Diagnosis method based on switch cabinet

Abstract

The invention relates to the technical field of electrical equipment diagnosis, and discloses a diagnosis method based on a switch cabinet, and the method is characterized in that the method comprises the following steps: S1, obtaining the partial discharge data and temperature data of the switch cabinet; s2, performing corresponding feature extraction on the partial discharge data and the temperature data; and S3, establishing an evaluation model, inputting the features extracted from the partial discharge data and the features extracted from the temperature data into the evaluation model, and outputting a diagnosis result by the evaluation model. The method can make correct judgments in the case of complex faults, and improves the accuracy of fault diagnosis.

Description

Technical Field

[0001] The present application relates to the technical field of power equipment diagnosis, and in particular to a switchgear-based diagnosis method. Background Art

[0002] Accurately monitoring the operating status of 10kV switchgear is crucial in power systems. Conventional switchgear monitoring systems typically use a single sensor to detect the switchgear's operating status. Each sensor detects the switchgear's parameters, and the data from each sensor is processed and displayed independently. This lacks cross-correlation analysis between the data. Existing technologies are insensitive to complex switchgear faults (such as insulation aging accompanied by localized overheating), making them prone to misdiagnosis or missed diagnosis in complex fault scenarios, impacting the accuracy of fault diagnosis. Summary of the Invention

[0003] This application provides a switch cabinet-based diagnostic method, which aims to improve the ability to make correct judgments when faced with complex fault situations and improve the accuracy of fault diagnosis.

[0004] In one solution, a switchgear-based diagnostic method is provided, comprising the following steps:

[0005] S1. Obtaining partial discharge data and temperature data of the switchgear;

[0006] S2. performing corresponding feature extraction on the partial discharge data and the temperature data;

[0007] S3. Establish an evaluation model, input the features extracted from the partial discharge data and the features extracted from the temperature data into the evaluation model to perform fusion processing on the features, and the evaluation model outputs a diagnosis result.

[0008] In one embodiment, the feature extraction of the partial discharge data includes phase asymmetry, skewness, and kurtosis; the phase asymmetry is obtained by the formula Obtain, among which, and Correspondingly represents the average discharge amount in the positive half cycle and the average discharge amount in the negative half cycle; and They correspond to the positive half-cycle standard deviation and the negative half-cycle standard deviation respectively; the skewness is expressed by the formula S k = E[( φ-μ ) 3 ] σ 3 Obtain, among which, is the discharge pulse phase, μ is the phase mean, is the standard deviation; the kurtosis is expressed by the formula K u = E[( φ-μ ) 4 ] σ 4 - 3 Obtain, among which is the discharge pulse phase, μ is the phase mean, is the standard deviation.

[0009] Specifically, Reflects the quantized positive and negative half-cycle phase offset, A large value (e.g. > 0.3) indicates asymmetric discharge (e.g. creeping discharge). Reflecting the symmetry of the phase distribution, Left deviation (such as corona discharge), Right bias (such as internal air gap discharge), Symmetry (normal or slightly imperfect). Reflects the concentration of discharge, Sharp (clustered discharges, such as metal particles), Smooth (discharge dispersion, such as moisture), Close to normal (normal).

[0010] In one solution, environmental data is also acquired in step S1, and the features extracted from the partial discharge data are corrected according to predetermined data based on the environmental data.

[0011] In one embodiment, the environmental data includes environmental temperature data and environmental humidity data; the predetermined data is obtained by the formula To determine, Features extracted from the PD data 、 、 ; is the environmental variable factor;

[0012] ; T is the real-time temperature; is the reference temperature; is the maximum temperature; H is the real-time humidity; is the reference humidity; Maximum humidity.

[0013] In one embodiment, the environmental data also includes current data of the switch cabinet; the theoretical temperature rise data of the switch cabinet is obtained based on the current data, and the actual temperature rise data is obtained based on the temperature data; the temperature rise coefficient is obtained based on the ratio of the theoretical temperature rise data to the actual temperature rise data; the temperature rise coefficient is input into the evaluation model and the corresponding diagnostic result is output.

[0014] In one embodiment, the theoretical temperature rise data is obtained by the formula Obtain, among which, is the thermal resistance coefficient (K / W); is the real-time current; is the contact resistance (Ω); is the dielectric loss compensation coefficient.

[0015] In one embodiment, the actual temperature rise data is obtained by the formula Obtain, among which, is the ambient temperature, The real-time temperature.

[0016] Specifically, the actual temperature rise data is also obtained by the set thermal imager, that is, .

[0017] In one embodiment, the temperature rise coefficient is given by the formula Obtain, among which is the actual temperature rise value, is the theoretical temperature rise value.

[0018] In one embodiment, the partial discharge data is provided by a TEV sensor, a UHF sensor, and an AE sensor installed in the switch cabinet; the temperature data is provided by multiple RFID sensors and a thermal imager; the multiple RFID sensors are respectively installed on the upper contact arms and lower contact arms of the A-phase, B-phase, and C-phase circuits of the switch cabinet; and the thermal imager can provide thermal imaging data of the entire switch cabinet.

[0019] In one solution, weight calculation is performed based on the temperature data and the partial discharge data to obtain a comprehensive weight, a weighting matrix is obtained based on the comprehensive weight, a positive ideal solution and a negative ideal solution are obtained based on the weighting matrix, a progress of the patch is obtained based on the positive ideal solution and the negative ideal solution, and a health degree is obtained based on the progress of the patch. The health degree is calculated by the health degree rule formula. Output the corresponding health status.

[0020] Beneficial effects of this application:

[0021] By extracting features from temperature and partial discharge data and constructing an evaluation model, the features are input into the evaluation model, which then fuses the input feature data to perform cross-correlation analysis and output the corresponding diagnostic results. By fusing multi-source feature data, the evaluation model can comprehensively grasp the operating status of the equipment, identify potential complex fault hazards, and increase sensitivity to complex faults in switchgear (such as insulation aging accompanied by local overheating). This application method can make accurate judgments when faced with complex fault situations, improving the accuracy of fault diagnosis and enhancing the reliability, stability, and predictability of equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0023] Figure 1 This is a flow chart of a diagnostic method in one embodiment of the present application;

[0024] Figure 2 This is a schematic diagram of the detection process of the diagnostic method in one embodiment of the present application;

[0025] Figure 3 This is a comparison chart of actual temperature rise and actual temperature rise in one embodiment of the present application;

[0026] Figure 4 This is a diagnostic method display page in an embodiment of the present application; DETAILED DESCRIPTION

[0027] The specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application, but are not intended to limit the scope of the present application. Similarly, the following examples are only some embodiments of the present application and not all embodiments. All other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of this application.

[0028] In the present invention, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples without contradiction.

[0029] This application makes improvements and innovations and proposes the following embodiments.

[0030] In some embodiments, see Figure 1-3 , provides a switchgear-based diagnostic method, including the following steps:

[0031] S1. Obtaining partial discharge data and temperature data of the switchgear;

[0032] S2. performing corresponding feature extraction on the partial discharge data and the temperature data;

[0033] S3. Establish an evaluation model, input the features extracted from the partial discharge data and the features extracted from the temperature data into the evaluation model to perform fusion processing on the features, and the evaluation model outputs a diagnosis result.

[0034] By extracting features from temperature and partial discharge data and constructing an evaluation model, the features are input into the evaluation model, which then fuses the input feature data to perform cross-correlation analysis and output the corresponding diagnostic results. By fusing multi-source feature data, the evaluation model can comprehensively grasp the operating status of the equipment, identify potential complex fault hazards, and increase sensitivity to complex faults in switchgear (such as insulation aging accompanied by local overheating). This application method can make accurate judgments when faced with complex fault situations, improving the accuracy of fault diagnosis and enhancing the reliability, stability, and predictability of equipment operation.

[0035] In one embodiment, the feature extraction of the partial discharge data includes phase asymmetry, skewness, and kurtosis; the phase asymmetry is obtained by the formula Obtain, among which, and Correspondingly represents the average discharge amount in the positive half cycle and the average discharge amount in the negative half cycle; and They correspond to the positive half-cycle standard deviation and the negative half-cycle standard deviation respectively;

[0036] The skewness is given by the formula S k = E[( φ-μ ) 3 ] σ 3 Obtain, among which, is the discharge pulse phase, μ is the phase mean, is the standard deviation;

[0037] The kurtosis is given by the formula K u = E[( φ-μ ) 4 ] σ 4 - 3 Obtain, among which is the discharge pulse phase, μ is the phase mean, is the standard deviation.

[0038] Specifically, Reflects the quantized positive and negative half-cycle phase offset. When it is large (e.g. > 0.3), it indicates asymmetric discharge (e.g. creeping discharge).

[0039] Reflecting the symmetry of the phase distribution, Left deviation (such as corona discharge), Right deviation (such as internal air gap discharge), when Symmetry (normal or slightly imperfect).

[0040] Reflects the concentration of discharge, when Sharp time-frequency bands indicate discharge clusters, such as metal particles. Smooth, indicating that the discharge is dispersed, such as when it is damp. The frequency band is close to normal, indicating a normal state.

[0041] In one solution, environmental data is also obtained in step S1, and the features extracted from the partial discharge data are corrected according to predetermined data based on the environmental data. By correcting the features extracted from the partial discharge data, the output result can be adjusted more reasonably.

[0042] In one embodiment, the environmental data includes environmental temperature data and environmental humidity data; the predetermined data is obtained by the formula To determine, Features extracted from the PD data 、 、 ; is the environmental variable factor;

[0043] ; T is the real-time temperature; is the reference temperature; is the maximum temperature; H is the real-time humidity; is the reference humidity; Maximum humidity. Depending on the PD characteristic data, the corresponding corrections vary to ensure that the final output is reasonable and normal. Based on continuous experiments and calculations, the temperature coefficient of the environmental variable factor is set to 0.2 and the humidity coefficient is set to 0.8 to provide accurate correction data for the PD data characteristics, ensuring that the final output is normal and reasonable.

[0044] In one embodiment, the environmental data also includes the switchgear current data; theoretical temperature rise data for the switchgear is obtained based on the current data, and actual temperature rise data is obtained based on the temperature data; a temperature rise coefficient is obtained based on the ratio of the theoretical temperature rise data to the actual temperature rise data; and the temperature rise coefficient is input into the evaluation model to output a corresponding diagnostic result. Obtaining the temperature coefficient provides a data source for the evaluation model input, making the final output structure of the evaluation model more reasonable and accurate.

[0045] The current data is collected through the current transformer.

[0046] In one embodiment, the theoretical temperature rise data is obtained by the formula Obtain, among which, The theoretical temperature rise is obtained according to the thermal effect formula of current, but with the addition of , which can more reasonably obtain the theoretical temperature rise.

[0047] In one embodiment, the actual temperature rise data is obtained by the formula Obtain, among which, is the ambient temperature, The real-time temperature can be provided by a temperature sensor provided on the circuit.

[0048] Specifically, the actual temperature rise data is also obtained by the set thermal imager, that is, .

[0049] In one embodiment, the temperature rise coefficient is given by the formula Obtain, among which is the actual temperature rise value, is the theoretical temperature rise value. This temperature rise coefficient can be directly input into the evaluation model, providing the corresponding data source for the evaluation model.

[0050] Specifically, Figure 3 The theoretical and actual temperature rise data are shown in Figure 3 After the test starts, the theoretical temperature rise value and the actual temperature rise value will deviate over time, but the overall trend is consistent. As long as the deviation value is within a certain range (i.e. the yellow line part, the deviation value is ≤10℃), it is normal.

[0051] In one solution, the partial discharge data is provided by a TEV (Transient Earth Voltage Sensor), a UHF (Ultra-High Frequency Partial Discharge Sensor), and an AE (Acoustic Emission) sensor installed in the switchgear. The temperature data is provided by multiple RFID (Radio Frequency Identification Temperature Sensors) and a thermal imager. The multiple RFID sensors are respectively installed on the upper and lower contact arms of the A-phase, B-phase, and C-phase circuits of the switchgear. The thermal imager can provide thermal imaging data of the entire switchgear.

[0052] Specifically, the TEV sensor has a bandwidth of 3-100 MHz, a sampling rate of 100 MS / s, and a measurement range of 0-60 dBmV;

[0053] The frequency band of the UHF sensor is 300-1500MHz, and its detection sensitivity is ≤-80dBm;

[0054] The AE sensor has a center frequency of 40kHz and a measurement range of 20-80dB. The three PD sensors cover as many frequency bands as possible, increasing the detection range and enabling more sensitive detection of PD signals.

[0055] In one solution, weight calculation is performed based on the temperature data and the partial discharge data to obtain a comprehensive weight, a weighting matrix is obtained based on the comprehensive weight, a positive ideal solution and a negative ideal solution are obtained based on the weighting matrix, a progress of the patch is obtained based on the positive ideal solution and the negative ideal solution, and a health degree is obtained based on the progress of the patch. The health degree is calculated by the health degree rule formula. Output the corresponding health status.

[0056] Specifically, the evaluation model is implemented by formula An evaluation matrix is constructed for the partial discharge data; by the formula An evaluation matrix is constructed for the temperature data, where: is the upper contact arm temperature, is the lower contact arm temperature, is the thermal imaging temperature of one phase in the A-phase, B-phase, and C-phase circuits of the switch cabinet; wherein TEV represents the phase asymmetry corresponding to the TEV sensor , skewness and kurtosis UHF represents the phase asymmetry corresponding to the UHF sensor. , skewness and kurtosis AE represents the phase asymmetry corresponding to the AE sensor. , skewness and kurtosis data.

[0057] When calculating the weight, first calculate the information entropy of the partial discharge data and temperature data respectively. ,in ; Then perform conflict calculation on the partial discharge data and temperature data, ;in, is the correlation coefficient of feature j, k.

[0058] According to the obtained partial discharge information entropy and conflict, as well as the temperature information entropy and conflict, the formula Perform comprehensive weight calculation.

[0059] The weighted matrix is given by the formula get;

[0060] The positive ideal solution is given by the formula R + =[ max V 1j , max V 2j , max V 3j ] get;

[0061] The negative ideal solution is obtained by R - =[ min V 1j , min V 2j , min V 3j ] get;

[0062] The closeness is calculated by the formula get;

[0063] Health is calculated by the formula get.

[0064] By obtaining the information entropy and conflict of partial discharges, as well as the information entropy and conflict of temperature, and performing a comprehensive weighted calculation, the corresponding weighting is applied. The weight distribution of temperature data and monthly partial discharge data is adjusted to obtain reasonable positive and negative ideal solutions, thereby obtaining the corresponding patching progress and ultimately the corresponding health. The corresponding results are output according to the health rules.

[0065] In one scenario, Figure 2 It shows the health value of the output after the switch cabinet is tested using the method of this application. Figure 2 The blue line is the real-time health value line, and the yellow line is the "attention" threshold line of the health formula. 0.25; the orange line is the "warning" threshold line of the health formula ( 0.45); the red line is the “failure” threshold line of the health formula ( 0.7). The corresponding yellow, orange, and red line thresholds can be adjusted according to customer needs to meet different switchgear detection requirements.

[0066] Figure 2 As shown in the figure, as the health value exceeds yellow, it will decrease to below the yellow threshold line. The reason is that when the health value reaches the yellow line, the evaluation model will output the "attention" status to remind the staff to take corresponding measures. After the treatment is completed, the real-time health value will naturally drop to below the yellow line.

[0067] Figure 4 What is displayed is the final result display page of the switch cabinet using the method of the present application, and all the results output by the evaluation model can be found in this page.

[0068] Specifically, when the health is tested three times in a row When the health level is ≤0.25, it indicates that the device is in a “healthy” state; when the health level rises to = 0.45 triggers the "early warning" state (need to increase the monitoring frequency); when = 0.7, it enters the "fault" state (requiring power outage for maintenance).

[0069] The threshold adjustment is shown in the following table;

[0070]

[0071] For example, in a scenario where the ambient humidity is over 90% and the load rate is 95%, =0.38 S. The original evaluation model was to trigger the "attention state"; however, after adjusting the threshold according to the table above: =0.4>0.38, therefore, the evaluation model outputs a diagnostic result of maintaining a "healthy state".

[0072] The above are merely optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application. Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. Those skilled in the art may make changes, modifications, replacements, and variations to the above embodiments within the scope of the present invention.

Claims

1. A diagnostic method based on a switch cabinet, characterized in that: The steps include: S1. Obtaining partial discharge data and temperature data of the switchgear; S2. performing corresponding feature extraction on the partial discharge data and the temperature data; S3. Establish an evaluation model, input the features extracted from the partial discharge data and the features extracted from the temperature data into the evaluation model to perform fusion processing on the features, and the evaluation model outputs a diagnosis result.

2. The diagnostic method according to claim 1, wherein The feature extraction of the partial discharge data includes phase asymmetry, skewness, and kurtosis; the phase asymmetry is obtained by the following formula: in, and Correspondingly represents the average discharge amount in the positive half cycle and the average discharge amount in the negative half cycle; and They correspond to the positive half-cycle standard deviation and the negative half-cycle standard deviation respectively; The skewness is obtained by the following formula: in, is the discharge pulse phase, μ is the phase mean, is the standard deviation; The kurtosis is obtained by the formula; in, is the discharge pulse phase, μ is the phase mean, is the standard deviation.

3. The diagnostic method according to claim 2, characterized in that In step S1 , environmental data is also acquired, and features extracted from the partial discharge data are corrected according to predetermined data based on the environmental data.

4. The diagnostic method according to claim 3, characterized in that The environmental data includes environmental temperature data and environmental humidity data; The predetermined data is determined by the following formula: in, Features extracted from the PD data 、 、 ; is the environmental variable factor; Wherein, T is the real-time temperature; is the reference temperature; is the maximum temperature; H is the real-time humidity; is the reference humidity; Maximum humidity.

5. The diagnostic method according to claim 4, characterized in that The environmental data also includes current data of the switch cabinet; Acquire theoretical temperature rise data of the switch cabinet according to the current data, and acquire actual temperature rise data according to the temperature data; Obtaining a temperature rise coefficient according to a ratio of the theoretical temperature rise data to the actual temperature rise data; After the temperature rise coefficient is input into the evaluation model, a corresponding diagnosis result is output.

6. The diagnostic method according to claim 5, characterized in that The theoretical temperature rise data is obtained by the following formula: in, .

7. The diagnostic method according to claim 6, characterized in that The actual temperature rise data is obtained by the following formula: in, is the ambient temperature, The real-time temperature.

8. The diagnostic method according to claim 7, characterized in that The temperature rise coefficient is obtained by the following formula: in, is the actual temperature rise value, is the theoretical temperature rise value.

9. The diagnostic method according to claim 8, characterized in that The partial discharge data is provided by a TEV sensor, a UHF sensor and an AE sensor arranged in the switch cabinet; The temperature data is provided by a plurality of RFID sensors and a thermal imager; the plurality of RFID sensors are respectively installed on the upper contact arm and the lower contact arm of the A phase, B phase, and C phase circuits of the switch cabinet; The thermal imager can provide thermal imaging data of the entire switch cabinet.

10. The diagnostic method according to any one of claims 1 to 9, characterized in that Performing weight calculations based on the temperature data and the partial discharge data to obtain a comprehensive weight; Obtaining a weighting matrix according to the comprehensive weight, and obtaining a positive ideal solution and a negative ideal solution according to the weighting matrix; Obtaining a patch progress according to the positive ideal solution and the negative ideal solution, and obtaining a health degree according to the patch progress; The health is determined by the health rule Output the corresponding health status.

Images