Equipment state evaluation method based on fuzzy comprehensive evaluation

Abstract

The invention discloses an equipment state evaluation method based on fuzzy comprehensive evaluation, and relates to the technical field of fuzzy evaluation, and the method comprises the following steps: obtaining the state data of power equipment of a modular substation and the evaluation index of the power equipment, calculating the optimal inflection point of a trapezoidal distribution function through a gradient descent algorithm, and calculating the optimal inflection point of the trapezoidal distribution function; a membership function of the improved power equipment index quality grade is obtained, the membership of each power equipment evaluation index is calculated, and a single-factor evaluation matrix of the power equipment is obtained; and obtaining a power equipment fuzzy comprehensive evaluation vector based on the single-factor evaluation matrix of the power equipment and the comprehensive weight of the power equipment evaluation index, and carrying out grading evaluation on the state of the power equipment of the modular substation. The problems that according to traditional fuzzy evaluation, when a trapezoidal distribution function is selected as a power equipment membership function, calculation of the membership is not accurate enough generally based on experience or fixed parameters, and actual changes of the state of power equipment cannot be effectively reflected are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of fuzzy evaluation, and more specifically, to an equipment status evaluation method based on fuzzy comprehensive evaluation. Background Art

[0002] Modular substation is a modern power facility. Its core concept is to achieve rapid construction, flexible deployment, intelligent operation and efficient maintenance of substations through modular design. With the continuous growth of power demand and the development of smart grids, modular substations have gradually become an important supplement to traditional substations, especially in the fields of urban expansion, power supply in remote areas, emergency power supply and renewable energy access. Through systematic evaluation of equipment status, potential problems can be discovered in time, the risk of failure can be reduced, and the reliability and service life of equipment can be improved.

[0003] In the traditional fuzzy comprehensive evaluation of the equipment status of modular substations, the selection of the trapezoidal distribution function as the membership function of the power equipment is usually based on experience or fixed parameters, which leads to the inaccurate calculation of the membership and cannot effectively reflect the actual changes in the status of the power equipment. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an equipment status evaluation method based on fuzzy comprehensive evaluation, which solves the problem that when traditional fuzzy evaluation selects a trapezoidal distribution function as the membership function of power equipment, the membership calculation is usually not accurate enough based on experience or fixed parameters, and cannot effectively reflect the actual changes in the status of the power equipment.

[0005] To achieve the above object, the present invention provides the following technical solutions: A method for evaluating equipment status based on fuzzy comprehensive evaluation, characterized in that it includes the following steps: obtaining evaluation indicators of modular substation power equipment, and initially assigning N quality levels of the evaluation indicators of the power equipment; using a gradient descent algorithm to calculate the optimal inflection point of a trapezoidal distribution function, and obtaining an improved membership function of the power equipment indicator quality level; based on the improved membership function of the power equipment indicator quality level, calculating the membership of each power equipment evaluation indicator, and obtaining a single factor evaluation matrix of the power equipment; calculating the comprehensive weight of the power equipment evaluation indicator, and obtaining a fuzzy comprehensive evaluation vector of the power equipment based on the single factor evaluation matrix of the power equipment; and performing a graded evaluation on the status of modular substation power equipment based on the fuzzy comprehensive evaluation vector of the power equipment.

[0006] In a preferred embodiment, the gradient descent algorithm is used to calculate the optimal inflection point of the trapezoidal distribution function to obtain the improved membership function of the power equipment indicator quality grade, specifically: the trapezoidal distribution function is selected as the power equipment membership function, and the fuzzy C-means clustering algorithm is used to obtain the initial inflection point of the trapezoidal distribution function; based on the initial inflection point of the trapezoidal distribution function, the gradient descent algorithm is used to obtain the optimal inflection point of the trapezoidal distribution function to obtain the improved membership function of the power equipment indicator quality grade.

[0007] In a preferred embodiment, a trapezoidal distribution function is selected as the power equipment membership function, and a fuzzy C-means clustering algorithm is used to obtain the initial inflection point of the trapezoidal distribution function, specifically: the number of clusters is set to N, and the power equipment membership matrix is ​​initialized to obtain the initialized power equipment membership matrix; Based on the initialized power equipment membership matrix, the cluster centers of N clusters are calculated; based on the cluster centers of the N clusters, the initialized power equipment membership matrix is ​​updated; by continuously calculating the cluster centers of N clusters and updating the initialized power equipment membership matrix, clustering is iterated based on a preset threshold to obtain the final cluster centers of N clusters; the final cluster centers of N clusters are used as the initial inflection points of the trapezoidal distribution function.

[0008] In a preferred embodiment, a gradient descent algorithm is used based on the initial inflection point of the trapezoidal distribution function to obtain the optimal inflection point of the trapezoidal distribution function, and the membership function of the improved power equipment indicator quality grade is obtained, specifically: an error function is set, and a gradient descent algorithm is used to solve the partial derivative of the error function to obtain a gradient value; based on the gradient value, a gradient update is performed, and when the error function converges to a preset threshold, the iteration is stopped to obtain the optimal inflection points of N trapezoidal distribution functions.

[0009] In a preferred embodiment, based on the improved membership function of the power equipment indicator quality level, the membership of each power equipment evaluation indicator is calculated to obtain a single factor judgment matrix of the power equipment, specifically: for the N quality levels of the power equipment evaluation indicators, based on the trapezoidal distribution function formula, the membership values ​​of the N quality levels of the power equipment evaluation indicators are calculated; based on the membership values ​​of the N quality levels of the power equipment evaluation indicators, a single factor judgment matrix of the power equipment is constructed.

[0010] In a preferred embodiment, the comprehensive weight of the power equipment evaluation index is calculated, specifically: the hierarchical analysis method is used to calculate the first-level index weight of the power equipment evaluation index; the entropy weight method is used to calculate the second-level index weight of the power equipment evaluation index; the first-level index weight of the power equipment index and the second-level index weight of the power equipment index are linearly weighted averaged to obtain the comprehensive weight of the power equipment index.

[0011] In a preferred embodiment, the hierarchical analysis method is used to calculate the first-level indicator weights of the power equipment evaluation indicators, specifically: based on the 1-9 scaling method, a power equipment state evaluation judgment matrix is ​​established, and the power equipment state evaluation judgment matrix is ​​normalized to obtain a normalized power equipment state evaluation judgment matrix; the weight vector of the normalized power equipment state evaluation judgment matrix is ​​calculated to obtain an initial weight vector for power equipment evaluation; the consistency ratio of the initial weight vector is calculated, and the consistency ratio is compared with a preset consistency ratio to obtain the first-level indicator weight of the power equipment after the consistency test is passed.

[0012] In a preferred embodiment, the entropy weight method is used to calculate the secondary indicator weights of the power equipment evaluation indicators, specifically: calculate the probability value of each power equipment evaluation indicator appearing at N quality levels to obtain a power equipment status evaluation probability matrix; based on the power equipment status evaluation probability matrix, calculate the information utility value of each power equipment evaluation indicator; based on the information utility value of each power equipment evaluation indicator, calculate the secondary indicator weights of the power equipment evaluation indicators.

[0013] The technical effects and advantages of the equipment status evaluation method based on fuzzy comprehensive evaluation of the present invention are as follows: 1. The present invention obtains the evaluation index of the power equipment and performs initial assignment, thereby ensuring that the basic data of the evaluation is accurate, can reflect the real state of the equipment, and uses the gradient descent algorithm to calculate the optimal inflection point of the trapezoidal distribution function, optimizes the calculation method of the traditional membership function, makes the evaluation model more flexible and accurate, and can better adapt to the dynamic changes of the equipment operation; secondly, based on the improved membership function, the membership of each indicator is calculated to form a single factor evaluation matrix of the power equipment, which can quantify the influence of each indicator in detail, and provide solid data support for subsequent comprehensive evaluation. The transparency and operability of this method improve the credibility of the evaluation results, facilitate the understanding and application of operation and maintenance personnel, and in the calculation process of the comprehensive weight, the relative importance of each evaluation index can be comprehensively considered to ensure the scientificity and objectivity of the evaluation results; finally, the fuzzy comprehensive evaluation vector is used to perform a hierarchical evaluation of the power equipment status, which not only helps to quickly identify the health status of the equipment, but also provides a reference for maintenance decisions, and improves the efficiency and accuracy of equipment management. Overall, this method optimizes the equipment evaluation process through data-driven decision support, and helps to achieve efficient operation and maintenance of modular substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a schematic diagram of the structure of a device status evaluation method based on fuzzy comprehensive evaluation. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Embodiment 1, Figure 1 The present invention provides a device status evaluation method based on fuzzy comprehensive evaluation, comprising the following steps: S1, obtaining evaluation indicators of power equipment in a modular substation, and initially assigning N quality levels of the evaluation indicators of the power equipment; In this example, the evaluation index of the modular substation power equipment is obtained as follows: The evaluation indicators of the modular substation power equipment include temperature and humidity, dielectric loss factor, historical fault frequency, current and voltage, and power factor.

[0017] In this example, the N quality levels of the evaluation indicators of the power equipment are initially assigned values, specifically: Temperature and humidity: divided according to the normal operating temperature and humidity range and the abnormal temperature and humidity range. For example, the temperature of 20-25℃ and the humidity of 40%-50% can be classified as excellent, the temperature of 25-30℃ and the humidity of 50%-60% can be classified as good, the temperature of 30-35℃ and the humidity of 60%-70% can be classified as medium, and the temperature above 25℃ and the humidity above 70% can be classified as poor. The specific definition and classification can be made according to different power equipment.

[0018] Dielectric loss factor: Generally, a low dielectric loss factor indicates good insulation, while a high dielectric loss factor indicates reduced insulation performance; Historical failure frequency: The equipment is graded based on the number or frequency of past failures. A higher failure frequency usually means a poorer health status of the equipment. Current and voltage: evaluate the extent to which voltage and current deviate from standard values; Power Factor: Deviations from normal values ​​may indicate inefficiency or abnormalities in the equipment.

[0019] It should be noted that "excellent" means that the performance of the power equipment under this indicator is optimal and there is basically no risk; "good" means that the performance of the power equipment under this indicator is relatively good and the risk is relatively low; "medium" means that the performance of the power equipment under this indicator is medium and there is a certain risk; "poor" means that the performance of the power equipment under this indicator is poor and requires special attention or maintenance.

[0020] S2, using the gradient descent algorithm to calculate the optimal inflection point of the trapezoidal distribution function, and obtaining the improved membership function of the quality level of the power equipment index; In this example, the gradient descent algorithm is used to calculate the optimal inflection point of the trapezoidal distribution function to obtain the improved membership function of the quality level of the power equipment index, which is specifically: Selecting a trapezoidal distribution function as the membership function of the power equipment, and using a fuzzy C-means clustering algorithm to obtain an initial inflection point of the trapezoidal distribution function; Based on the initial inflection point of the trapezoidal distribution function, a gradient descent algorithm is used to obtain the optimal inflection point of the trapezoidal distribution function, and the improved membership function of the quality grade of the power equipment index is obtained.

[0021] In this example, the trapezoidal distribution function is selected as the membership function of the power equipment, and the fuzzy C-means clustering algorithm is used to obtain the initial inflection point of the trapezoidal distribution function, specifically: The number of clusters is set to N, and the power equipment membership matrix is ​​initialized to obtain the initialized power equipment membership matrix; Based on the initialized power equipment membership matrix, calculating cluster centers of N clusters; Based on the cluster centers of the N clusters, updating the initialized power equipment membership matrix; By continuously calculating the cluster centers of the N clusters and updating the initialized power equipment membership matrix, clustering is iterated based on a preset threshold to obtain the final cluster centers of the N clusters; The cluster centers of the final N clusters are used as initial inflection points of the trapezoidal distribution function.

[0022] In this example, the cluster number is set to N, and the power equipment membership matrix is ​​initialized to obtain the initialized power equipment membership matrix, which is specifically: The number of quality grades corresponding to the evaluation index of the power equipment is taken as the number of clusters; The membership degree of the ith data point belonging to the jth cluster is randomly generated to obtain the initialized power equipment membership matrix.

[0023] In this example, based on the initialized power equipment membership matrix, the cluster centers of N clusters are calculated, specifically:

[0024] in, is the cluster center of the jth cluster, is the membership degree of the i-th power equipment status data point to the j-th cluster, is the i-th power equipment status data point, m is the fuzzification coefficient, and T is the total number of power equipment status data points.

[0025] In this example, by continuously calculating the cluster centers of the N clusters and updating the initialized power equipment membership matrix, clustering is iterated based on a preset threshold to obtain the final cluster centers of the N clusters, specifically:

[0026] in, is the membership degree of the i-th power equipment status data point to the j-th cluster, is the i-th power equipment status data point, is the cluster center of the jth cluster, is the cluster center of the kth class, is the Euclidean distance between the i-th power equipment status data point and the k-th cluster center, is the Euclidean distance between the i-th power equipment status data point and the j-th cluster center.

[0027] It should be noted that selecting the trapezoidal distribution function as the membership function of the power equipment and obtaining its initial inflection point through the fuzzy C-means clustering algorithm helps to build a membership model that is more in line with the actual state of the equipment. The flexibility of the trapezoidal distribution function can more accurately describe the fuzzy transition between different quality levels of equipment performance indicators, and the fuzzy C-means clustering can adaptively find the data clustering center as the inflection point of the trapezoidal function, making the model building process more automated and objective, and reducing the subjective bias of manually set parameters. This method not only improves the rationality of the membership function, but also makes the equipment health status assessment more accurate and effective.

[0028] In this example, based on the initial inflection point of the trapezoidal distribution function, a gradient descent algorithm is used to obtain the optimal inflection point of the trapezoidal distribution function, and the improved membership function of the power equipment indicator quality level is obtained, which is specifically: Setting an error function, and using a gradient descent algorithm to solve a partial derivative of the error function to obtain a gradient value; Based on the gradient value, a gradient update is performed, and when the error function converges to a preset threshold, the iteration is stopped to obtain the best inflection points of N trapezoidal distribution functions.

[0029] In this example, the error function is set as follows:

[0030] Among them, L is the value of the loss function, N is the total number of quality levels of the power equipment evaluation indicators, The membership function of the improved power equipment quality level is at point The predicted value at is the actual membership degree of the i-th power equipment evaluation index.

[0031] It should be noted that by setting the error function and using the gradient descent algorithm to solve the gradient value to optimize the inflection point of the trapezoidal distribution function, the optimal parameters can be effectively and automatically found to make the trapezoidal membership function more suitable for the power equipment status data. The gradient descent method gradually reduces the value of the error function and stops iterating when it converges to the threshold, thereby finding the optimal inflection point for each trapezoidal function. This method not only improves the optimization efficiency, but also avoids the deviation of subjectively setting the inflection point in the traditional method, making the evaluation model more accurate and adaptable, thereby improving the reliability and accuracy of equipment status evaluation.

[0032] S3, based on the improved membership function of the power equipment index quality grade, calculating the membership of each power equipment evaluation index to obtain a single factor evaluation matrix of the power equipment; In this example, based on the improved membership function of the power equipment index quality level, the membership of each power equipment evaluation index is calculated to obtain a single factor evaluation matrix of the power equipment, specifically: For the m quality levels of the power equipment evaluation index, based on the trapezoidal distribution function formula, calculating the membership values ​​of the m quality levels of the power equipment evaluation index; Based on the membership values ​​of the m quality levels of the power equipment evaluation indicators, a single factor evaluation matrix of the power equipment is constructed.

[0033] In this example, the membership function formula based on the improved power equipment indicator quality level is specifically:

[0034] Among them, a, b, c, d are the inflection point parameters of the membership function of the four improved power equipment index quality levels, x is the evaluation index of the power equipment, is the membership value of the m quality levels of the power equipment evaluation index.

[0035] In this example, based on the membership values ​​of the m quality levels of the power equipment evaluation index, a single factor evaluation matrix of the power equipment is constructed, specifically: Each element of the single factor evaluation matrix of the electric power equipment represents the degree of membership of the ith electric power equipment state evaluation index at the jth quality level, and the degrees of membership are all within the interval of [0, 1].

[0036] S4, calculating the comprehensive weight of the evaluation index of the power equipment, and obtaining the fuzzy comprehensive evaluation vector of the power equipment based on the single factor evaluation matrix of the power equipment; In this example, 6. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 1 is characterized in that the comprehensive weight of the power equipment evaluation index is calculated as follows: Calculate the primary index weight of the power equipment evaluation index by using the hierarchical analysis method; The entropy weight method is used to calculate the secondary index weight of the power equipment evaluation index; The first-level indicator weight of the power equipment indicator and the second-level indicator weight of the power equipment indicator are linearly weighted averaged to obtain the comprehensive weight of the power equipment indicator.

[0037] In this example, the hierarchical analysis method is used to calculate the primary indicator weight of the power equipment evaluation indicator, specifically: Establishing a power equipment state evaluation judgment matrix based on the 1-9 scaling method, and normalizing the power equipment state evaluation judgment matrix to obtain a normalized power equipment state evaluation judgment matrix; Calculating the weight vector of the normalized power equipment state assessment judgment matrix to obtain an initial weight vector for power equipment assessment; The consistency ratio of the initial weight vector is calculated, and the consistency ratio is compared with a preset consistency ratio to obtain the primary indicator weight of the power equipment after the consistency test passes.

[0038] It should be noted that the use of the analytic hierarchy process (AHP) to calculate the subjective weights of the evaluation indicators of power equipment can effectively structure complex problems and quantify the importance of each indicator. AHP decomposes the various evaluation indicators of power equipment from the overall goal to the specific factors step by step by constructing a hierarchical model, which facilitates a clear understanding of the relationship between the indicators. Experts can form a comparison matrix based on their empirical judgment of the relative importance of each indicator. The AHP method further ensures the rationality of the judgment through consistency testing to avoid the deviation caused by subjectivity. In addition, the weight results calculated by AHP can not only reflect the relative importance of each indicator of the equipment, but also provide an intuitive quantitative basis, which is helpful to accurately evaluate the health status of the equipment and guide resource allocation and maintenance decisions. In general, the analytic hierarchy process systematizes and structures complex multi-indicator problems, ensures the scientificity and rationality of weight allocation, and thus improves the accuracy of power equipment status evaluation and the reliability of decision-making.

[0039] In this example, the entropy weight method is used to calculate the secondary index weight of the power equipment evaluation index, specifically: Calculate the probability value of each power equipment evaluation index appearing at m quality levels to obtain the power equipment state evaluation probability matrix; Based on the power equipment state assessment probability matrix, calculating the information utility value of each power equipment evaluation index; The secondary indicator weight of the electric power equipment evaluation indicator is calculated based on the information utility value of each electric power equipment evaluation indicator.

[0040] It should be noted that the entropy weight method calculates the weights of the secondary indicators of the power equipment evaluation indicators through the concept of information entropy, which can effectively reduce human interference and make the weight allocation more scientific and objective. Different from subjective weighting, the entropy weight method uses the variability of each indicator data to measure the importance of the indicator, that is, it assigns higher weights to indicators with greater data differences, thereby reflecting the greater impact of the indicator on the overall evaluation results. By calculating the entropy value of each indicator, the entropy weight method can identify which indicators in the data set have a large amount of information, and automatically reduce the weights of indicators that do not change much in different data points. This objective weight determination method avoids human subjective bias and improves the reliability of weight calculation. At the same time, the entropy weight method can adapt to the data characteristics under different times and environments, ensuring that the weight allocation can accurately reflect the actual situation of the equipment status data. Therefore, the entropy weight method provides a more accurate and objective weight allocation basis for the evaluation of the power equipment status, which is suitable for large-scale and diversified power equipment status monitoring data analysis, and helps to improve the scientificity and practicality of equipment operation status evaluation.

[0041] S5, performing hierarchical evaluation on the status of the power equipment in the modular substation based on the fuzzy comprehensive evaluation vector of the power equipment.

[0042] Based on the fuzzy comprehensive evaluation vector of the electric power equipment, the maximum membership degree in the fuzzy comprehensive evaluation vector of the electric power equipment is calculated by adopting the maximum membership degree principle; Based on the maximum membership degree, the state level of the corresponding power equipment is determined.

[0043] It should be noted that the hierarchical evaluation of the state of power equipment in modular substations based on the fuzzy comprehensive evaluation vector of power equipment has many advantages. First, this method can effectively deal with the fuzziness and subjectivity in the state evaluation of power equipment by processing uncertainty through fuzzy logic, and improve the reliability and scientificity of the evaluation results. Secondly, the fuzzy comprehensive evaluation vector combines multiple evaluation indicators. By comprehensively considering different indicators, the evaluation is more comprehensive and objective, and can reflect the actual operating status of the equipment. Thirdly, this method allows a certain weight distribution between each indicator to ensure that important indicators occupy a corresponding position in the evaluation, thereby improving the accuracy of the evaluation. At the same time, the hierarchical evaluation results facilitate intuitive decision support, helping operation and maintenance personnel to quickly identify the health status of the equipment, take maintenance measures in time, reduce the risk of equipment failure, and extend the service life of the equipment. In addition, the dynamically updated fuzzy evaluation mechanism makes this method highly adaptable and can be adjusted in real time according to the equipment status and environmental changes, meeting the growing intelligent management needs of modern power systems. In summary, the state hierarchical evaluation method based on the fuzzy comprehensive evaluation vector not only improves the scientificity and accuracy of the evaluation, but also provides strong support for the intelligent management of power equipment.

[0044] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0045] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0046] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0047] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0048] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0049] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A device status evaluation method based on fuzzy comprehensive evaluation, characterized in that: The steps include: Acquire state data of power equipment of the modular substation and evaluation indicators of the power equipment, and initially assign N quality levels of the evaluation indicators of the power equipment; The gradient descent algorithm is used to calculate the optimal inflection point of the trapezoidal distribution function, and the improved membership function of the quality grade of power equipment indicators is obtained; Based on the improved membership function of the power equipment index quality grade, the membership of each power equipment evaluation index is calculated to obtain a single factor evaluation matrix of the power equipment; Calculating the comprehensive weight of the power equipment evaluation index, and obtaining the power equipment fuzzy comprehensive evaluation vector based on the single factor evaluation matrix of the power equipment; The status of the power equipment in the modular substation is evaluated in a graded manner based on the power equipment fuzzy comprehensive evaluation vector.

2. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 1 is characterized in that: The gradient descent algorithm is used to calculate the optimal inflection point of the trapezoidal distribution function to obtain the improved membership function of the power equipment index quality level, which is specifically: Selecting a trapezoidal distribution function as the membership function of the power equipment, and using a fuzzy C-means clustering algorithm to obtain an initial inflection point of the trapezoidal distribution function; Based on the initial inflection point of the trapezoidal distribution function, a gradient descent algorithm is used to obtain the optimal inflection point of the trapezoidal distribution function, and the improved membership function of the quality grade of the power equipment index is obtained.

3. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 2 is characterized in that: The trapezoidal distribution function is selected as the membership function of the power equipment, and the fuzzy C-means clustering algorithm is used to obtain the initial inflection point of the trapezoidal distribution function, which is specifically: The number of clusters is set to N, and the power equipment membership matrix is ​​initialized to obtain the initialized power equipment membership matrix; Based on the initialized power equipment membership matrix, calculating cluster centers of N clusters; Based on the cluster centers of the N clusters, updating the initialized power equipment membership matrix; By continuously calculating the cluster centers of the N clusters and updating the initialized power equipment membership matrix, clustering is iterated based on a preset threshold to obtain the final cluster centers of the N clusters; The cluster centers of the final N clusters are used as initial inflection points of the trapezoidal distribution function.

4. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 2 is characterized in that: Based on the initial inflection point of the trapezoidal distribution function, a gradient descent algorithm is used to obtain the optimal inflection point of the trapezoidal distribution function, and the improved membership function of the quality level of the power equipment index is obtained, which is specifically: Setting an error function, and using a gradient descent algorithm to solve a partial derivative of the error function to obtain a gradient value; Based on the gradient value, a gradient update is performed, and when the error function converges to a preset threshold, the iteration is stopped to obtain the best inflection points of N trapezoidal distribution functions.

5. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 4 is characterized in that: Based on the improved membership function of the power equipment index quality level, the membership of each power equipment evaluation index is calculated to obtain the single factor evaluation matrix of the power equipment, which is specifically: For the N quality levels of the power equipment evaluation index, based on the improved membership function formula of the power equipment index quality level, the membership values ​​of the N quality levels of the power equipment evaluation index are calculated; Based on the membership values ​​of the N quality levels of the power equipment evaluation indicators, a single factor evaluation matrix of the power equipment is constructed.

6. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 1 is characterized in that: The comprehensive weight of the power equipment evaluation index is calculated as follows: Calculate the primary index weight of the power equipment evaluation index by using the hierarchical analysis method; The entropy weight method is used to calculate the secondary index weight of the power equipment evaluation index; The first-level indicator weight of the power equipment indicator and the second-level indicator weight of the power equipment indicator are linearly weighted averaged to obtain the comprehensive weight of the power equipment indicator.

7. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 6 is characterized in that: The use of the analytic hierarchy process to calculate the primary index weights of the power equipment evaluation index is specifically: Establishing a power equipment state evaluation judgment matrix based on the 1-9 scaling method, and normalizing the power equipment state evaluation judgment matrix to obtain a normalized power equipment state evaluation judgment matrix; Calculating the weight vector of the normalized power equipment state assessment judgment matrix to obtain an initial weight vector for power equipment assessment; The consistency ratio of the initial weight vector of the electric power equipment evaluation is calculated, and the consistency ratio is compared with a preset consistency ratio to obtain the primary indicator weight of the electric power equipment after the consistency test passes.

8. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 6 is characterized in that: The entropy weight method is used to calculate the secondary index weight of the power equipment evaluation index, specifically: Calculate the probability value of each power equipment evaluation index appearing at N quality levels to obtain the power equipment state evaluation probability matrix; Based on the power equipment state assessment probability matrix, calculating the information utility value of each power equipment evaluation index; The secondary indicator weight of the electric power equipment evaluation indicator is calculated based on the information utility value of each electric power equipment evaluation indicator.

9. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 2 is characterized in that: The membership function formula based on the improved power equipment index quality level is specifically: Among them, a, b, c, d are the inflection point parameters of the membership function of the four improved power equipment index quality levels, and x is the evaluation index of the power equipment.

10. The device status evaluation method based on fuzzy comprehensive evaluation according to claim 5 is characterized in that: The specific calculation formula of the error function is: Among them, L is the value of the loss function, N is the total number of quality levels of the power equipment evaluation indicators, The membership function of the improved power equipment quality level is at point The predicted value at is the actual membership degree of the i-th power equipment evaluation index.

Images