Intelligent power equipment state evaluation method

Abstract

The invention provides an intelligent power equipment state evaluation method. The method comprises the following steps: S1, collecting and classifying historical operation parameters of intelligent power equipment; s2, dynamically screening out the key indexes according to the influence degrees of the key indexes; s3, determining the weight of each key index according to the correlation between the key indexes and the equipment fault, and carrying out adaptive adjustment; s4, forming a new dynamic cloud matter element model based on the combination of the cloud model and the dynamic matter element, and formulating an evaluation index system; s5, outputting the initial membership degree at the moment t1; s6, obtaining a real-time membership degree at the moment t2; and S7, comparing the initial membership degree with the real-time membership degree, outputting the state of the intelligent power equipment at the moment, and predicting the index development trend based on the evaluation index system. According to the invention, the real-time operation parameters of the intelligent power equipment at multiple moments can be efficiently compared through the constructed dynamic matter-element model; the current state of the intelligent power equipment can be clearly presented, and the development trend of the equipment state can be accurately predicted.

Description

Technical Field

[0001] This invention relates to the status of intelligent power equipment, and more particularly to a method for evaluating the status of intelligent power equipment. Background Technology

[0002] One intelligent power equipment condition assessment method comprehensively utilizes sensor technology to collect equipment data in real time, and evaluates the equipment's operating status through data analysis and artificial intelligence algorithms. This method can promptly detect potential faults, provide accurate basis for equipment maintenance, and improve the reliability and stability of the power system.

[0003] While existing methods for assessing the condition of intelligent power equipment collect operating parameters from multiple sensors and analyze all data to derive key criteria relevant to equipment failures, and even construct fuzzy cloud matter-element theory by combining fuzzy matter-element theory with cloud models, which simplifies the condition assessment process to some extent, this method has significant shortcomings. On the one hand, the fuzzy cloud matter-element model can only obtain the current state of the equipment and cannot accurately predict the development trend of the equipment's state. This makes it difficult to plan equipment maintenance and management strategies in advance in practical applications, increasing the risk of sudden equipment failures. On the other hand, when data from multiple time periods needs to be processed, the data must be reprocessed, which is extremely complex, consumes a lot of time and manpower, and seriously affects the efficiency of data processing.

[0004] Therefore, it is necessary to provide a new method for evaluating the condition of intelligent power equipment to solve the above-mentioned technical problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for evaluating the condition of intelligent power equipment.

[0006] The intelligent power equipment condition evaluation method provided by this invention includes the following steps:

[0007] S1. Collect historical operating parameters of intelligent power equipment and classify the historical operating parameters according to the operating status of the equipment. For example, collect and classify data of intelligent power equipment through devices such as temperature sensors, current sensors, voltage sensors and vibration sensors.

[0008] S2. The historical operating parameters of intelligent power equipment when a fault occurs are dynamically filtered according to the influence of key indicators. Key indicators that have a significant impact on the equipment status are filtered out, while indicators that have a small or no impact on the equipment status are filtered out. This reduces the amount of data that needs to be processed, reduces the burden on computing equipment, and improves computing efficiency.

[0009] S3. Determine the weight of each key indicator based on its correlation with equipment failure and make adaptive adjustments to increase the weight of indicators with higher correlation with equipment failure, so as to make the results obtained when processing data more accurate.

[0010] S4. A new dynamic cloud-element model is formed by combining cloud models and dynamic matter-element models, and an evaluation index system for intelligent power equipment is formulated based on key indicators: The cloud model is constructed according to the evaluation index system for key indicators formulated in step S2. Then, the cloud droplets of key indicators are generated using the forward cloud generator algorithm of the cloud model. Then, the classical domain and section domain of the dynamic matter-element model are determined based on the expectation and entropy of the cloud droplets, thereby completing the construction of the dynamic cloud-element model. By combining the dynamic matter-element model with the evaluation index system, real-time operating parameters can be processed quickly. At the same time, the dynamic matter-element model can memorize data, thereby simplifying the steps of performing a second test on the equipment in a short time and improving work efficiency.

[0011] S5. Input the key indicators that have a significant impact on the equipment status collected at time t1 into the dynamic cloud object model to obtain the initial membership degree.

[0012] S6. Collect the real-time operating parameters of key indicators that have a significant impact on the equipment status at time t2 and input them into the dynamic cloud object model to obtain new real-time membership degrees.

[0013] S7. By comparing the initial membership degree with the real-time membership degree, the current status of the intelligent power equipment is output, and the development trend of the indicators is predicted based on the evaluation index system.

[0014] Preferably, the dynamic filtering of data from intelligent power devices in step S2 includes the following steps:

[0015] The data collected in step S1 is standardized to ensure that the data of different indicators have the same scale.

[0016] Calculate the covariance matrix based on the standardized data;

[0017] Eigenvalue decomposition is performed on the covariance matrix to obtain eigenvalues ​​and eigenvectors;

[0018] The number of principal components is determined based on the magnitude of the eigenvalues, and principal components with larger contribution rates are selected.

[0019] By calculating the loading coefficient of each original indicator in the principal component, the weight of the indicator is determined. Indicators with larger weights contribute more to the principal component and can be regarded as key indicators.

[0020] Preferably, the adaptive adjustment process of the key indicators in step S3 includes the following steps:

[0021] The key indicators obtained in step S2 are identified using data analysis and fault diagnosis techniques, and the fault type is determined.

[0022] For each fault type, the Pearson correlation coefficient statistical method was used to analyze the correlation between each indicator and the fault.

[0023] Weight adjustment rules are defined based on the relevance of the indicators;

[0024] The adjusted weights are normalized to ensure that the sum of all indicator weights is 1.

[0025] Preferably, the process of processing the indicators obtained from the cloud object model in step S4 includes the following steps:

[0026] Construct a cloud model based on the key indicator data obtained in step S2;

[0027] Generate cloud droplets for key metrics using the forward cloud generator algorithm of the cloud model;

[0028] The classical domain and nodal domain of the dynamic matter-element model are determined based on the expectation and entropy of cloud droplets.

[0029] Preferably, the data processing method used in step S6 for processing the dynamic cloud object model is the exponential weighted average method.

[0030] Preferably, the step of formulating the evaluation index system for intelligent power equipment in step S4 includes:

[0031] Key indicator data are collected based on the key indicators when the equipment is working normally.

[0032] The expected value and entropy of key indicator data are calculated, and the data range of key indicator data when the equipment is working normally is determined based on the expected value and entropy.

[0033] The evaluation index system for intelligent power equipment is determined based on the data range.

[0034] Preferably, the process of standardizing the data includes the following steps:

[0035] The standardization method was determined to be the Z-score standardization method;

[0036] For the historical operating parameter data of the collected smart power equipment, calculate the mean and variance of the historical operating parameter data for each indicator;

[0037] For each data point of each indicator, calculations are performed according to a standardized formula, so that the data of different indicators have the same scale.

[0038] Preferably, the process of analyzing the correlation between each indicator and the fault includes the following steps:

[0039] Randomly extract some historical operating parameters of this indicator from historical data;

[0040] By combining the status of the intelligent power equipment, the historical operating parameters of this part are analyzed and fault data is obtained. The fault data is represented in binary, with 1 recorded if a fault occurs and 0 recorded if no fault occurs.

[0041] Calculate the average of historical operating parameters and the average of fault conditions;

[0042] The Pearson correlation coefficient between this index and the fault type was calculated.

[0043] Preferably, the data processing procedure for the dynamic cloud object model in step S5 includes the following steps:

[0044] The weighting parameter α is determined based on the equipment's operational stability and the rate of data change.

[0045] The real-time operating parameters obtained at time t1 are named the initial data, and the center value E of the classical domain of the initial data is obtained. x1 The range of the classical domain kE n1 and the range H of the segment. e1 ;

[0046] Record the real-time running data x at time t2;

[0047] The classical domain and nodal domain are analyzed and processed using the triangular membership function to obtain the membership degree.

[0048] Compared with related technologies, the intelligent power equipment condition evaluation method provided by this invention has the following beneficial effects:

[0049] 1. This invention provides a method for evaluating the condition of intelligent power equipment. In specific implementation, by filtering out data that has low correlation with faults of intelligent power equipment, the amount of data processing is reduced and the efficiency of data processing is improved.

[0050] 2. At the same time, the constructed dynamic matter-element model can efficiently compare the real-time operating parameters of intelligent power equipment at multiple times, clearly present the current state of intelligent power equipment, and accurately predict the development trend of equipment state.

[0051] 3. The dynamic matter-element model also has a powerful data memory function, which greatly simplifies the tedious process of multiple data processing steps. Attached Figure Description

[0052] Figure 1 A wireframe flowchart illustrating the overall steps of the intelligent power equipment condition evaluation method provided by this invention.

[0053] Figure 2 A wireframe flowchart illustrating the dynamic screening process for the intelligent power equipment condition evaluation method provided by this invention.

[0054] Figure 3 A wireframe flowchart for constructing a dynamic cloud object model for the intelligent power equipment condition evaluation method provided by this invention;

[0055] Figure 4 A wireframe flowchart illustrating the adaptive adjustment of the intelligent power equipment condition evaluation method provided by this invention.

[0056] Figure 5 A wireframe flowchart illustrating the equipment condition prediction method for the intelligent power equipment condition assessment provided by this invention. Detailed Implementation

[0057] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0058] Please refer to the following: Figure 1 — Figure 5 ,in, Figure 1 A wireframe flowchart illustrating the overall steps of the intelligent power equipment condition evaluation method provided by this invention. Figure 2 A wireframe flowchart illustrating the dynamic screening process for the intelligent power equipment condition evaluation method provided by this invention. Figure 3 A wireframe flowchart for constructing a dynamic cloud object model for the intelligent power equipment condition evaluation method provided by this invention; Figure 4 A wireframe flowchart illustrating the adaptive adjustment of the intelligent power equipment condition evaluation method provided by this invention. Figure 5 A wireframe flowchart illustrating the equipment condition prediction method for the intelligent power equipment condition assessment provided by this invention.

[0059] In its implementation, a method for evaluating the condition of intelligent power equipment includes the following steps:

[0060] S1. Collect historical operating parameters of intelligent power equipment and classify the historical operating parameters according to the operating status of the equipment;

[0061] S2. The historical operating parameters of intelligent power equipment when it fails are dynamically screened according to the influence of key indicators to identify key indicators that have a significant impact on the equipment status, and an evaluation index system for intelligent power equipment is developed based on the key indicators.

[0062] S3. Determine the weight of each key indicator based on the correlation between key indicators and equipment failure, and make adaptive adjustments.

[0063] S4. A new dynamic cloud matter-element model is formed by combining cloud model and dynamic matter-element: Based on the evaluation index system of key indicators established in step S2, a cloud model is constructed. Then, the cloud model's forward cloud generator algorithm is used to generate cloud droplets of key indicators. Then, based on the expectation and entropy of the cloud droplets, the classical domain and section domain of the dynamic matter-element model are determined, thereby completing the construction of the dynamic cloud matter-element model.

[0064] S5. Input the key indicators that have a significant impact on the equipment status in step S2 into the dynamic cloud object model to obtain the initial membership degree.

[0065] S6. Collect the real-time operating parameters of the key indicators that have a significant impact on the equipment status in step S2 and input them into the dynamic cloud object model to obtain new real-time membership degrees.

[0066] S7. By comparing the initial membership degree with the real-time membership degree, the current status of the intelligent power equipment is output, and the development trend of the indicators is predicted based on the evaluation index system.

[0067] It should be noted that the dynamic filtering of data from smart power devices in step S2 includes the following steps:

[0068] The method for standardizing the data collected in step S1 is determined to be the Z-score standardization method.

[0069] For the historical operating parameter data of the collected smart power equipment, calculate the mean and variance of the historical operating parameter data for each indicator;

[0070] For each data point of each indicator, calculations are performed according to a standardized formula, so that the data of different indicators have the same scale;

[0071] Calculate the covariance matrix based on the standardized data;

[0072] Eigenvalue decomposition is performed on the covariance matrix to obtain eigenvalues ​​and eigenvectors;

[0073] The number of principal components is determined based on the magnitude of the eigenvalues, and principal components with larger contribution rates are selected.

[0074] By calculating the loading coefficient of each original indicator in the principal component, the weight of the indicator is determined. Indicators with larger weights contribute more to the principal component and can be regarded as key indicators.

[0075] Step S4, which involves developing an evaluation index system for smart power equipment, includes the following steps:

[0076] A cloud model is constructed based on key indicator data collected by key indicator acquisition equipment during normal operation.

[0077] The expected value and entropy of the historical operating parameters of key indicators are calculated, and the data range of key indicator data when the equipment is working normally is determined based on the expected value and entropy.

[0078] The evaluation index system for intelligent power equipment is determined based on the data scope. This is achieved by defining the scope of multiple key indicators.

[0079] The adaptive adjustment process for key indicators in step S3 includes the following steps:

[0080] The key indicators obtained in step S2 are identified using data analysis and fault diagnosis techniques, and the fault type is determined.

[0081] For each fault type, randomly extract n historical operating parameters X = {x1, x2, x3, ..., xn} from the historical data of that fault type. n-1 ,x n}, and the equipment status F = {f1, f2, f3, ..., f} corresponding to the historical operating parameters. n-1 ,f n} Calculate the average of historical operating parameters and the average of the fault conditions And the Pearson correlation coefficient r,

[0082] When the absolute value of the Pearson correlation coefficient r is less than 0.3, it is generally considered to be uncorrelated; between 0.3 and 0.5, it is generally considered to be low correlation; between 0.5 and 0.8, it is considered to be moderate correlation; and above 0.8, it is considered to be high correlation.

[0083] For example: Extract random data of oil temperature from 10 smart power devices, as follows:

[0084] X={50, 55, 60, 52, 58, 53, 57, 54, 56, 51}

[0085] The equipment status is represented in binary. A value of 0 indicates normal operation, and a value of 1 indicates a fault. The fault data corresponding to the oil temperature mentioned above is as follows:

[0086] F={0, 0, 1, 0, 0, 0, 1, 0, 0, 0}

[0087] The average historical operating parameters and the average fault conditions are calculated as follows:

[0088] Average oil temperature:

[0089]

[0090] Average number of failure scenarios:

[0091]

[0092] Pearson correlation coefficient:

[0093]

[0094] Then, by calculation, the Pearson correlation coefficient r≈0.51 between this index and the fault type was obtained. Therefore, it can be determined that oil temperature is moderately correlated with equipment faults, that is, oil temperature may have a certain effect on equipment faults.

[0095] The weight adjustment rules are defined based on the correlation of the indicators. Based on the Pearson correlation coefficient of the key indicator type obtained in step S2, the adjusted weights are normalized to ensure that the sum of the weights of all indicators is 1.

[0096] The processing steps for the indicators obtained from the cloud object model in step S4 include the following steps:

[0097] A cloud model is constructed based on the mean and standard deviation of the historical operating parameters obtained in step S2;

[0098] The cloud model's forward cloud generator algorithm generates cloud droplets representing key indicators, and the expected value E of the cloud droplets is obtained by calculating historical data and real-time operating parameters. x Entropy E n ;

[0099] The classical domain and nodal domain of the dynamic matter-element model are determined based on the expectation and entropy of cloud droplets, with the expectation of cloud droplets used as the intermediate value E of the classical domain. x1 The range of the classical domain is (E x -k1E n E x +k1E n k1 is determined based on the Pearson correlation coefficient r. When the absolute value of the Pearson correlation coefficient r is between 0.3 and 0.5, k1 can be 2; when the absolute value of r is between 0.5 and 0.8, k1 can be 1.5; and when the absolute value of r is above 0.8, k1 can be 1. This ensures the range of the classical domain. The range of the section domain is (E... x -k2E n E x +k2E n k2 is determined based on the historical data of the equipment, the operating environment, and the correlation between the indicator and the equipment status. The value of k2 should be such that the historical data of the smart power equipment is exactly within the range of the node. k2 can be appropriately increased according to the operating environment of the smart power equipment.

[0100] The classical domain and section domain of the dynamic matter-element model are updated in real time based on the new data collected.

[0101] In step S5, the data processing method used for processing the dynamic cloud object model is the exponential weighted average method.

[0102] Step 5, the data processing procedure for the dynamic cloud object model, includes the following steps:

[0103] The weighting parameter α is determined based on the equipment's operational stability and the rate of data change.

[0104] Record the obtained historical running parameters and name them as initial data, and calculate the center value E of the classical domain of the initial data. x1 Classical Domain E n1 Scope (E) x -k1E n1 E x +k1E n1 ) and the domain H e1 Scope (E) x -k2E n1 E x +k2E n1 );

[0105] After a period of time, the real-time running data x at time t2 is recorded and the dynamic matter-element model is updated. The relevant data of the updated real-time running data x are as follows:

[0106] The central value of a classical field: E x2 =αx+(1-α)E x1 ,

[0107] entropy:

[0108] That is, the classical domain range of the device operating parameters at time t2 is (E x2 -k1E n2 E x2 +k1E n2 );

[0109] That is, the range of the device operating parameters at time t2 is (E x2 -k2E n2 E x2 +k2E n2 ).

[0110] The obtained classical domain and nodal domain are analyzed and processed by combining the triangular membership function to obtain the membership degree;

[0111] Substitute the obtained real-time operating parameter x into the above formula to obtain the real-time membership degree of the indicator at this time.

[0112] The process of analyzing the obtained data in step S6 includes the following steps:

[0113] The initial membership degree of the initial data after the filtering in step S5 at time t1 is obtained by processing the dynamic matter-element model;

[0114] The new data at time t2 after the filtering in step S2 are processed by the dynamic matter-element model to obtain new membership degrees;

[0115] By comparing the real-time membership degree with the initial membership degree, the current operating status of the intelligent power equipment and the development trend of the operating status can be obtained.

[0116] It should be noted that, assuming the weight parameter α = 0.3, the real-time operating parameter at time t1 is x = 56℃, and the real-time operating parameter at time t2 is x = 58℃, from the above example, k1 = 1.5, E x =54.6, E n1 ≈3.1, H e1 Since r≈0.51, this indicates that the indicator is moderately correlated with whether the equipment is working properly. Therefore, k2 can be taken as 2.5, and E x1 =54.6, Classical domain E n1 The range is (54.6-1.5×3.1, 54.6+1.5×3.1), which is (49.95, 59.25), and the node range H is... e1 The range is (54.6-2.5×3.1, 54.6+2.5×3.1), which is (46.85, 62.35). It is stipulated that when the equipment temperature is within the range (46.85, 52.25), the equipment is considered to be in poor operating condition; when the equipment temperature is within the range (49.95, 59.25), the equipment is considered to be in good operating condition; and when the equipment temperature is within the range (55.95, 62.35), the equipment is considered to be in possible faulty condition.

[0117] When the equipment is in poor operating condition: the membership degree increases linearly from 0 to 1 within the interval (46.85, 50.85), and decreases linearly from 1 to 0 within the interval (50.85, 52.25).

[0118] When the equipment is in good operating condition: the membership degree increases linearly from 0 to 1 within the interval (49.95, 54.6), and decreases linearly from 1 to 0 within the interval (54.6, 59.25);

[0119] When equipment malfunctions, the membership degree is specified to increase linearly from 0 to 1 within the interval (55.95, 59.25) and decrease linearly from 1 to 0 within the interval (59.25, 62.35).

[0120] Then, by calculating the membership function of the triangle, it is found that when the temperature x = 56℃, the membership degree of the equipment with poor operating status is 0; the membership degree of the equipment with good operating status is 0.7; and the membership degree of the equipment that may fail is 0.02. Therefore, it is considered that when the temperature is 56℃, the equipment is in good operating status.

[0121] Then, the temperature of 56℃ was recorded in the historical data, and the classical domain of the equipment operating parameters at time t2 was calculated to be (51.35, 58.7), and the sub-domain of the equipment operating parameters at time t2 was (45.82, 64.22). The evaluation index system was then updated. The updated evaluation index system is as follows: when the equipment temperature is in the range (45.82, 54.65), the equipment is considered to be in poor operating condition; when the equipment temperature is in the range (51.35, 58.7), the equipment is considered to be in good operating condition; when the equipment temperature is in the range (55.4, 64.22), the equipment is considered to be in possible fault condition.

[0122] When the equipment is in poor operating condition: the membership degree increases linearly from 0 to 1 within the interval (45.82, 49.82), and decreases linearly from 1 to 0 within the interval (49.82, 51.35).

[0123] When the equipment is in good operating condition: the membership degree increases linearly from 0 to 1 within the interval (51.35, 55.02), and decreases linearly from 1 to 0 within the interval (55.02, 58.7).

[0124] When equipment malfunctions, the membership degree increases linearly from 0 to 1 within the interval (52.1, 59.4), and decreases linearly from 1 to 0 within the interval (59.4, 64.22).

[0125] Then, by using the triangular membership function, it was found that when the temperature x = 58℃, the membership degree for poor equipment operation was 0; the membership degree for good equipment operation was 0.2; and the membership degree for equipment failure was 0.8. Therefore, it was concluded that the equipment might fail when the temperature was 58℃.

[0126] Compared to time t1, the membership degree changed significantly at time t2. Therefore, it can be considered that the equipment is trending towards failure, and technicians need to promptly inspect and maintain the intelligent power equipment.

[0127] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general-purpose hardware platform, or of course by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the related technology, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0128] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0129] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for evaluating the condition of intelligent power equipment, characterized in that, Includes the following steps: S1. Collect historical operating parameters of intelligent power equipment and classify the historical operating parameters according to the operating status of the equipment; S2. Dynamically filter the historical operating parameters of intelligent power equipment when it fails based on the impact of key indicators, and screen out the key indicators that have a greater impact on the equipment status. S3. Determine the weight of each key indicator based on the correlation between key indicators and equipment failure, and make adaptive adjustments. S4. Based on the combination of cloud model and dynamic matter element, a new dynamic cloud matter element model is formed, and an evaluation index system for smart power equipment is formulated based on key indicators: The cloud model is constructed according to the evaluation index system of key indicators formulated in step S2. Then, the cloud droplets of key indicators are generated by using the forward cloud generator algorithm of the cloud model. Then, the classical domain and section domain of the dynamic matter element model are determined based on the expectation and entropy of the cloud droplets, thereby completing the construction of the dynamic cloud matter element model. S5. Input the key indicators that have a significant impact on the equipment status collected at time t1 into the dynamic cloud object model to obtain the initial membership degree. S6. Collect the real-time operating parameters of key indicators that have a significant impact on the equipment status at time t2 and input them into the dynamic cloud object model to obtain new real-time membership degrees. S7. By comparing the initial membership degree with the real-time membership degree, the current status of the intelligent power equipment is output, and the development trend of the indicators is predicted based on the evaluation index system.

2. The intelligent power equipment condition evaluation method according to claim 1, characterized in that, The dynamic filtering of data from smart power devices in step S2 includes the following steps: The data collected in step S1 is standardized to ensure that the data of different indicators have the same scale. Calculate the covariance matrix based on the standardized data; Eigenvalue decomposition is performed on the covariance matrix to obtain eigenvalues ​​and eigenvectors; The number of principal components is determined based on the magnitude of the eigenvalues, and principal components with larger contribution rates are selected. By calculating the loading coefficient of each original indicator in the principal component, the weight of the indicator is determined. Indicators with larger weights contribute more to the principal component and can be regarded as key indicators.

3. The intelligent power equipment condition evaluation method according to claim 1, characterized in that, The adaptive adjustment process for the key indicators in step S3 includes the following steps: The key indicators obtained in step S2 are identified using data analysis and fault diagnosis techniques, and the fault type is determined. For each fault type, the Pearson correlation coefficient statistical method was used to analyze the correlation between each indicator and the fault. Weight adjustment rules are defined based on the relevance of the indicators; The adjusted weights are normalized to ensure that the sum of all indicator weights is 1.

4. The intelligent power equipment condition evaluation method according to claim 1, characterized in that, The process of processing the indicators obtained from the cloud object model in step S4 includes the following steps: Construct a cloud model based on the evaluation index system of key indicators established in step S2; Generate cloud droplets for key metrics using the forward cloud generator algorithm of the cloud model; The classical domain and nodal domain of the dynamic matter-element model are determined based on the expectation and entropy of cloud droplets.

5. The intelligent power equipment condition evaluation method according to claim 1, characterized in that, The data processing method used in step S5 for processing the dynamic cloud object model is the exponential weighted average method.

6. The method for evaluating the condition of intelligent power equipment according to claim 1, characterized in that, The steps in step S4 for developing an evaluation index system for intelligent power equipment include: Key indicator data are collected based on the key indicators when the equipment is working normally. The expected value and entropy of key indicator data are calculated, and the data range of key indicator data when the equipment is working normally is determined based on the expected value and entropy. The evaluation index system for intelligent power equipment is determined based on the data range.

7. The intelligent power equipment condition evaluation method according to claim 2, characterized in that, The process of standardizing the data includes the following steps: The standardization method was determined to be the Z-score standardization method; For the historical operating parameter data of the collected smart power equipment, calculate the mean and variance of the historical operating parameter data for each indicator; For each data point of each indicator, calculations are performed according to a standardized formula, so that the data of different indicators have the same scale.

8. The method for evaluating the condition of intelligent power equipment according to claim 3, characterized in that, The process of analyzing the correlation between various indicators and faults includes the following steps: Randomly extract some historical operating parameters of this indicator from historical data; By combining the status of the intelligent power equipment, the historical operating parameters of this part are analyzed and fault data is obtained. The fault data is represented in binary, with 1 recorded if a fault occurs and 0 recorded if no fault occurs. Calculate the average of historical operating parameters and the average of fault conditions; The Pearson correlation coefficient between this index and the fault type was calculated.

9. The method for evaluating the condition of intelligent power equipment according to claim 5, characterized in that, The data processing procedure for the dynamic cloud object model in step S5 includes the following steps: The weighting parameter α is determined based on the equipment's operational stability and the rate of data change. The real-time operating parameters obtained at time t1 are named the initial data, and the center value of the classical domain, the range of the classical domain, and the range of the section domain of the initial data are obtained. Record the real-time running data x at time t2; The classical domain and nodal domain are analyzed and processed using the triangular membership function to obtain the membership degree.

Images