Maintenance decision-making method of hydroelectric equipment based on FMEA
By building multiple libraries and applying multiple algorithms, objective evaluation of hydropower equipment status and automation of maintenance decision-making, solving the problem of separation of equipment status evaluation and maintenance decision-making processes in the existing technology, and improving the intelligence of maintenance plans and objectivity.
Patent Information
- Application Number
- CN202411024839.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing FMEA-based maintenance method of hydropower equipment has failed to achieve the integration of equipment status evaluation process and maintenance decision-making process, resulting in a lack of automated and intelligent equipment status evaluation and maintenance decision-making, relying on manual experience and regular maintenance.
Using FMEA-based maintenance decision-making method, the equipment structure library, function library, operation feature library, fault mode library and optimization measure library are constructed, combined with hierarchical clustering, association rule mining, principal component analysis, Bayesian network, and fuzzy comprehensive evaluation , decision tree and particle swarm optimization algorithm, generate a list of equipment maintenance items, recommended maintenance levels and automatically generate maintenance plans, and realize objective diagnosis and maintenance planning of equipment status.
It realizes the objective evaluation of equipment status and the automation of maintenance decisions, reduces the subjectivity of manual analysis, improves the objectivity and intelligence of maintenance plans, can conduct quantitative analysis and automatically transmit results based on the real-time status of the equipment, and optimizes maintenance projects and the choice of grades.
Smart Images

Figure CN119090478A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the fields of industrial Internet platform, equipment maintenance, hydropower equipment, and FMEA risk management, and in particular to a maintenance decision-making method for hydropower equipment based on FMEA. Background Art
[0002] FMEA-based maintenance is a method of performing equipment or system maintenance using Failure Mode and Effects Analysis (FMEA). FMEA is a risk management tool used to identify potential equipment failure modes, assess their impacts, and develop appropriate repair measures.
[0003] In FMEA-based maintenance, a comprehensive failure mode analysis is first performed on the equipment or system, that is, identifying the possible failure modes and the possible consequences. This can be done by analyzing the working principle of the equipment, historical failure data, and operating manuals.
[0004] Next, based on the analysis results, each failure mode is evaluated, including determining its severity, frequency of occurrence, and detection capability. This can help determine which failure modes have the greatest impact on the operational safety and reliability of the equipment or system.
[0005] Finally, based on the evaluation results, develop corresponding maintenance plans and measures. This may include regular maintenance, replacement of key components, adding monitoring devices, or improving operation and maintenance procedures. Through FMEA-based maintenance, the risk of equipment or system failure can be minimized and its reliability and maintainability can be improved.
[0006] Most of the existing technical solutions have two directions. One is to conduct FMEA analysis based on fault type and fault cause, to classify the importance of the components of the transformer equipment level and to assess the transformer risk, but it is not connected with the maintenance decision content; the other is that the existing maintenance decision is based on the status evaluation and scoring results. After evaluating the equipment status, the maintenance decision is made, which basically uses manual experience to connect the overall process. However, there is still no way to connect the equipment status evaluation process with the maintenance decision process, so as to realize automated and intelligent equipment status evaluation and maintenance decision, and achieve a more objective evaluation of equipment and formulate a status maintenance mode for different maintenance projects. Summary of the invention
[0007] The main purpose of the present invention is to provide a maintenance decision method for hydropower equipment based on FMEA, so as to solve the problem that the existing maintenance planning is based on equipment status evaluation results, etc., and more on regular detection and analysis of equipment conditions to obtain maintenance planning content.
[0008] In order to solve the above technical problems, the technical solution adopted by the present invention is: a maintenance decision method for hydropower equipment based on FMEA, the maintenance decision method comprising: S1. Conduct FMEA analysis. The analysis steps are as follows: a. Build equipment structure library, equipment function library, operation characteristic library, failure mode library, analysis method library and optimization measure library; b. FMEA version construction: by configuring the prepared information, the FMEA version that needs to be analyzed is formed, and the scope of the subsequent maintenance project generation is clarified; c. FMEA analysis process: by instantiating the generated FMEA version, the equipment that needs risk analysis is identified, and finally a list of equipment maintenance items that need to be performed is generated based on the maintenance items associated with the risk analysis results; S2. Select the maintenance items. The FMEA analysis results will intelligently recommend a list of items that need maintenance, including standard items and non-standard items. S3. Determine the maintenance level, and intelligently recommend the maintenance level of the unit based on the risk analysis results and the items that need maintenance; S4. Prepare maintenance plans and automatically generate maintenance plans based on intelligent scheduling tools through intelligently recommended maintenance level results.
[0009] In the preferred solution, the equipment structure library is used to clarify the hierarchical relationship and role of the hydropower station equipment system. The hierarchical clustering algorithm is used to clarify the hierarchical relationship and role of the hydropower station equipment system. Starting from a single device, similar devices or levels are gradually merged to form a hierarchical structure. The expression formula is: (1); in, Yes Equipment i and equipment j The distance between and is the value of the device and on the th feature, is the number of features; Used to calculate the distance between devices to determine which devices or levels should be merged and build a clear system hierarchy.
[0010] In the preferred solution, the equipment function library conducts a detailed analysis of the equipment functions, establishes a direct relationship between the functions and the equipment structure through function decomposition, decomposes the functions into various structures in detail, and uses the association rule mining algorithm to establish the relationship between the functions and the equipment structure. The construction formula is: (2); (3); in, T is a transaction set, is included A The transaction set, is a transaction set that includes both A and B; Support To measure the universality of a rule, Confidence It is used to measure the certainty of rules and help discover strong association rules between device structure and function.
[0011] In the preferred solution, the operation characteristic library stores the operation characteristic requirements of the device, and the principal component analysis algorithm is used to construct the operation characteristic library, converting multiple related operation characteristic variables into a group of fewer unrelated variables, namely, the principal components, and the construction formula is: (4); in, X is a variable vector, is the mean vector, is the covariance matrix; used to calculate the covariance matrix of the variables to determine the direction and importance of the principal components.
[0012] In the preferred solution, the fault mode library, based on the equipment function library, identifies the fault cause, fault mode and fault impact for the functional failure of each hierarchical structural element, and establishes a relationship chain of fault cause-mode-impact to form an equipment fault network. The Bayesian network is used to construct the fault mode library. Based on the equipment function library, nodes are established to represent the fault cause, mode and impact, and the conditional probability relationship between nodes is defined. The construction formula is: (5); in, Under the condition that B occurs A The probability of occurrence is used to calculate conditional probabilities and determine the strength of the relationship between failure causes, modes, and effects.
[0013] In the preferred solution, the analysis method library is a method library that collects calculation methods of the three dimensions of severity, occurrence and detection to quantitatively analyze the risks that may be caused by the failure mode of the equipment. The fuzzy comprehensive evaluation algorithm is used to perform quantitative analysis of the risks, and the formula is constructed as follows: (6); Among them, W is the weight vector, R is the evaluation matrix, Represents fuzzy synthesis operation; it is used to comprehensively evaluate the risk of failure modes by considering factors in multiple dimensions such as severity, occurrence and detection.
[0014] In the preferred solution, the optimization measures library includes: preventive measures library: based on the maintenance guidelines of hydropower stations, it is necessary to consider sorting out maintenance items and maintenance level information, and at the same time, it is necessary to consider the linkage with advanced maintenance decision-making applications. Based on the decision tree algorithm, the maintenance items and level information are sorted out, and the linkage with advanced maintenance decision-making applications is considered. The construction formula is: (7); (8); (9); in, is the information entropy of data set D, It is the information entropy after being divided based on attribute A. is the information gain of attribute A; The constructed formula is used to select the optimal attributes for splitting the decision tree, thereby determining effective preventive measures.
[0015] In the preferred solution, the detection measure library needs to consider the relevant information of the periodic test items, and a linear programming algorithm is used to consider the relevant information of the periodic test items. The construction formula is: (10); (11); in, c The objective function coefficient vector, x is the decision variable vector, A is the constraint matrix, b is the constraint vector; The algorithm is used to maximize the objective function under various constraints to determine the optimal detection measure plan.
[0016] In the preferred solution, the corrective action library: needs to consider the relevant information of handling defects, and uses integer programming algorithm to consider the relevant information of handling defects.
[0017] In the preferred embodiment, the steps bc The specific steps are: A1. Use genetic algorithm to construct the FMEA version. First, encode the prepared information to form the initial population; Then, through selection, crossover and mutation operations, the population is continuously optimized to obtain the optimal configuration and form the required FMEA version; A2. Use the particle swarm optimization algorithm to perform analysis and initialize a group of particles, each of which represents a possible equipment selection and risk analysis solution; The particle updates its speed and position according to its own optimal position and the optimal position of the entire group, so as to find the optimal risk analysis results and related maintenance items. The analysis process formula is: (12); (13); in, is the velocity of the particle, is the position of the particle, is the inertia weight, and is the learning factor, and is a random number in the range [0, 1], is the individual optimal position, is the optimal position of the group; In the analysis process formula, formula (12) is used to update the particle velocity, and formula (13) is used to update the particle position; By continuously iterating these two formulas, the particle swarm gradually converges to the optimal solution, thereby determining the equipment that requires risk analysis and generating the final equipment maintenance item list.
[0018] The present invention provides a maintenance decision method for hydropower equipment based on FMEA. The present invention diagnoses the real-time status of the equipment, draws objective conclusions, and then makes maintenance plans to make up for the original subjective and periodic maintenance of the equipment.
[0019] In addition, the existing maintenance planning process requires more manual experience and more links for manual analysis. Due to different manual experience and standards, this application can connect processes in series through quantitative results, so the links that require manual analysis can be replaced by the industrial analysis tools provided by the industrial Internet platform, reducing the subjective analysis component of manual experience, improving objective quantitative analysis, and automatically transmitting the results of quantitative analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is the main structural diagram of the cleaning process of the present invention; Figure 2 It is the overall appearance structure diagram of the present invention; Figure 3 It is a structural diagram of the ejection protection shell of the ejection device of the present invention; Figure 4 It is a disassembled structural diagram of the cover plate of the general protective shell of the present invention. DETAILED DESCRIPTION
[0021] Example 1 like Figures 1 to 4 As shown, a maintenance decision method for hydropower equipment based on FMEA, the maintenance decision method includes: S1. Conduct FMEA analysis. The analysis steps are as follows: a. Build equipment structure library, equipment function library, operation characteristic library, failure mode library, analysis method library and optimization measure library; b. FMEA version construction: by configuring the prepared information, the FMEA version that needs to be analyzed is formed, and the scope of the subsequent maintenance project generation is clarified; c. FMEA analysis process: by instantiating the generated FMEA version, the equipment that needs risk analysis is identified, and finally a list of equipment maintenance items that need to be performed is generated based on the maintenance items associated with the risk analysis results; S2. Select the maintenance items. The FMEA analysis results will intelligently recommend a list of items that need maintenance, including standard items and non-standard items. S3. Determine the maintenance level, and intelligently recommend the maintenance level of the unit based on the risk analysis results and the items that need maintenance; S4. Prepare maintenance plans and automatically generate maintenance plans based on intelligent scheduling tools through intelligently recommended maintenance level results.
[0022] In the preferred solution, the equipment structure library is used to clarify the hierarchical relationship and role of the hydropower station equipment system. The hierarchical clustering algorithm is used to clarify the hierarchical relationship and role of the hydropower station equipment system. Starting from a single device, similar devices or levels are gradually merged to form a hierarchical structure. The expression formula is: (1); in, Yes Equipment i and equipment j The distance between and is the value of the device and on the th feature, is the number of features; Used to calculate the distance between devices to determine which devices or levels should be merged and build a clear system hierarchy.
[0023] In the preferred solution, the equipment function library conducts a detailed analysis of the equipment functions, establishes a direct relationship between the functions and the equipment structure through function decomposition, decomposes the functions into various structures in detail, and uses the association rule mining algorithm to establish the relationship between the functions and the equipment structure. The construction formula is: (2); (3); in, T is a transaction set, is included A The transaction set, is a transaction set that includes both A and B; Support To measure the universality of a rule, Confidence It is used to measure the certainty of rules and help discover strong association rules between device structure and function.
[0024] In the preferred solution, the operation characteristic library stores the operation characteristic requirements of the device, and the principal component analysis algorithm is used to construct the operation characteristic library, converting multiple related operation characteristic variables into a group of fewer unrelated variables, namely, the principal components, and the construction formula is: (4); in, X is a variable vector, is the mean vector, is the covariance matrix; used to calculate the covariance matrix of the variables to determine the direction and importance of the principal components.
[0025] In the preferred solution, the fault mode library, based on the equipment function library, identifies the fault cause, fault mode and fault impact for the functional failure of each hierarchical structural element, and establishes a relationship chain of fault cause-mode-impact to form an equipment fault network. The Bayesian network is used to construct the fault mode library. Based on the equipment function library, nodes are established to represent the fault cause, mode and impact, and the conditional probability relationship between nodes is defined. The construction formula is: (5); in, Under the condition that B occurs A The probability of occurrence is used to calculate conditional probabilities and determine the strength of the relationship between failure causes, modes, and effects.
[0026] In the preferred solution, the analysis method library is a method library that collects calculation methods of the three dimensions of severity, occurrence and detection to quantitatively analyze the risks that may be caused by the failure mode of the equipment. The fuzzy comprehensive evaluation algorithm is used to perform quantitative analysis of the risks, and the formula is constructed as follows: (6); Among them, W is the weight vector, R is the evaluation matrix, Represents fuzzy synthesis operation; it is used to comprehensively evaluate the risk of failure modes by considering factors in multiple dimensions such as severity, occurrence and detection.
[0027] In the preferred solution, the optimization measures library includes: preventive measures library: based on the maintenance guidelines of hydropower stations, it is necessary to consider sorting out maintenance items and maintenance level information, and at the same time, it is necessary to consider the linkage with advanced maintenance decision-making applications. Based on the decision tree algorithm, the maintenance items and level information are sorted out, and the linkage with advanced maintenance decision-making applications is considered. The construction formula is: (7); (8); (9); in, is the information entropy of data set D, It is the information entropy after being divided based on attribute A. is the information gain of attribute A; The constructed formula is used to select the optimal attributes for splitting the decision tree, thereby determining effective preventive measures.
[0028] In the preferred solution, the detection measure library needs to consider the relevant information of the periodic test items, and a linear programming algorithm is used to consider the relevant information of the periodic test items. The construction formula is: (10); (11); in, c The objective function coefficient vector, x is the decision variable vector, A is the constraint matrix, b is the constraint vector; The algorithm is used to maximize the objective function under various constraints to determine the optimal detection measure plan.
[0029] In the preferred solution, the corrective action library: needs to consider the relevant information of handling defects, and uses integer programming algorithm to consider the relevant information of handling defects.
[0030] In the preferred embodiment, the steps bc The specific steps are: A1. Use genetic algorithm to construct the FMEA version. First, encode the prepared information to form the initial population; Then, through selection, crossover and mutation operations, the population is continuously optimized to obtain the optimal configuration and form the required FMEA version; A2. Use the particle swarm optimization algorithm to perform analysis and initialize a group of particles, each of which represents a possible equipment selection and risk analysis solution; The particle updates its speed and position according to its own optimal position and the optimal position of the entire group, so as to find the optimal risk analysis results and related maintenance items. The analysis process formula is: (12); (13); in, is the velocity of the particle, is the position of the particle, is the inertia weight, and is the learning factor, and is a random number in the range [0, 1], is the individual optimal position, is the optimal position of the group; In the analysis process formula, formula (12) is used to update the particle velocity, and formula (13) is used to update the particle position; By continuously iterating these two formulas, the particle swarm gradually converges to the optimal solution, thereby determining the equipment that requires risk analysis and generating the final equipment maintenance item list.
[0031] Example 2 Further illustrate with reference to Example 1, Figure 1-4 The structure shown in the figure, this application realizes the maintenance decision of the unit based on FMEA, comprehensively opens up the equipment status evaluation and maintenance decision business process, realizes the objective evaluation of the equipment health status, and can automatically pass the evaluation results to the maintenance decision process. The maintenance decision process will automatically plan the entire process based on the information input by FMEA, and realize intelligent state maintenance. The maintenance decision method is as follows: Step 1. Conduct FMEA analysis; Step 1.1 Build the equipment structure library, equipment function library, operation characteristic library, failure mode library, analysis method library and optimization measure library; The composition of hydropower station equipment is usually a complex system composed of one or more subsystems. If a certain link or a basic unit fails, it may indirectly or directly lead to the functional failure of the upper system, or even the functional failure of the entire system. Therefore, it is necessary to build an equipment structure library to clarify the hierarchical relationship and role of the hydropower station equipment system to facilitate FMEA analysis.
[0032] The equipment function library conducts a detailed analysis of equipment functions. Through functional decomposition, it establishes a direct relationship between functions and equipment structures, and decomposes the functions into each structure in detail.
[0033] Since the equipment characteristic parameter requirements are the operating characteristic requirements required for the realization of functions, they are physical quantities that can be quantitatively characterized, such as temperature, torque, current, etc. When the characteristics are not realized or not met, they can be regarded as failure modes or failure causes. Therefore, it is necessary to build an operation characteristic library to store the equipment operation characteristic requirements.
[0034] Based on the equipment function library, the fault mode library identifies the fault causes, fault modes and fault impacts for the functional failures of structural elements at each level, and establishes a relationship chain of fault causes, modes and impacts to form an equipment fault network.
[0035] The analysis method library is a method library that brings together calculation methods in three dimensions: severity, occurrence, and detection to conduct quantitative analysis on the risks that may be caused by the failure modes of equipment.
[0036] The optimization measures library includes: preventive measures library, detection measures library and maintenance measures library.
[0037] Preventive measures library: It is necessary to sort out maintenance items, maintenance levels and other information based on the maintenance guidelines of hydropower stations, and it is also necessary to consider the linkage with advanced maintenance decision-making applications.
[0038] 2) Detection measures library: Relevant information of regular test items needs to be considered.
[0039] 3) Corrective action library: Relevant information for handling defects needs to be considered.
[0040] Step 1.2 FMEA version construction: By configuring the prepared information, form the FMEA version that needs to be analyzed and clarify the scope of the subsequent maintenance project generation; The specific method is to create a FMEA project, clarify the research object to be analyzed, configure the relevant business library for basic information preparation, clarify the processing team and completion time of the final optimization suggestions, and provide a basic version for subsequent FMEA instantiation operation.
[0041] Step 1.3 FMEA analysis process, by instantiating the generated FMEA version, clarify the equipment that needs risk analysis, and finally generate a list of equipment maintenance items that need to be carried out based on the maintenance items associated with the risk analysis results.
[0042] Based on the operation of the instance FMEA analysis, the final RPN risk situation is output, and corresponding optimization measures are pushed based on the RPN risk. The optimization measures are mapped with the maintenance items, such as Figure 1 shown.
[0043] Step 2. Select the maintenance item; In particular, the FMEA analysis results will intelligently recommend a list of items that need to be repaired, including standard items and non-standard items, such as Figure 2 shown.
[0044] Step 3. Determine the maintenance level; In particular, based on the risk analysis results and the items that need to be repaired, the maintenance level of the unit is intelligently recommended, such as Figure 3 shown.
[0045] Step 4. Prepare a maintenance plan In particular, through the intelligently recommended maintenance level results, the maintenance plan is automatically generated based on the intelligent scheduling tool, such as Figure 4 shown.
[0046] Through the FMEA risk analysis results of this application, data-driven technologies such as artificial intelligence are introduced to make the equipment status assessment more objective. At the same time, the more objective equipment status assessment is deeply associated with the measures. Different measures can be associated with different fault causes for refined management. At the same time, since the maintenance items are associated with relevant risk analysis results, intelligent recommendations for maintenance levels can be made, and automatic scheduling of maintenance plans can be achieved through the determined maintenance levels.
[0047] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.
Claims
1. A maintenance decision method for hydropower equipment based on FMEA, characterized by: Maintenance decision-making methods include: S1. Conduct FMEA analysis. The analysis steps are as follows: a. Build equipment structure library, equipment function library, operation characteristic library, failure mode library, analysis method library and optimization measure library; b. FMEA version construction: by configuring the prepared information, the FMEA version that needs to be analyzed is formed, and the scope of the subsequent maintenance project generation is clarified; c. FMEA analysis process: by instantiating the generated FMEA version, the equipment that needs risk analysis is identified, and finally a list of equipment maintenance items that need to be performed is generated based on the maintenance items associated with the risk analysis results; S2. Select the maintenance items. The FMEA analysis results will intelligently recommend a list of items that need maintenance, including standard items and non-standard items. S3. Determine the maintenance level, and intelligently recommend the maintenance level of the unit based on the risk analysis results and the items that need maintenance; S4. Prepare maintenance plans and automatically generate maintenance plans based on intelligent scheduling tools through intelligently recommended maintenance level results.
2. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: The equipment structure library is used to clarify the hierarchical relationship and role of the hydropower station equipment system. The hierarchical clustering algorithm is used to clarify the hierarchical relationship and role of the hydropower station equipment system. Starting from a single device, similar devices or levels are gradually merged to form a hierarchical structure. The expression formula is: (1); in, Yes Equipment i and equipment j The distance between and is the value of the device and on the th feature, is the number of features; Used to calculate the distance between devices to determine which devices or levels should be merged and build a clear system hierarchy.
3. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: The equipment function library analyzes the equipment functions in detail, establishes the direct relationship between the functions and the equipment structure through function decomposition, decomposes the functions into various structures in detail, and uses the association rule mining algorithm to establish the relationship between the functions and the equipment structure. The construction formula is: (2); (3); in, T is a transaction set, is included A The transaction set, is a transaction set that includes both A and B; Support To measure the universality of a rule, Confidence It is used to measure the certainty of rules and help discover strong association rules between device structure and function.
4. According to claim 1, a maintenance decision method for hydropower equipment based on FMEA is characterized by: running The characteristic library stores the equipment operation characteristic requirements. The principal component analysis algorithm is used to construct the operation characteristic library, which converts multiple related operation characteristic variables into a set of fewer unrelated variables, namely the principal components. The construction formula is: (4); in, X is a variable vector, is the mean vector, is the covariance matrix; used to calculate the covariance matrix of the variables to determine the direction and importance of the principal components.
5. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: Based on the equipment function library, the fault mode library identifies the fault causes, fault modes and fault impacts for the functional failures of the structural elements at each level, and establishes a relationship chain of fault causes, modes and impacts to form an equipment fault network. The Bayesian network is used to build the fault mode library. Based on the equipment function library, nodes are established to represent the fault causes, modes and impacts, and the conditional probability relationship between nodes is defined. The construction formula is: (5); in, Under the condition that B occurs A The probability of occurrence is used to calculate conditional probabilities and determine the strength of the relationship between failure causes, modes, and effects.
6. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: The analysis method library is a method library that collects the calculation methods of the three dimensions of severity, occurrence and detection to quantitatively analyze the risks that may be caused by the failure mode of the equipment. The fuzzy comprehensive evaluation algorithm is used to perform quantitative analysis of the risks. The construction formula is: (6); Among them, W is the weight vector, R is the evaluation matrix, Represents fuzzy synthesis operation; it is used to comprehensively evaluate the risk of failure modes by considering factors in multiple dimensions such as severity, occurrence and detection.
7. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: The optimization measures library includes: Preventive measures library: Based on the maintenance guidelines for hydropower stations, it is necessary to consider sorting out maintenance items and maintenance level information, and at the same time, it is necessary to consider the linkage with advanced maintenance decision-making applications. Based on the decision tree algorithm, it sorts out maintenance items and level information, and considers the linkage with advanced maintenance decision-making applications. The construction formula is: (7); (8); (9); in, is the information entropy of data set D, It is the information entropy after being divided based on attribute A. is the information gain of attribute A; The constructed formula is used to select the optimal attributes for splitting the decision tree, thereby determining effective preventive measures.
8. According to claim 1, a maintenance decision method for hydropower equipment based on FMEA is characterized by: detecting Measures library: It is necessary to consider the relevant information of the periodic test items. The linear programming algorithm is used to consider the relevant information of the periodic test items. The construction formula is: (10); (11); in, c The objective function coefficient vector, x is the decision variable vector, A is the constraint matrix, b is the constraint vector; The algorithm is used to maximize the objective function under various constraints to determine the optimal detection measure plan.
9. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: Corrective action library: It is necessary to consider the relevant information of handling defects, and use integer programming algorithm to consider the relevant information of handling defects.
10. The maintenance decision method for hydropower equipment based on FMEA according to claim 1 is characterized by: step bc The specific steps are: A1. Use genetic algorithm to construct the FMEA version. First, encode the prepared information to form the initial population; Then, through selection, crossover and mutation operations, the population is continuously optimized to obtain the optimal configuration and form the required FMEA version; A2. Use the particle swarm optimization algorithm to perform analysis and initialize a group of particles, each of which represents a possible equipment selection and risk analysis solution; The particle updates its speed and position according to its own optimal position and the optimal position of the entire group, so as to find the optimal risk analysis results and related maintenance items. The analysis process formula is: (12); (13); in, is the velocity of the particle, is the position of the particle, is the inertia weight, and is the learning factor, and is a random number in the range [0, 1], is the individual optimal position, is the optimal position of the group; In the analysis process formula, formula (12) is used to update the particle velocity, and formula (13) is used to update the particle position; By continuously iterating these two formulas, the particle swarm gradually converges to the optimal solution, thereby determining the equipment that requires risk analysis and generating the final equipment maintenance item list.
Citation Information
Patent Citations
Rail transit train system multi-component dynamic maintenance strategy optimization method
CN112785010A
Power transformation maintenance plan intelligent scheduling strategy method based on machine learning
CN115689177A
Power equipment state maintenance strategy optimization method
CN116128145A
Reliability maintenance analysis method for hydroelectric equipment
CN116882965A
Method and system for constructing FMEA knowledge base of hydroelectric equipment
CN117454981A
Cited By
Intelligent compilation system for maintenance plan budget based on historical data
CN122288227A
