly comparable. The closer the score is to 1, the higher the reliability; the closer it is to 0, the higher the risk. This provides an intuitive and quantitative evaluation benchmark for engineering maintenance decisions. Simultaneously, this invention uses a graphical risk display module to output the compromise ranking, total reliability fluctuation, reliability score, and influence relationship matrix in the form of visual charts, transforming complex multidimensional data into intuitive decision support information, significantly improving the efficiency and accuracy of fault location and maintenance decisions. Attached Figure Description
[0014] Figure 1 This is a schematic diagram of the process structure of an embodiment of the present invention. Detailed Implementation
[0015] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0016] like Figure 1 As shown, a method for evaluating the reliability of hydrogen fuel cells under complex operating conditions includes the following steps: (1) Collect key indicator data of hydrogen fuel cells under typical operating conditions, and establish a matrix of the relationship between failure modes, failure causes and key components within the key indicator data; Key performance indicators (KPIs) include typical failure modes, failure causes and frequencies, reliability indicators, key components of hydrogen fuel cells, and data on the influence relationships between hydrogen fuel cell components, failure modes, and failure causes.
[0017] Collect hydrogen fuel cell data under typical operating conditions A typical failure mode, One cause of failure Key components, failure causes, and frequency of key component failures Frequency of failure modes caused by component failure and reliability index data And store them in a two-dimensional relational table format; in: Indicates the i-th cause of failure; , These represent the i-th and j-th key components, respectively. This represents the j-th fault mode; Based on historical failure data, the correlations between failure causes, failure modes, and key components are determined. A HoR cross-correlation matrix natural language value-to-numerical conversion table and a HoR autocorrelation matrix scoring standard are then established. Failure cause influence relationship matrix Failure Mode and Effect Matrix and The key component influence relationship matrix. The HoR cross-correlation matrix natural language value-numerical conversion table and the HoR autocorrelation matrix scoring criteria are shown in Tables 1 and 2. Table 1: HoR Cross-Correlation Matrix Natural Language Value-Numerical Transformation Table Table 2: HoR Autocorrelation Matrix Scoring Criteria (2) Establish a set of natural description language terms and corresponding quantitative fuzzy numbers for each risk factor in the FMEA method, which can be used as the quantitative evaluation criteria for each risk factor; Risk factors include severity (S), incidence (O), and difficulty of detection (D); the quantitative evaluation criteria for severity (S), incidence (O), and difficulty of detection (D) are shown in Tables 3 to 5.
[0018] Table 3: Severity S Quantitative Evaluation Criteria Table 4: Quantitative Evaluation Criteria for Occurrence Rate Table 5: Quantitative Evaluation Criteria for Detection Difficulty D Based on the quantitative evaluation criteria corresponding to each risk factor, the failure modes, failure causes, and risk factors of key components of hydrogen fuel cells are scored. The scores corresponding to each risk factor are summarized as the original dataset of risk factor scores for hydrogen fuel cell failure modes, failure causes, and key components. Using rough set theory, the score values in the original dataset are transformed into interval numbers, and these interval numbers are then fused to obtain the lower bounds of the interval numbers for each failure cause, failure mode, and key component under various risk factors. Upper bound of the sum of interval numbers ,in: ; ; In the formula, For the kth failure cause, failure mode, or critical component; for The number of objects included in the lower approximation; This is the score for the kth failure cause, failure mode, or critical component. for The lower approximate set; for The number of objects included in the upper approximation; for The upper approximate set; Calculate the number of fusion intervals [ , ],in: ; ; In the formula, To be the lower bound of the number of fusion intervals, for The corresponding lower bound of the interval number; To be the upper bound of the number of fusion intervals, for The corresponding upper bound of the interval number; The optimal values of each failure mode, failure cause, and key component in each quantitative evaluation criterion are taken as the ideal solution. The worst values of each failure mode, failure cause, and key component in each quantitative evaluation criterion are taken as the negative ideal solution. ; Calculate the distance between the rough number and the ideal and negative ideal solutions for each failure mode, failure cause, and key component under various risk factors in the evaluation interval. : ; In the formula, [ , [Is the] number The failure mode is related to the first Rough score of each risk factor; Combined with the number of fusion intervals [ , Calculate the weights of each failure mode, failure cause, and key component. ; And by analyzing the weights of each failure mode, failure cause, and key components. Calculate the corresponding group utility maximization ranking Individual Regret Minimization Sort ; ; Ranking based on the maximum group utility of each failure mode, failure cause, and critical components. Individual Regret Minimization Sort Calculate the corresponding compromise sort : ; In the formula, , , ; The weight of the "majority criterion" strategy or the maximum group utility weight; When, it means that a decision is made based on the opinion of the majority; When, it means that a decision is made based on the level of agreement; When this is the case, it indicates that a decision will be made based on the circumstances of refusal. Compromise sorting The failure modes, causes of failure, and severity of critical components were ranked and then sorted using a compromise approach. The smaller the value, the higher the severity of the failure mode.
[0019] In the FMEA method, the risk priority number RPN is: In the formula, S represents severity, O represents incidence, D represents detection difficulty, C represents critical factor, and f represents a weighted nonlinear combination function.
[0020] (3) Construct a reliability analysis model for hydrogen fuel cells by combining the impact relationship matrix of failure modes, failure causes, and key component data, as well as the compromise ranking; Specifically, the reliability analysis model is: RM = <V,R,W>; In the formula, V is the set of all nodes in the reliability analysis model, and V = { }, It serves as the central node of the reliability analysis model; These are nodes representing reliability indices, failure causes, key components, and failure modes in the reliability analysis model. R is the set of all relation matrices in the reliability analysis model, and R = { }; The weight matrix of the nodes causing the failure; This is a weight matrix representing the correlation between failure causes; This is a weighted matrix representing the correlation between failure causes and key components; The weight matrix for key component nodes; This is a weight matrix representing the relationships between key components. This is a weighted matrix representing the correlation between key components and failure modes. The weight matrix for the fault mode nodes; This is a weight matrix representing the correlation between failure modes; This is the weight matrix representing the correlation between failure modes and the model's central node; The method for determining each matrix in the set R of all relation matrices in the reliability analysis model is as follows: Based on key performance indicator data for hydrogen fuel cells, the presence and frequency of failures of critical components caused by failure factors are used as the basis for determining the correlation matrix and weight matrix between failure factors and critical components in the reliability analysis model; the presence and frequency of failure modes caused by critical component failures are used as the basis for determining the correlation matrix and weight matrix between the critical component layer and failure modes in the reliability analysis model; the correlation matrix includes the correlation matrix between reliability indicators and failure factors. Correlation matrix between failure causes and failure modes Correlation matrix between key components and failure modes And the matrix of relationships between key components and systems. The weight matrix includes the weights between reliability metrics and failure causes. Weight matrix of the relationship between failure causes and key components Weight matrix of correlation between key components and failure modes And the matrix of relationships between key components and systems. ; Based on the key performance data of hydrogen fuel cells, as well as the correlation matrix between failure causes, the correlation data between key components, and the correlation data between failure modes, a correlation matrix between failure causes was determined. ; Correlation matrix between failure modes Relationship matrix between key components ; The weight matrix of the failure cause node is determined by a compromise ranking based on failure cause, critical components, and failure mode severity. Weight matrix of key component nodes Weight matrix of fault mode nodes ; (4) Based on historical operating data, experiments and expert prior knowledge, the normal range of reliability index data is set as the reliability threshold. Calculate the distance between the measured value of the reliability index data and the reliability threshold. If the distance is greater than the safe distance, then the reliability index data is... Add to the set of motivations ; ; In the formula, The distance between the measured value of the reliability index data and the reliability threshold is generally calculated using a membership function; This is the safe distance at which the indicator takes effect without causing failure. Traversing the set of motivations Each element in A breadth-first search strategy is used to find all nodes in the reliability analysis model that reach the central node. The pathways, all of which constitute the cut set of the reliability analysis model; Based on the cut set of the reliability analysis model, calculate the set of drivers. Total fluctuation in system reliability caused by propagation along all pathways: The VIKOR algorithm is used to rank the total reliability fluctuations and trade-offs. Mapped to (0, 1), serving as a score for the reliability of the hydrogen fuel cell; The closer the score is to 1, the higher the reliability; the closer it is to 0, the higher the risk. This provides an intuitive and quantitative evaluation benchmark for engineering maintenance decisions.
[0021] (5) Map the compromise ranking, total reliability fluctuation, reliability score of hydrogen fuel cell and influence relationship matrix into a visual chart and output it as a file to complete the reliability evaluation of hydrogen fuel cell.
[0022] Based on the above-mentioned reliability evaluation method for hydrogen fuel cells under complex operating conditions, the present invention also provides a reliability evaluation system for hydrogen fuel cells under complex operating conditions, including a data acquisition module, a model building module, a reliability assessment module and a graphical risk display module connected sequentially via a data bus or API. The model building module combines various risk factors in the FMEA method to obtain a trade-off ranking of each failure mode, failure cause, and key component, and combines key indicator parameters and influence relationship matrices collected by the data acquisition module to build a reliability analysis model; the reliability assessment module combines the reliability analysis model and trade-off ranking to output a reliability score for the hydrogen fuel cell. Finally, the graphical risk display module outputs visual charts in a visual manner, including trade-off ranking, total reliability fluctuation, hydrogen fuel cell reliability scores, and impact relationship matrix mappings.
[0023] Specific implementation examples are given below.
[0024] Taking a certain model of 300kW hydrogen fuel cell as an example: Step 1: Collect real-time data during operation (sampling frequency of 10Hz) to form a parameter matrix including temperature, voltage, and current; Step 2: Combine historical fault records and experimental test results to construct an FMEA table and introduce key factor scores; Step 3: Establish a reliability analysis model based on the thermal and electrical coupling relationships between modules; Step 4: Use the VIKOR algorithm (multi-objective decision algorithm) to generate a reliability score for the hydrogen fuel cell; Step 5: The results show that the anode diffusion layer and electrolyte membrane are high-risk components, and the reliability score of the hydrogen fuel cell is 0.81. Recommendations for optimizing the cooling structure are proposed.
[0025] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present invention should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for evaluating the reliability of hydrogen fuel cells under complex operating conditions, characterized in that, Includes the following steps: S1. Collect key indicator data of hydrogen fuel cells under typical operating conditions, and establish a matrix of the relationship between failure modes, failure causes and key components within the key indicator data. S2. Use the FMEA method to obtain the trade-off ranking of each failure mode, failure cause, and key components; S3. A reliability analysis model for hydrogen fuel cells is constructed by combining the impact relationship matrix of failure modes, failure causes, and key component data, as well as a compromise ranking. S4. Set the reliability threshold for the reliability index data within the key index data, and establish a set of drivers based on the measured values of the reliability index data and the reliability threshold. The total reliability fluctuation of hydrogen fuel cells caused by the set of driving factors is calculated by a reliability analysis model, and the VIKOR algorithm is used to map the total reliability fluctuation and the compromise ranking to the interval (0,1) as the reliability score of hydrogen fuel cells. S5. Map the compromise ranking, total reliability fluctuation, reliability score of hydrogen fuel cell, and influence relationship matrix into a visual chart and output it as a file to complete the reliability evaluation of hydrogen fuel cell.
2. The method for evaluating the reliability of hydrogen fuel cells under complex operating conditions according to claim 1, characterized in that, In step S1, key indicator data include typical failure modes, failure causes and frequencies, reliability indicator data, key component data of hydrogen fuel cells, and data on the influence relationships between hydrogen fuel cell components, failure modes and failure causes.
3. The method for evaluating the reliability of hydrogen fuel cells under complex operating conditions according to claim 1, characterized in that, Step S1 is as follows: S11. Collect data on hydrogen fuel cells under typical operating conditions. A typical failure mode, One cause of failure Key components, failure causes, and frequency of key component failures Frequency of failure modes caused by component failure and reliability index data And store them in a two-dimensional relational table format; in: Indicates the i-th cause of failure; , These represent the i-th and j-th key components, respectively. This represents the j-th fault mode; S12. Based on historical failure data, determine the correlations between failure causes, failure modes, and key components. Establish corresponding tables using the HoR cross-correlation matrix natural language value-to-numerical conversion table and the HoR autocorrelation matrix scoring criteria. Failure cause influence relationship matrix Failure Mode and Effect Matrix The key component influence relationship matrix.
4. The method for evaluating the reliability of hydrogen fuel cells under complex operating conditions according to claim 1, characterized in that, Step S2 is as follows: S21. Establish a natural description language terminology set and corresponding quantitative fuzzy numbers for each risk factor in the FMEA method, as the quantitative evaluation criteria for each risk factor. Risk factors include severity (S), incidence (O), and difficulty of detection (D); S22. Based on the quantitative evaluation criteria corresponding to each risk factor, score the failure modes, failure causes, and risk factors of key components of hydrogen fuel cells. The scores corresponding to each risk factor are summarized as the original dataset of risk factor scores for hydrogen fuel cell failure modes, failure causes, and key components. S23. Using rough set theory, the score values in the original dataset are transformed into interval numbers, and the interval numbers are fused to obtain the lower bound of the interval numbers for each failure cause, failure mode, and key component under various risk factors. Upper bound of the sum of interval numbers ,in: ； ； In the formula, For the kth failure cause, failure mode, or critical component; for The number of objects included in the lower approximation; This is the score for the kth failure cause, failure mode, or critical component. for The lower approximate set; for The number of objects included in the upper approximation; for The upper approximate set; Calculate the number of fusion intervals [ , ],in: ； ； In the formula, To be the lower bound of the number of fusion intervals, for The corresponding lower bound of the interval number; To be the upper bound of the number of fusion intervals, for The corresponding upper bound of the interval number; S24. The optimal values of each failure mode, failure cause, and key component in each quantitative evaluation criterion are taken as the ideal solution. The worst values of each failure mode, failure cause, and key component in each quantitative evaluation criterion are taken as the negative ideal solution. ; S25. Calculate the distance between the roughness number and the ideal solution and negative ideal solution for each failure mode, failure cause, and key component under each risk factor in the evaluation interval. : ； In the formula, [ , [Is the] number The failure mode is related to the first Rough score of each risk factor; S26, Combining the number of fusion intervals [ , Calculate the weights of each failure mode, failure cause, and key component. ; And by analyzing the weights of each failure mode, failure cause, and key components. Calculate the corresponding group utility maximization ranking Individual Regret Minimization Sort : ； ； Ranking based on the maximum group utility of each failure mode, failure cause, and critical components. Individual Regret Minimization Sort Calculate the corresponding compromise sort : ； In the formula, , , , ; The weight of the "majority criterion" strategy or the maximum group utility weight; When, it means that a decision is made based on the opinion of the majority; When, it means that a decision is made based on the level of agreement; When this is the case, it indicates that a decision will be made based on the circumstances of refusal. Compromise sorting The failure modes, causes of failure, and severity of critical components are ranked.
5. The method for evaluating the reliability of hydrogen fuel cells under complex operating conditions according to claim 1, characterized in that, In step S3, the reliability analysis model is: RM = <V,R,W>; In the formula, V is the set of all nodes in the reliability analysis model, and V = { }, It serves as the central node of the reliability analysis model; These are nodes representing reliability indices, failure causes, key components, and failure modes in the reliability analysis model. R is the set of all relation matrices in the reliability analysis model, and R = { }; The weight matrix of the nodes causing the failure; This is a weight matrix representing the correlation between failure causes; This is a weighted matrix representing the correlation between failure causes and key components; The weight matrix for key component nodes; This is a weighted matrix representing the relationships between key components. This is a weighted matrix representing the correlation between key components and failure modes. The weight matrix for the fault mode nodes; This is a weight matrix representing the correlation between failure modes; This is the weight matrix representing the correlation between failure modes and the model's central node; The method for determining each matrix in the set R of all relation matrices in the reliability analysis model is as follows: Based on key performance indicator data for hydrogen fuel cells, the presence and frequency of failures of critical components caused by failure factors are used as the basis for determining the correlation matrix and weight matrix between failure factors and critical components in the reliability analysis model; the presence and frequency of failure modes caused by critical component failures are used as the basis for determining the correlation matrix and weight matrix between the critical component layer and failure modes in the reliability analysis model; the correlation matrix includes the correlation matrix between reliability indicators and failure factors. Correlation matrix between failure causes and failure modes Correlation matrix between key components and failure modes And the matrix of relationships between key components and systems. The weight matrix includes the weights between reliability metrics and failure causes. Weight matrix of the relationship between failure causes and key components Weight matrix of correlation between key components and failure modes And the matrix of relationships between key components and systems. ; Based on the key performance data of hydrogen fuel cells, as well as the correlation matrix between failure causes, the correlation data between key components, and the correlation data between failure modes, a correlation matrix between failure causes was determined. Correlation matrix between failure modes Relationship matrix between key components ; The weight matrix of the failure cause node is determined by a compromise ranking based on failure cause, critical components, and failure mode severity. Weight matrix of key component nodes Weight matrix of fault mode nodes .
6. The method for evaluating the reliability of hydrogen fuel cells under complex operating conditions according to claim 1, characterized in that, Step S4 is as follows: S41. Based on historical operating data, experiments, and expert prior knowledge, set the normal range of reliability index data as the reliability threshold. Calculate the distance between the measured value of the reliability index data and the reliability threshold. If the distance is greater than the safe distance, then the reliability index data is... Add to the set of motivations ; ； In the formula, The distance between the measured value of the reliability index data and the reliability threshold is generally calculated using a membership function; This is the safe distance at which the indicator takes effect without causing failure. S42, Traversing the set of motivations Each element in A breadth-first search strategy is used to find all nodes in the reliability analysis model that reach the central node. The pathways, all of which constitute the cut set of the reliability analysis model; S43. Calculate the set of drivers based on the cut set of the reliability analysis model. Total fluctuation in system reliability caused by propagation along all pathways: The VIKOR algorithm is used to rank the total reliability fluctuations and trade-offs. Mapped to (0, 1), it serves as a score for the reliability of the hydrogen fuel cell.
7. A reliability evaluation system for hydrogen fuel cells under complex operating conditions, characterized in that, It includes a data acquisition module, a model building module, a reliability assessment module, and a graphical risk display module, which are connected sequentially via a data bus or API. The model building module combines various risk factors in the FMEA method to obtain a trade-off ranking of each failure mode, failure cause, and key component, and combines key indicator parameters and influence relationship matrices collected by the data acquisition module to build a reliability analysis model; the reliability assessment module combines the reliability analysis model and trade-off ranking to output a reliability score for the hydrogen fuel cell. Finally, the graphical risk display module outputs visual charts in a visual manner, including trade-off ranking, total reliability fluctuation, hydrogen fuel cell reliability scores, and influence relationship matrix mappings.