Water turbine state evaluation method fusing fuzzy theory and comprehensive weighting

Abstract

According to the water turbine state evaluation method integrating the fuzzy theory and the comprehensive weighting, historical monitoring data are fully utilized, so that the weighting result is more reasonable; each monitoring index limit value of the unit is determined based on the Gaussian threshold, the characteristics of the unit are fully considered, and the problem that the index limit values under different power stations, different types of units and different unit working conditions are difficult to accurately obtain is solved. According to the method, the index limit value determined based on the Gaussian threshold is subjected to section division, and the corresponding relation between the divided sections and each evaluation grade of the triangular-semi-trapezoidal fuzzy evaluation function is established, so that the membership function design is more scientific; the evaluation state of the middle layer is used as the basis for evaluating the overall state of the unit system, and the defect that the influence degree of multiple abnormal indexes on the overall state of the equipment cannot be accurately reflected by a variable weight mode is effectively overcome.

Description

Technical Field

[0001] The invention relates to the field of health status analysis of hydropower units, and in particular to a method for evaluating the status of a hydropower unit integrating fuzzy theory and comprehensive weighting. Background Art

[0002] Hydropower units are key equipment for hydropower energy conversion. In recent years, with the continuous development of my country's hydropower industry, the installed capacity of power stations has become larger and larger, and the unit structure has become increasingly complex. In addition, most of the units put into production early have entered the middle and late stages of their service life, which puts forward higher requirements for operation and maintenance. Therefore, it is of great significance to reasonably evaluate the operating status of hydropower units and promptly discover safety hazards of units to ensure the stable operation of units.

[0003] Condition monitoring and analysis are important links to ensure the safe and stable operation of hydropower units. Most domestic power stations have installed real-time and effective condition monitoring systems, which determine the real-time operation status of the unit by judging whether the monitoring indicators are out of limit. At present, the evaluation of unit status is mostly based on the signal analysis of a single measuring point, focusing on local diagnosis, and lacking the evaluation and analysis of the overall status of the system. The condition monitoring system of hydropower units has many measuring points, and there are more normal samples and fewer fault samples. How to use the existing monitoring data to conduct a comprehensive and effective evaluation of the unit is an issue that needs to be explored in depth in the hydropower field.

[0004] Combining hierarchical analysis with fuzzy comprehensive evaluation method based on fuzzy mathematics is a hot research and application topic in the current equipment status analysis direction. However, existing technologies often have problems such as excessive reliance on expert experience in index weighting and lack of objectivity; difficulty in determining index limits, lack of detailed regulations on the limits of various monitoring quantities for different types of units, unclear limit ranges for some monitoring indicators, and inability to effectively evaluate; and unreasonable correspondence between index degradation degree and fuzzy membership interval. Summary of the invention

[0005] The main purpose of the present invention is to provide a method for evaluating the state of a turbine by integrating fuzzy theory and comprehensive weighting, so as to solve the problem of the insufficiency of the current equipment state analysis technology in the field of hydropower.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for evaluating the state of a turbine integrating fuzzy theory and comprehensive weighting, the method relies on hierarchical analysis and Gaussian threshold to improve fuzzy evaluation, and the method includes the following steps: S1. According to the unit structure and measurement point arrangement, determine the turbine hierarchical analysis system and use the hierarchical analysis method to calculate the subjective weight of each component; Obtain historical data of the monitoring system, calculate the objective weight of each component through the CRITIC method based on the historical data of the monitoring system, and combine the subjective and objective weights to obtain the comprehensive weight; S2. According to the historical data of the unit, the upper and lower limits of each monitoring indicator of the unit are determined by using the Gaussian threshold method and the degradation degree of the real-time monitoring indicator is calculated. The degradation degree is introduced into the fuzzy evaluation model, and the state level in the model is divided into four types. According to the characteristics of the indicator interval in the Gaussian threshold method, the corresponding relationship between the indicator degradation degree and the state level is established; S3. After obtaining the fuzzy judgment matrix of each indicator, the indicator fuzzy judgment matrix is ​​weighted to obtain the overall fuzzy judgment matrix of the system, and the operating state of the unit is judged according to the maximum membership principle.

[0007] In the preferred solution, step 1 is to obtain the comprehensive weight of the index according to the minimum discriminant information principle. ,Right now: (1); i is the subjective weight, i is the objective weight, i and j are the i-th and j-th monitoring indicators respectively, then the comprehensive weight vector is: ; In step 2, the status levels are divided into four states: good, qualified, caution, and abnormal. The corresponding relationship between the indicator limit and each status level is determined. Based on the triangle-semi-trapezoidal membership function, the membership of each monitoring indicator of the hydropower unit to each level is obtained to construct a fuzzy evaluation system for the unit.

[0008] In the preferred embodiment, the specific process of the analytic hierarchy process is as follows: A1. Establish a hierarchical model, and stratify the multiple indicators contained in the object to be evaluated according to different attributes according to the structural characteristics of the research object; A2. Construct a judgment matrix. Assume that this layer has Monitoring indicators , judge the importance between two indicators through expert experience, Relative to The importance of is expressed on a scale of 1-9, denoted by ,in ,and , then the judgment matrix: (2); A3. Weight calculation and consistency check, calculation of judgment matrix The maximum eigenvalue of and the corresponding eigenvector , normalize the feature vector to get the corresponding indicator weight vector , calculated as: After the judgment matrix is ​​constructed, it needs to be checked for consistency and the consistency check index is defined. : ,like , then the judgment matrix has complete consistency, if If it approaches 0, it has good consistency. The larger it is, the worse the consistency is; define the consistency ratio: ; In the formula, The value is related to the order of the judgment matrix; if the judgment matrix does not meet the consistency condition, that is, , then the judgment matrix needs to be modified until it meets the consistency.

[0009] In the preferred embodiment, the step of calculating the objective weight of each component by the CRITIC method is: B1. Index normalization, for the Monitoring indicators , normalize it and remove the indicator dimension: (3); So the normalized index is: ; B2. Calculate the amount of indicator information. The indicator information includes indicator differences and indicator conflicts. The differences are expressed by standard deviation, denoted as ; Assumption indicator With indicators The Pearson correlation coefficient is , then the indicator The conflict is: (4); Index information for: (5) B3. Calculate the objective weight of the indicator, normalize the amount of information of the indicator, and obtain the objective weight of each indicator: (6); is the objective weight vector of the indicator.

[0010] In the preferred embodiment, the turbine hierarchical analysis system includes a target layer, a project layer and an indicator layer.

[0011] In the preferred solution, in step S2, the degradation degree is calculated by the ratio of the actual value of the indicator to the limit value of the indicator, so as to characterize the degree of degradation of the indicator.

[0012] In the preferred solution, the degradation degree is calculated as follows: (7); In the formula, is the measured value of the indicator, is the optimal value of the indicator, is the lower limit of the indicator, and , g is the index degradation degree.

[0013] The preferred solution includes: (8); According to formula (8), the fuzzy evaluation matrix of indicators is constructed ; f1.f2,f3,f4 are the membership degrees of the indicators belonging to good, qualified, caution, and abnormal states respectively, and g is the indicator degradation degree.

[0014] In the preferred scheme, when all components of the unit are evaluated as normal, the overall evaluation is normal; when the status of any component is warning status, abnormal status or serious status, the overall evaluation should be the most serious status, and the most serious status of the sub-project evaluation results in the project layer is taken as the final status of the turbine system.

[0015] The present invention provides a method for evaluating the state of a hydraulic turbine by integrating fuzzy theory and comprehensive weighting. The beneficial effects of the present invention include: (1) The analytic hierarchy process and entropy weight method are used to determine the subjective and objective weights of each sub-item of the unit. While respecting expert experience, historical monitoring data are fully utilized to make the weight results more reasonable.

[0016] (2) According to the normal distribution phenomenon of the unit monitoring values ​​under the big data model, the limit values ​​of each monitoring indicator of the unit are determined based on the Gaussian threshold, which fully considers the characteristics of the unit itself and solves the problem that the limit values ​​of indicators under different power stations, different types of units and different unit operating conditions are difficult to obtain accurately.

[0017] (3) The indicator limit values ​​determined based on the Gaussian threshold are divided into segments, and the corresponding relationship between the divided segments and the evaluation levels of the triangular-semi-trapezoidal fuzzy evaluation function is established, so that the design of the membership function has a more scientific mathematical basis and avoids the drawback of the lack of reasonable explanation of the membership function caused by the general specification of the correspondence between the indicator degradation degree and the state membership in the fuzzy evaluation.

[0018] (4) In view of the attenuation of the weight transfer of the bottom-level indicators in the hierarchical analysis system, the intermediate-level evaluation status is used as the basis for evaluating the overall status of the unit system, which effectively solves the problem that the dependent variable weight method cannot accurately reflect the impact of multiple abnormal indicators on the overall status of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a flow chart of the water turbine state analysis of the present invention.

[0020] Figure 2 It is a hierarchical analysis system diagram of the turbine system of the present invention.

[0021] Figure 3 This is a relationship diagram between the membership degree and the degradation degree after the Gaussian threshold is improved in the present invention. DETAILED DESCRIPTION

[0022] Example 1 like Figures 1 to 3 As shown, a method for evaluating the status of a turbine that integrates fuzzy theory and comprehensive weighting makes full use of historical monitoring data to make the weight results more reasonable; this application determines the limits of each monitoring indicator of the unit based on the Gaussian threshold, fully considers the characteristics of the unit itself, and solves the problem that the indicator limits are difficult to obtain accurately under different power stations, different types of units, and different unit operating conditions. This application divides the indicator limits determined based on the Gaussian threshold into sections, and establishes a corresponding relationship between the divided sections and the evaluation levels of the triangle-semi-trapezoidal fuzzy evaluation function, making the design of the membership function more scientific; this application uses the intermediate layer evaluation status as the basis for evaluating the overall status of the unit system, effectively solving the problem that the dependent variable weight method cannot accurately reflect the degree of influence of multiple abnormal indicators on the overall status of the equipment.

[0023] The method relies on hierarchical analysis and Gaussian threshold to improve fuzzy evaluation, and the method includes the following steps: S1. According to the unit structure and measurement point arrangement, determine the turbine hierarchical analysis system and use the hierarchical analysis method to calculate the subjective weight of each component; Obtain historical data of the monitoring system, calculate the objective weight of each component through the CRITIC method based on the historical data of the monitoring system, and combine the subjective and objective weights to obtain the comprehensive weight; S2. According to the historical data of the unit, the upper and lower limits of each monitoring indicator of the unit are determined by using the Gaussian threshold method and the degradation degree of the real-time monitoring indicator is calculated. The degradation degree is introduced into the fuzzy evaluation model, and the state level in the model is divided into four types. According to the characteristics of the indicator interval in the Gaussian threshold method, the corresponding relationship between the indicator degradation degree and the state level is established; S3. After obtaining the fuzzy judgment matrix of each indicator, the indicator fuzzy judgment matrix is ​​weighted to obtain the overall fuzzy judgment matrix of the system, and the operating state of the unit is judged according to the maximum membership principle.

[0024] In the preferred solution, step 1 is to obtain the comprehensive weight of the index according to the minimum discriminant information principle. ,Right now: (1); i is the subjective weight, i is the objective weight, i and j are the i-th and j-th monitoring indicators respectively, then the comprehensive weight vector is: ; In step 2, the status levels are divided into four states: good, qualified, caution, and abnormal. The corresponding relationship between the indicator limit and each status level is determined. Based on the triangle-semi-trapezoidal membership function, the membership of each monitoring indicator of the hydropower unit to each level is obtained to construct a fuzzy evaluation system for the unit.

[0025] In the preferred embodiment, the specific process of the analytic hierarchy process is as follows: A1. Establish a hierarchical model, and stratify the multiple indicators contained in the object to be evaluated according to different attributes according to the structural characteristics of the research object; A2. Construct a judgment matrix. Assume that this layer has Monitoring indicators , judge the importance between two indicators through expert experience, Relative to The importance of is expressed on a scale of 1-9, denoted by ,in ,and , then the judgment matrix: (2); A3. Weight calculation and consistency check, calculation of judgment matrix The maximum eigenvalue of and the corresponding eigenvector , normalize the feature vector to get the corresponding indicator weight vector , calculated as: After the judgment matrix is ​​constructed, it needs to be checked for consistency and the consistency check index is defined. : ,like , then the judgment matrix has complete consistency, if If it approaches 0, it has good consistency. The larger it is, the worse the consistency is; define the consistency ratio: ; In the formula, The value is related to the order of the judgment matrix; if the judgment matrix does not meet the consistency condition, that is, , then the judgment matrix needs to be modified until it meets the consistency.

[0026] In the preferred embodiment, the step of calculating the objective weight of each component by the CRITIC method is: B1. Index normalization, for the Monitoring indicators , normalize it and remove the indicator dimension: (3); So the normalized index is: ; B2. Calculate the amount of indicator information. The indicator information includes indicator differences and indicator conflicts. The differences are expressed by standard deviation, denoted as ; Assumption indicator With indicators The Pearson correlation coefficient is , then the indicator The conflict is: (4); Index information for: (5) B3. Calculate the objective weight of the indicator, normalize the amount of information of the indicator, and obtain the objective weight of each indicator: (6); is the objective weight vector of the indicator.

[0027] In the preferred embodiment, the turbine hierarchical analysis system includes a target layer, a project layer and an indicator layer.

[0028] In the preferred solution, in step S2, the degradation degree is calculated by the ratio of the actual value of the indicator to the limit value of the indicator, so as to characterize the degree of degradation of the indicator.

[0029] In the preferred solution, the degradation degree is calculated as follows: (7); In the formula, is the measured value of the indicator, is the optimal value of the indicator, is the lower limit of the indicator, and , g is the index degradation degree.

[0030] The preferred solution includes: (8); According to formula (8), the fuzzy evaluation matrix of indicators is constructed ; f1.f2,f3,f4 are the membership degrees of the indicators belonging to good, qualified, caution, and abnormal states respectively, and g is the indicator degradation degree.

[0031] In the preferred scheme, when all components of the unit are evaluated as normal, the overall evaluation is normal; when the status of any component is warning status, abnormal status or serious status, the overall evaluation should be the most serious status, and the most serious status of the sub-project evaluation results in the project layer is taken as the final status of the turbine system.

[0032] Example 2 Further illustrate with reference to Example 2, Figure 1-3 The structure shown in the figure comprises the following steps: Step 1: According to the unit structure and measurement point layout, determine the turbine hierarchical analysis system, use the hierarchical analysis method to calculate the subjective weight of each component, and use the CRITIC method to calculate the objective weight of each component based on the historical data of the monitoring system, and combine the subjective and objective weights to obtain the comprehensive weight; Step 2: Based on the historical data of the unit, the Gaussian threshold method is used to determine the upper and lower limits of each monitoring indicator of the unit and calculate the degradation degree of the real-time monitoring indicator. The degradation degree is introduced into the fuzzy evaluation model, and the state levels in the model are divided into four types. According to the characteristics of the indicator interval in the Gaussian threshold method, the corresponding relationship between the indicator degradation degree and the state level is established.

[0033] Step 3: After obtaining the fuzzy judgment matrix of each indicator, the indicator fuzzy judgment matrix is ​​weighted to obtain the overall fuzzy judgment matrix of the system, and the operating status of the unit is judged according to the maximum membership principle.

[0034] Among them, Figure 1 As shown, it can be divided into two stages: offline and online.

[0035] (1) Offline stage: Step 1: According to the unit structure and measurement point layout, determine the turbine hierarchical analysis system and divide the turbine into three layers from top to bottom: target layer - project layer - indicator layer; Figure 2 shown.

[0036] Step 2: Use the analytic hierarchy process and CRITIC method to determine the comprehensive weight of each layer; The core of the hierarchical analysis method is to decompose the complex system from the whole to the part, then construct a judgment matrix and solve the maximum eigenvalue of the matrix and the corresponding eigenvector to obtain the weight of each layer of indicators. The specific process is: 1) Establish a hierarchical model. According to the structural characteristics of the research object, the multiple indicators contained in the object to be evaluated are orderly layered according to different attributes.

[0037] 2) Construct the judgment matrix. Assume that this layer has Monitoring indicators , the importance between two indicators is judged through expert experience. Relative to The importance of is expressed on a scale of 1-9, denoted by The 1-9 scaling rules are shown in Table 1, where ,and .

[0038] Table 1 1-9 scaling method

[0039] Then the judgment matrix is: ; (3) Weight calculation and consistency check. Calculate the judgment matrix The maximum eigenvalue of and the corresponding eigenvector , normalize the feature vector to get the corresponding indicator weight vector , calculated as: ; After the judgment matrix is ​​constructed, it needs to be checked for consistency and the consistency check index is defined. : ; like , then the judgment matrix has complete consistency, if If it approaches 0, it has good consistency. The larger the value, the worse the consistency. Define the consistency ratio: ; In the formula, The value is related to the order of the judgment matrix, and the value is referred to Table 2. Table 2 Value reference table

[0040] If the judgment matrix does not meet the consistency condition, that is, , then the judgment matrix needs to be modified until it meets the consistency.

[0041] The CRITIC (Criteria Importance Through Intercriteria Correlation) method is an objective weighting method that is superior to the entropy weight method. This method uses the amount of information contained in the indicator as the basis for determining the indicator weight. The indicator information includes the differences between the same indicator and the conflict between the indicator and different indicators. Indicator difference refers to the range of indicator change. The larger the range of change, the greater the indicator difference, which is generally expressed by the standard deviation. Indicator conflict refers to the degree of correlation between the indicator and other indicators. The greater the conflict, the lower the degree of correlation. The CRITIC method determines the indicator weight, which not only considers the amount of information contained in the same indicator, but also considers the correlation between different indicators, making the calculation results more objective and reasonable. The steps to determine the indicator weight using the CRITIC method are: (1) Normalization of indicators. Monitoring indicators , normalize it and remove the indicator dimension: (9); mean is the average function So the normalized index is:

[0042] (2) Calculate the amount of indicator information. In the CRITIC method, indicator information includes indicator differences and indicator conflicts. The differences are expressed by standard deviation, which is expressed as ; Assumption indicator With indicators The Pearson correlation coefficient is , then the indicator The conflict is:

[0043] Index information for: ; (3) Calculate the objective weight of the indicator. Normalize the amount of information of the indicator to obtain the objective weight of each indicator: ; is the objective weight vector of the indicator.

[0044] The subjective weight obtained by AHP The objective weight obtained by CRITIC method Combine and obtain the comprehensive weight of indicators The comprehensive weight should be as close as possible to the subjective weight and the objective weight, without giving preference to any one of them. The present invention determines the comprehensive weight based on the minimum discriminant information principle. The calculation method is: ; The comprehensive weight vector is: .

[0045] Step 3: Determine the operating limits of each indicator of the unit based on the Gaussian threshold method; Assumptions ,but The probability distribution function of is: (10) Due to the randomness of measurement errors, the monitoring signal of the hydropower unit has obvious normal distribution characteristics. , any set of real-time monitoring values ​​of this indicator Falling in the range The probability of falling within the interval is 99.74%, and the probability of falling outside the interval is 0.26%, that is, the indicator monitoring value falls within the interval It is a low-probability event. In the normal operation of the hydropower unit, the low-probability event is almost impossible to occur. As the normal operating limit of the unit monitoring index, As the overall limit of the unit monitoring indicators, is the optimal value, where: (11) Step 4: Divide the unit project evaluation level into four states: good, qualified, caution, and abnormal, determine the corresponding relationship between the indicator limit and each state level, and construct the unit fuzzy evaluation system based on the triangle-semi-trapezoidal membership function.

[0046] The fuzzy comprehensive evaluation method is based on fuzzy mathematics. After fuzzy transformation of the unit's measured values ​​and the evaluation criteria, the comprehensive evaluation results of the system are obtained. The evaluation results are scientific, accurate and authentic. In fuzzy evaluation, fuzzy transformation is achieved by constructing a membership function. The equipment status membership is , the index degradation degree is The degradation degree is calculated by the ratio of the actual value of the indicator to the limit value of the indicator, which represents the degree of degradation of the indicator. For the larger the better indicator, the degradation degree is calculated as follows: (12) In the formula, is the measured value of the indicator, is the optimal value of the indicator, is the lower limit of the indicator, and .

[0047] Correspondingly, the calculation method of the smaller the better type indicator degradation degree is: (13) In the formula, is the optimal value of the indicator, is the upper limit of the indicator, and .

[0048] The Gaussian threshold method is used to determine the operating limit of the unit monitoring index. , the optimal value is , if the measured value of the indicator , in this case, the larger the better the indicator, the indicator degradation degree is calculated as shown in formula (12). Similarly, if the actual measured value of the indicator is , in this case, the smaller the better the index, and the index degradation degree calculation is shown in formula (13). The unit and index evaluation status are divided into four levels, namely, status I: good, status II: qualified, status III: attention, and status IV: abnormal. The evaluation levels of each unit status are shown in Table 3.

[0049] Table 3 Hydropower unit status evaluation table

[0050] Based on the triangle-semi-trapezoidal membership function, the membership of each monitoring indicator of the hydropower unit to each evaluation level is obtained. When using the traditional triangle-semi-trapezoidal membership function for equipment evaluation, the degradation degree The value of is mostly determined based on expert experience, and there is a lack of reasonable explanation for the corresponding relationship with different evaluation levels. To solve this problem, the unit index limit determined by the Gaussian threshold method is divided into four sections, namely: , , , .in , , , is the segment boundary value, and the corresponding relationship between the boundary value and each state level is shown in Table 4.

[0051] Table 4 Membership function boundary conditions According to the corresponding relationship in Table 4, the improved triangle-semi-trapezoidal membership function is constructed as follows: Figure 3 As shown, Figure 3 In the figure, I~IV are different states of the object to be evaluated.

[0052] The corresponding analytical expression is: (14); In the formula, , , , According to formula (14), the fuzzy evaluation matrix of indicators is constructed: , determine the current value of the indicator relative to each state membership.

[0053] By weighting the fuzzy evaluation matrix of each indicator, we can get the overall fuzzy evaluation matrix of the system.

[0054] (2) Online stage: Step 1: Calculate the real-time degradation degree of the monitoring indicator according to the indicator operation limit; Step 2: Calculate the membership of the index in the membership function relative to each state interval according to the degradation degree, and construct the index fuzzy evaluation matrix; Step 3: Weight the fuzzy evaluation matrix of each indicator at the indicator layer and calculate the fuzzy evaluation matrix at the project layer; Step 4: Weight the project-level fuzzy evaluation matrix and calculate the target-level fuzzy evaluation matrix; Step 5: Based on the maximum membership principle, the overall status of the unit is evaluated through the obtained fuzzy judgment matrix.

[0055] When all components of the unit are evaluated as normal, the overall evaluation is normal; when the status of any component is a warning state, an abnormal state or a serious state, the overall evaluation should be the most serious state. The present invention takes the most serious state of the sub-project evaluation results in the project layer as the final state of the turbine system.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0057] 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 method for evaluating the condition of a hydraulic turbine integrating fuzzy theory and comprehensive weighting, characterized by: The method relies on hierarchical analysis and Gaussian threshold to improve fuzzy evaluation, and the method includes the following steps: S1. According to the unit structure and measurement point arrangement, determine the turbine hierarchical analysis system and use the hierarchical analysis method to calculate the subjective weight of each component; Obtain historical data of the monitoring system, calculate the objective weight of each component through the CRITIC method based on the historical data of the monitoring system, and combine the subjective and objective weights to obtain the comprehensive weight; S2. According to the historical data of the unit, the upper and lower limits of each monitoring indicator of the unit are determined by using the Gaussian threshold method and the degradation degree of the real-time monitoring indicator is calculated. The degradation degree is introduced into the fuzzy evaluation model, and the state level in the model is divided into four types. According to the characteristics of the indicator interval in the Gaussian threshold method, the corresponding relationship between the indicator degradation degree and the state level is established; S3. After obtaining the fuzzy judgment matrix of each indicator, the indicator fuzzy judgment matrix is ​​weighted to obtain the overall fuzzy judgment matrix of the system, and the operating state of the unit is judged according to the maximum membership principle.

2. A method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 1, characterized in that: in, Step 1: Obtain the comprehensive weight of the index based on the minimum discriminant information principle ,Right now: （1）； i is the subjective weight, i is the objective weight, i and j are the i-th and j-th monitoring indicators respectively, then the comprehensive weight vector is: ; In step 2, the status levels are divided into four states: good, qualified, caution, and abnormal. The corresponding relationship between the indicator limit and each status level is determined. Based on the triangle-semi-trapezoidal membership function, the membership of each monitoring indicator of the hydropower unit to each level is obtained to construct a fuzzy evaluation system for the unit.

3. The method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 1 is characterized in that: The specific process of the hierarchical analysis method is as follows: A1. Establish a hierarchical model, and stratify the multiple indicators contained in the object to be evaluated according to different attributes according to the structural characteristics of the research object; A2. Construct a judgment matrix. Assume that this layer has Monitoring indicators , judge the importance between two indicators through expert experience, Relative to The importance of is expressed on a scale of 1-9, denoted by ,in ,and , then the judgment matrix: （2）； A3. Weight calculation and consistency check, calculation of judgment matrix The maximum eigenvalue of and the corresponding eigenvector , normalize the feature vector to get the corresponding indicator weight vector , calculated as: After the judgment matrix is ​​constructed, it needs to be checked for consistency and the consistency check index is defined. : ,like , then the judgment matrix has complete consistency, if If it approaches 0, it has good consistency. The larger it is, the worse the consistency; Define the consistency ratio: ; In the formula, The value is related to the order of the judgment matrix; if the judgment matrix does not meet the consistency condition, that is, , then the judgment matrix needs to be modified until it meets the consistency.

4. The method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 1 is characterized in that: The steps of calculating the objective weight of each component by the CRITIC method are: B1. Index normalization, for the Monitoring indicators , normalize it and remove the indicator dimension: （3）； So the normalized index is: ; B2. Calculate the amount of indicator information. The indicator information includes indicator differences and indicator conflicts. The differences are expressed by standard deviation, denoted as ; Assumption indicator With indicators The Pearson correlation coefficient is , then the indicator The conflict is: （4）； Index information for: （5） B3. Calculate the objective weight of the indicator, normalize the amount of information of the indicator, and obtain the objective weight of each indicator: （6）； is the objective weight vector of the indicator.

5. The method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 1 is characterized in that: The hydroturbine hierarchical analysis system includes a target layer, a project layer and an indicator layer.

6. The method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 1 is characterized in that: In step S2, the degradation degree is calculated by the ratio of the actual value of the indicator to the limit value of the indicator, which represents the degree of degradation of the indicator.

7. The method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 6 is characterized by: The degradation degree is calculated as follows: （7）； In the formula, is the measured value of the indicator, is the optimal value of the indicator, is the lower limit of the indicator, and , g is the index degradation degree.

8. The method for evaluating a hydraulic turbine state by integrating fuzzy theory and comprehensive weighting according to claim 2 is characterized by: include: （8）； According to formula (8), the fuzzy evaluation matrix of indicators is constructed ; f1.f2,f3,f4 are the membership degrees of the indicators belonging to good, qualified, caution, and abnormal states respectively, and g is the indicator degradation degree.

9. The method for evaluating the condition of a hydraulic turbine integrating fuzzy theory and comprehensive weighting according to claim 1 is characterized in that: When all components of the unit are evaluated as normal, the overall evaluation is normal; when the status of any component is warning status, abnormal status or serious status, the overall evaluation should be the most serious status among them, and the status with the most serious evaluation results of the sub-projects in the project layer shall be taken as the final status of the turbine system.

Images