A method and device for preventive testing of a capacitive voltage transformer

Abstract

The application discloses a kind of method and device for preventive testing of capacitive voltage transformer, it is related to voltage transformer preventive testing technical field, by training upper section error prediction model, middle section error prediction model and lower section error prediction model, read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to model, and using not to dismantle high voltage lead test method respectively obtains the upper section test dielectric loss value, middle section test dielectric loss value and lower section test dielectric loss value of voltage transformer to be tested, based on lower section error feature, lower section test dielectric loss value and lower section error prediction model, obtain lower section correction dielectric loss, and calculate out comprehensive dielectric loss;Ensure that when using non-high-precision dielectric loss instrument for preventive testing, also can reach the measurement effect of high-precision dielectric loss instrument, reduce the procurement cost of dielectric loss instrument.

Description

Technical Field

[0001] The present invention relates to the technical field of preventive testing of voltage transformers, and in particular to a method and device for preventive testing of a capacitive voltage transformer. Background Art

[0002] Preventive testing is an important part of the operation and maintenance of power equipment and an effective means to ensure the safe operation of power equipment. Preventive testing plays an important role in timely discovering and diagnosing equipment defects.

[0003] Whether it is high-voltage electrical equipment or safety equipment for live working, they all have their own insulation structures. During operation, these equipment and appliances are subject to internal and external overvoltages that are much higher than the normal rated operating voltage, which may cause defects in the insulation structure and become latent faults. On the other hand, as the insulation itself is heated and degraded due to aging under natural conditions during operation. Preventive testing is a complete set of systematic insulation performance diagnosis and testing methods developed to address these problems and possibilities and to prevent accidents caused by changes in the insulation performance of operating electrical equipment.

[0004] Current testing methods generally use a loss meter to regularly test the loss of voltage transformers to detect faults. However, during testing, the loss meter is often affected by environmental factors, external electromagnetic fields, and stray capacitance. Although some high-precision loss meters can avoid external interference and achieve high-precision measurements, such high-precision loss meters are often expensive. Therefore, a universal method for compensating for the loss error of loss meters with general accuracy is needed.

[0005] Chinese patent application publication number CN117607748A discloses a current or voltage transformer polarity test method, device, and medium; relating to the field of power testing technology; using a current / voltage sensor to collect a first signal and a second signal of a current / voltage transformer to be tested; performing a polarity logic judgment based on the first and second signals to obtain a polarity judgment result; reviewing the polarity judgment result; and outputting a successfully reviewed polarity judgment result; collecting the first and second signals of the current / voltage transformer to be tested by a sensor, and performing a polarity logic judgment based on the first and second signals to obtain a polarity judgment result, thereby avoiding manual climbing and outputting the polarity judgment result; and achieving non-contact direction detection; however, this solution fails to eliminate errors;

[0006] To this end, the present invention provides a method and device for preventive testing of a capacitor voltage transformer. Summary of the Invention

[0007] The present invention aims to address at least one of the technical problems existing in the prior art. To this end, the present invention provides a method and apparatus for preventive testing of capacitor voltage transformers. This method ensures that even when using a non-high-precision dielectric loss meter for preventive testing, the measurement results of a high-precision dielectric loss meter can be achieved, thereby reducing the procurement cost of the dielectric loss meter.

[0008] To achieve the above object, the present invention provides a method for preventive testing of a capacitor voltage transformer, comprising the following steps:

[0009] Step 1: Collect the upper section error training feature data and upper section error label data, the middle section error training feature data and middle section error label data, and the lower section error training feature data and lower section error label data for each dielectric loss meter in advance;

[0010] Step 2: For each type of dielectric loss meter:

[0011] Based on the error training feature data and the error label data of the previous section, the error prediction model of the previous section is trained; based on the error training feature data and the error label data of the middle section, the error prediction model of the middle section is trained; based on the error training feature data and the error label data of the next section, the error prediction model of the next section is trained;

[0012] Step 3: Collect the dielectric loss meter model of the voltage transformer to be tested, and read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to the model;

[0013] Step 4: Collect the upper section error characteristics, middle section error characteristics, and lower section error characteristics corresponding to the voltage transformer to be tested, and use the high-voltage lead test method without removing the high-voltage lead to obtain the upper section test dielectric loss value, middle section test dielectric loss value, and lower section test dielectric loss value of the voltage transformer to be tested respectively;

[0014] Step 5: Based on the error characteristics of the upper section, the dielectric loss value tested in the upper section, and the error prediction model of the upper section, obtain the corrected dielectric loss of the upper section; based on the error characteristics of the middle section, the dielectric loss value tested in the middle section, and the error prediction model of the middle section, obtain the corrected dielectric loss of the middle section; based on the error characteristics of the lower section, the dielectric loss value tested in the lower section, and the error prediction model of the lower section, obtain the corrected dielectric loss of the lower section;

[0015] Step 6: Calculate the comprehensive dielectric loss based on the corrected dielectric loss in the upper section, the middle section, and the lower section.

[0016] The method of collecting the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data for each dielectric loss meter is as follows:

[0017] Collect several groups of test voltage transformers for each type of dielectric loss meter, and collect the upper section error training feature set, middle section error training feature set, and lower section error training feature set corresponding to each group of test voltage transformers;

[0018] The feature elements in the upper section error training feature set include the temperature, humidity, stray capacitance, frequency mean, frequency variance, ground resistance and aging degree of the upper section capacitor of the test voltage transformer;

[0019] The feature elements in the middle section error training feature set include the temperature, humidity, frequency mean, frequency variance, ground resistance and aging degree of the middle section capacitor of the test voltage transformer;

[0020] The feature elements in the lower section error training feature set include the temperature, humidity, frequency mean, frequency variance, external electric field strength, grounding resistance and aging degree of the lower section capacitor of the test voltage transformer;

[0021] Use this type of dielectric loss meter to test the upper, middle, and lower capacitance of each set of test voltage transformers without removing the high-voltage leads to obtain the upper, middle, and lower capacitance values. The number of test voltage transformers collected by each type of dielectric loss meter is determined based on the actual number collected.

[0022] Then, a high-precision dielectric loss meter is used to measure the dielectric loss value of each test voltage transformer to obtain the upper section accurate dielectric loss value, the middle section accurate dielectric loss value and the lower section accurate dielectric loss value respectively;

[0023] For each set of tested voltage transformers, calculate the upper section test dielectric loss value divided by the upper section accurate dielectric loss value to obtain the upper section error label; calculate the middle section test dielectric loss value divided by the middle section accurate dielectric loss value to obtain the middle section error label; calculate the lower section test dielectric loss value divided by the lower section accurate dielectric loss value to obtain the lower section error label;

[0024] For each dielectric loss meter, all of its upper section error training feature sets constitute the upper section error training feature data, and all of its upper section error labels constitute the upper section error label data; for each dielectric loss meter, all of its middle section error training feature sets constitute the middle section error training feature data, and all of its middle section error labels constitute the middle section error label data; for each dielectric loss meter, all of its lower section error training feature sets constitute the lower section error training feature data, and all of its lower section error labels constitute the lower section error label data;

[0025] The method of testing the high voltage lead without removing it is as follows:

[0026] For the capacitor on the capacitor voltage transformer, from a safety perspective, the grounding switch cannot be opened without removing the high-voltage lead, so the top can only be grounded, and then the dielectric loss can be measured in combination with the reverse connection method; Figure 2 Shown is the experimental wiring diagram for testing the upper section capacitance using M4000.

[0027] For the middle section, the conventional connection method is adopted for measurement, and the measured data is the same as the disconnection data;

[0028] When measuring the capacitance of the lower section, the self-excitation method is used for measurement, and it is necessary to pay special attention to the insulation condition of the point;

[0029] The method of training the upper section error prediction model based on the upper section error training feature data and the upper section error label data; training the middle section error prediction model based on the middle section error training feature data and the middle section error label data; and training the lower section error prediction model based on the lower section error training feature data and the lower section error label data is as follows:

[0030] For each type of dielectric loss meter, the upper capacitance is:

[0031] Each group of upper section error training feature sets is used as input of an upper section error prediction model, the upper section error prediction model uses the predicted value of the upper section error corresponding to the upper section error training feature set as output, the upper section error label of the test voltage transformer corresponding to the upper section error training feature set is used as a prediction target, the difference between the predicted value of the upper section error and the upper section error label is used as a first prediction error, and minimizing the sum of squares of the first prediction errors is used as a training target; the upper section error prediction model is trained until the sum of squares of the first prediction errors reaches convergence and the training is stopped;

[0032] For each type of dielectric loss meter, the capacitance of the section is:

[0033] Each group of middle-section error training feature sets is used as input to a middle-section error prediction model, the middle-section error prediction model uses the predicted value of the middle-section error corresponding to the middle-section error training feature set as output, uses the middle-section error label of the test voltage transformer corresponding to the middle-section error training feature set as a prediction target, uses the difference between the predicted value of the middle-section error and the middle-section error label as a second prediction error, and uses minimizing the sum of squares of the second prediction errors as a training target; the middle-section error prediction model is trained until the sum of squares of the second prediction errors reaches convergence and the training is stopped;

[0034] For each model of capacitor:

[0035] Each group of next-section error training feature sets is used as input to a next-section error prediction model, the next-section error prediction model uses the predicted value of the next-section error corresponding to the next-section error training feature set as output, the next-section error label of the test voltage transformer corresponding to the next-section error training feature set as a prediction target, the difference between the predicted value of the next-section error and the next-section error label as a third prediction error, and minimizing the sum of squares of the third prediction errors as a training target; the next-section error prediction model is trained until the sum of squares of the third prediction errors reaches convergence and the training is stopped;

[0036] The method of collecting the upper section error characteristics, the middle section error characteristics and the lower section error characteristics corresponding to the voltage transformer to be tested is:

[0037] Before performing a preventive test on the voltage transformer to be tested, various characteristic elements of the upper section capacitance of the voltage transformer to be tested are collected to form an upper section error feature, various characteristic elements of the middle section capacitance of the voltage transformer to be tested are collected to form a middle section error feature, and various characteristic elements of the lower section capacitance of the voltage transformer to be tested are collected to form a lower section error feature;

[0038] The method of obtaining the corrected dielectric loss of the upper section based on the error characteristics of the upper section, the dielectric loss value tested in the upper section, and the error prediction model of the upper section; obtaining the corrected dielectric loss of the middle section based on the error characteristics of the middle section, the dielectric loss value tested in the middle section, and the error prediction model of the middle section; and obtaining the corrected dielectric loss of the lower section based on the error characteristics of the lower section, the dielectric loss value tested in the lower section, and the error prediction model of the lower section is as follows:

[0039] Input the error characteristics of the previous section into the error prediction model of the previous section to obtain the predicted value of the error of the previous section output by the error prediction model of the previous section; mark the predicted value of the error of the previous section as A1, and mark the dielectric loss value of the previous section as B1, then the calculation formula of the corrected dielectric loss X1 of the previous section is: ;

[0040] Input the middle section error characteristics into the middle section error prediction model to obtain the predicted value of the middle section error output by the middle section error prediction model; mark the predicted value of the middle section error as A2, and mark the middle section test dielectric loss value as B2, then the calculation formula of the middle section corrected dielectric loss X2 is: ;

[0041] Input the error characteristics of the next section into the error prediction model of the next section to obtain the predicted value of the next section error output by the error prediction model of the next section; mark the predicted value of the next section error as A3, and mark the dielectric loss value of the next section test as B3, then the calculation formula of the corrected dielectric loss X3 of the next section is: ;

[0042] The method for calculating the comprehensive dielectric loss based on the corrected dielectric loss in the upper section, the corrected dielectric loss in the middle section, and the corrected dielectric loss in the lower section is:

[0043] The comprehensive dielectric loss is marked as X, and the calculation formula of the comprehensive dielectric loss X is: , where C1, C2 and C3 are preset proportional coefficients.

[0044] The present invention also provides a preventive testing device for a capacitor voltage transformer, comprising a training data collection module, a model training module, and a preventive testing module; wherein the modules are electrically connected to each other;

[0045] A training data collection module is used to collect the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data for each dielectric loss meter in advance, and send the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data to the module;

[0046] Model training module, for each type of dielectric loss meter:

[0047] Based on the error training feature data and the error label data of the previous section, the error prediction model of the previous section is trained; based on the error training feature data and the error label data of the middle section, the error prediction model of the middle section is trained; based on the error training feature data and the error label data of the next section, the error prediction model of the next section is trained, and the error prediction model of the previous section, the error prediction model of the middle section and the error prediction model of the next section are sent to the preventive testing module;

[0048] A preventive testing module is used to collect the model of the dielectric loss meter used for testing the voltage transformer to be tested, and read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to the model, collect the upper section error characteristics, middle section error characteristics and lower section error characteristics corresponding to the voltage transformer to be tested, and use the high-voltage lead test method without removing the high-voltage lead to respectively obtain the upper section test dielectric loss value, middle section test dielectric loss value and lower section test dielectric loss value of the voltage transformer to be tested, based on the upper section error characteristics, upper section test dielectric loss value and upper section error prediction model, obtain the upper section corrected dielectric loss; based on the middle section error characteristics, middle section test dielectric loss value and middle section error prediction model, obtain the middle section corrected dielectric loss; based on the lower section error characteristics, lower section test dielectric loss value and lower section error prediction model, obtain the lower section corrected dielectric loss, and based on the upper section corrected dielectric loss, middle section corrected dielectric loss and lower section corrected dielectric loss, calculate the comprehensive dielectric loss.

[0049] The present invention further provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0050] The processor executes the above-mentioned method for preventive testing of a capacitor voltage transformer by calling the computer program stored in the memory.

[0051] The present invention also provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0052] When the computer program is run on a computer device, the computer device is enabled to execute the above-mentioned method for preventive testing of a capacitor voltage transformer.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The present invention collects upper section error training feature data and upper section error label data, middle section error training feature data and middle section error label data, and lower section error training feature data and lower section error label data for each dielectric loss meter in advance. For each dielectric loss meter, the present invention: trains an upper section error prediction model based on the upper section error training feature data and the upper section error label data; trains a middle section error prediction model based on the middle section error training feature data and the middle section error label data; trains a lower section error prediction model based on the lower section error training feature data and the lower section error label data, collects the dielectric loss meter model for testing the voltage transformer to be tested, reads the upper section error prediction model, the middle section error prediction model and the lower section error prediction model corresponding to the model, collects the upper section error feature, the middle section error feature and the lower section error feature corresponding to the voltage transformer to be tested, The high-voltage lead test method without removing the high-voltage lead is used to obtain the upper section test dielectric loss value, middle section test dielectric loss value and lower section test dielectric loss value of the voltage transformer to be tested respectively, and the upper section corrected dielectric loss is obtained based on the upper section error characteristics, the upper section test dielectric loss value and the upper section error prediction model; the middle section corrected dielectric loss is obtained based on the middle section error characteristics, the middle section test dielectric loss value and the middle section error prediction model; the lower section corrected dielectric loss is obtained based on the lower section error characteristics, the lower section test dielectric loss value and the lower section error prediction model, and the comprehensive dielectric loss is calculated based on the upper section corrected dielectric loss, the middle section corrected dielectric loss and the lower section corrected dielectric loss; by training the error prediction model for the test error prediction corresponding to the upper section, middle section and lower section for each type of dielectric loss meter, it is ensured that when a non-high-precision dielectric loss meter is used for preventive testing, the measurement effect of a high-precision dielectric loss meter can also be achieved, thereby reducing the procurement cost of the dielectric loss meter. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of a method for preventive testing of a capacitor voltage transformer in Example 1 of the present invention;

[0056] Figure 2 This is a test wiring diagram for testing the upper capacitance in Example 1 of the present invention;

[0057] Figure 3This is a test wiring diagram for testing the capacitance of the center node in Example 1 of the present invention;

[0058] Figure 4 This is a module connection diagram of a device for preventive testing of a capacitor voltage transformer in Example 2 of the present invention. DETAILED DESCRIPTION

[0059] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0060] Example 1

[0061] like Figure 1 As shown, a method for preventive testing of a capacitor voltage transformer comprises the following steps:

[0062] Step 1: Collect the upper section error training feature data and upper section error label data, the middle section error training feature data and middle section error label data, and the lower section error training feature data and lower section error label data for each dielectric loss meter in advance;

[0063] Step 2: For each type of dielectric loss meter:

[0064] Based on the error training feature data and the error label data of the previous section, the error prediction model of the previous section is trained; based on the error training feature data and the error label data of the middle section, the error prediction model of the middle section is trained; based on the error training feature data and the error label data of the next section, the error prediction model of the next section is trained;

[0065] Step 3: Collect the dielectric loss meter model of the voltage transformer to be tested, and read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to the model;

[0066] Step 4: Collect the upper section error characteristics, middle section error characteristics, and lower section error characteristics corresponding to the voltage transformer to be tested, and use the high-voltage lead test method without removing the high-voltage lead to obtain the upper section test dielectric loss value, middle section test dielectric loss value, and lower section test dielectric loss value of the voltage transformer to be tested respectively;

[0067] Step 5: Based on the error characteristics of the upper section, the dielectric loss value tested in the upper section, and the error prediction model of the upper section, obtain the corrected dielectric loss of the upper section; based on the error characteristics of the middle section, the dielectric loss value tested in the middle section, and the error prediction model of the middle section, obtain the corrected dielectric loss of the middle section; based on the error characteristics of the lower section, the dielectric loss value tested in the lower section, and the error prediction model of the lower section, obtain the corrected dielectric loss of the lower section;

[0068] Step 6: Calculate the comprehensive dielectric loss based on the corrected dielectric loss in the upper section, the middle section, and the lower section.

[0069] The method of collecting the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data for each dielectric loss meter is as follows:

[0070] Collect several groups of test voltage transformers for each type of dielectric loss meter, and collect the upper section error training feature set, middle section error training feature set, and lower section error training feature set corresponding to each group of test voltage transformers;

[0071] The characteristic elements in the upper section error training feature set include but are not limited to the temperature, humidity, stray capacitance, frequency mean, frequency variance, ground resistance, and aging degree of the upper section capacitor of the test voltage transformer; wherein the aging degree of the capacitor can be determined by measuring the dielectric loss tangent value, that is, by applying a standard voltage and measuring the charging current and admittance component of the transformer winding, the dielectric loss angle tanδ can be calculated; the larger the tanδ value, the more serious the dielectric aging, that is, tanδ can be used as an expression of the aging degree;

[0072] The feature elements in the mid-section error training feature set include but are not limited to the temperature, humidity, frequency mean, frequency variance, ground resistance and aging degree of the mid-section capacitor of the test voltage transformer;

[0073] The feature elements in the lower section error training feature set include but are not limited to the temperature, humidity, frequency mean, frequency variance, external electric field strength, grounding resistance and aging degree of the lower section capacitor of the test voltage transformer;

[0074] Use this type of dielectric loss meter to test the upper, middle, and lower capacitance of each set of test voltage transformers without removing the high-voltage leads to obtain the upper, middle, and lower capacitance values. The number of test voltage transformers collected by each type of dielectric loss meter is determined based on the actual number collected.

[0075] Then, a high-precision dielectric loss meter is used to measure the dielectric loss value of each test voltage transformer to obtain the upper section accurate dielectric loss value, the middle section accurate dielectric loss value, and the lower section accurate dielectric loss value. Specifically, the high-precision dielectric loss meter can be an imported M4000 or a domestic AI-6000 dielectric loss meter. Both instruments have a strong anti-interference function of different frequencies, which can effectively avoid interference from electric fields and other factors in the environment.

[0076] For each set of tested voltage transformers, the upper section test dielectric loss value is calculated and divided by the upper section precise dielectric loss value to obtain the upper section error label; the middle section test dielectric loss value is calculated and divided by the middle section precise dielectric loss value to obtain the middle section error label; the lower section test dielectric loss value is calculated and divided by the lower section precise dielectric loss value to obtain the lower section error label; it can be understood that by calculating the ratio between the dielectric loss value of each capacitance test and the precisely measured dielectric loss value, the measurement error of the dielectric loss meter of this model can be obtained;

[0077] For each dielectric loss meter, all of its upper section error training feature sets constitute the upper section error training feature data, and all of its upper section error labels constitute the upper section error label data; for each dielectric loss meter, all of its middle section error training feature sets constitute the middle section error training feature data, and all of its middle section error labels constitute the middle section error label data; for each dielectric loss meter, all of its lower section error training feature sets constitute the lower section error training feature data, and all of its lower section error labels constitute the lower section error label data;

[0078] It should be noted that the specific method of the non-removal high-voltage lead testing method is as follows:

[0079] First, for the capacitor on the capacitor voltage transformer, from a safety perspective, the grounding switch cannot be opened without removing the high-voltage lead, so the top can only be grounded, and then the reverse connection method can be used to measure the dielectric loss; Figure 2 The figure shows the test wiring diagram for the upper section capacitance test using M4000;

[0080] For the middle section, the conventional connection method is used for measurement, and the measured data is the same as the disconnection data; Figure 3 The following is the experimental wiring diagram for testing the middle section capacitance using M4000; Figure 2 and Figure 3 Where δ represents dielectric loss factor, X represents reactor, and PT represents voltage transformer;

[0081] When measuring the capacitance of the lower section, the self-excitation method is used for measurement, and it is necessary to pay special attention to the insulation condition of the point;

[0082] Furthermore, the method of training the upper section error prediction model based on the upper section error training feature data and the upper section error label data; training the middle section error prediction model based on the middle section error training feature data and the middle section error label data; and training the lower section error prediction model based on the lower section error training feature data and the lower section error label data is as follows:

[0083] For each type of dielectric loss meter, the upper capacitance is:

[0084] Each group of upper section error training feature sets is used as the input of the upper section error prediction model, the upper section error prediction model uses the predicted value of the upper section error corresponding to the upper section error training feature set as the output, the upper section error label of the test voltage transformer corresponding to the upper section error training feature set as the prediction target, the difference between the predicted value of the upper section error and the upper section error label as the first prediction error, and minimizing the sum of squares of the first prediction errors as the training target; the upper section error prediction model is trained until the sum of squares of the first prediction errors reaches convergence, and the training is stopped, and an upper section error prediction model is trained based on the upper section error training feature set to output the error that may be caused by using this type of dielectric loss meter; the upper section error prediction model is any one of the regression models, and the regression model includes but is not limited to a polynomial regression model, an SVR model, etc.; the sum of squares of the first prediction errors is the mean square error;

[0085] For each type of dielectric loss meter, the capacitance of the section is:

[0086] Each group of middle-section error training feature sets is used as the input of the middle-section error prediction model, the middle-section error prediction model uses the predicted value of the middle-section error corresponding to the middle-section error training feature set as the output, uses the middle-section error label of the test voltage transformer corresponding to the middle-section error training feature set as the prediction target, uses the difference between the predicted value of the middle-section error and the middle-section error label as the second prediction error, and uses minimizing the sum of squares of the second prediction errors as the training target; the middle-section error prediction model is trained until the sum of squares of the second prediction errors reaches convergence, and the training is stopped, and a middle-section error prediction model is trained based on the middle-section error training feature set to output the error that may be caused by using this type of dielectric loss meter; the middle-section error prediction model is any one of the regression models;

[0087] For each model of capacitor:

[0088] Each group of next-section error training feature sets is used as the input of the next-section error prediction model, the next-section error prediction model uses the predicted value of the next-section error corresponding to the next-section error training feature set as the output, the next-section error label of the test voltage transformer corresponding to the next-section error training feature set as the prediction target, the difference between the predicted value of the next-section error and the next-section error label as the third prediction error, and minimizing the sum of squares of the third prediction error as the training target; the next-section error prediction model is trained until the sum of squares of the third prediction error reaches convergence, and the training is stopped, and a next-section error prediction model is trained based on the next-section error training feature set to output the error that may be caused by using this type of dielectric loss meter; the next-section error prediction model is any one of the regression models;

[0089] Furthermore, the method of collecting the upper section error characteristics, the middle section error characteristics and the lower section error characteristics corresponding to the voltage transformer to be tested is:

[0090] Before the preventive test of the voltage transformer to be tested is performed, the characteristic elements of the upper section capacitance of the voltage transformer to be tested are collected to form the upper section error feature, the characteristic elements of the middle section capacitance of the voltage transformer to be tested are collected to form the middle section error feature, and the characteristic elements of the lower section capacitance of the voltage transformer to be tested are collected to form the lower section error feature; it can be understood that the specific characteristic elements are respectively consistent with the characteristic elements in the upper section error training feature set, the middle section error training feature set and the lower section error training feature set;

[0091] Furthermore, the method of obtaining the upper section corrected dielectric loss based on the upper section error characteristics, the upper section test dielectric loss value and the upper section error prediction model; obtaining the middle section corrected dielectric loss based on the middle section error characteristics, the middle section test dielectric loss value and the middle section error prediction model; and obtaining the lower section corrected dielectric loss based on the lower section error characteristics, the lower section test dielectric loss value and the lower section error prediction model is as follows:

[0092] Input the error characteristics of the previous section into the error prediction model of the previous section to obtain the predicted value of the error of the previous section output by the error prediction model of the previous section; mark the predicted value of the error of the previous section as A1, and mark the dielectric loss value of the previous section as B1, then the calculation formula of the corrected dielectric loss X1 of the previous section is: ;

[0093] Input the middle section error characteristics into the middle section error prediction model to obtain the predicted value of the middle section error output by the middle section error prediction model; mark the predicted value of the middle section error as A2, and mark the middle section test dielectric loss value as B2, then the calculation formula of the middle section corrected dielectric loss X2 is: ;

[0094] Input the error characteristics of the next section into the error prediction model of the next section to obtain the predicted value of the next section error output by the error prediction model of the next section; mark the predicted value of the next section error as A3, and mark the dielectric loss value of the next section test as B3, then the calculation formula of the corrected dielectric loss X3 of the next section is: ;

[0095] Furthermore, the method for calculating the comprehensive dielectric loss based on the upper section corrected dielectric loss, the middle section corrected dielectric loss and the lower section corrected dielectric loss is:

[0096] The comprehensive dielectric loss is marked as X, and the calculation formula of the comprehensive dielectric loss X is: ; Among them, C1, C2 and C3 are preset proportional coefficients;

[0097] It is understandable that by analyzing the error coefficient of each type of dielectric loss meter, it can be ensured that when a non-high-precision dielectric loss meter is used for preventive testing, the measurement effect of a high-precision dielectric loss meter can be achieved, thereby reducing the procurement cost of the dielectric loss meter.

[0098] Example 2

[0099] like Figure 4 As shown, a preventive testing device for a capacitor voltage transformer includes a training data collection module, a model training module, and a preventive testing module; wherein each module is electrically connected;

[0100] Among them, the training data collection module is mainly used to collect the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data for each dielectric loss meter in advance, and send the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data to the module;

[0101] The model training module is mainly used for each dielectric loss meter:

[0102] Based on the error training feature data and the error label data of the previous section, the error prediction model of the previous section is trained; based on the error training feature data and the error label data of the middle section, the error prediction model of the middle section is trained; based on the error training feature data and the error label data of the next section, the error prediction model of the next section is trained, and the error prediction model of the previous section, the error prediction model of the middle section and the error prediction model of the next section are sent to the preventive testing module;

[0103] Among them, the preventive test module is mainly used to collect the model of the dielectric loss meter for testing the voltage transformer to be tested, and read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to the model, collect the upper section error characteristics, middle section error characteristics and lower section error characteristics corresponding to the voltage transformer to be tested, and use the high-voltage lead test method without removing the high-voltage lead to respectively obtain the upper section test dielectric loss value, middle section test dielectric loss value and lower section test dielectric loss value of the voltage transformer to be tested, based on the upper section error characteristics, upper section test dielectric loss value and upper section error prediction model, obtain the upper section corrected dielectric loss; based on the middle section error characteristics, middle section test dielectric loss value and middle section error prediction model, obtain the middle section corrected dielectric loss; based on the lower section error characteristics, lower section test dielectric loss value and lower section error prediction model, obtain the lower section corrected dielectric loss, and based on the upper section corrected dielectric loss, middle section corrected dielectric loss and lower section corrected dielectric loss, calculate the comprehensive dielectric loss.

[0104] Example 3

[0105] According to another aspect of the present application, an electronic device is provided. The electronic device may include one or more processors and one or more memories. The memories may store computer-readable code that, when executed by the one or more processors, may implement the method for preventive testing of a capacitor voltage transformer as described above.

[0106] The methods or apparatus according to the embodiments of the present application can also be implemented using the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output components, a hard disk, and the like. A storage device in the electronic device, such as a ROM or hard disk, can store an implementation of the method for preventive testing of a capacitor voltage transformer provided in this application.

[0107] Furthermore, the electronic device may further include a user interface. Of course, when implementing different devices, one or more components in the electronic device may be omitted according to actual needs.

[0108] Example 4

[0109] According to one embodiment of the present application, a computer-readable storage medium is also provided. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by a processor, the method for preventive testing of a capacitor voltage transformer according to an embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc.

[0110] In addition, according to embodiments of the present application, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions capable of being executed by a processor to execute instructions corresponding to the steps of the method provided in the present application. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the method of the present application are performed.

[0111] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present application. Therefore, the present application also covers recording media that store programs for executing the methods according to the present application.

[0112] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0113] The above-described specific embodiments further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is merely a specific 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 shall be included within the scope of protection of the present invention.

[0114] The above preset parameters or preset thresholds are all set by those skilled in the art according to actual conditions or obtained through large amounts of data simulation.

[0115] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for preventive testing of a capacitor voltage transformer, characterized in that: The following steps are involved: Step 1: Collect the upper section error training feature data and upper section error label data, the middle section error training feature data and middle section error label data, and the lower section error training feature data and lower section error label data for each dielectric loss meter in advance; Step 2: For each dielectric loss meter: Based on the upper section error training feature data and the upper section error label data, train the upper section error prediction model; Based on the middle section error training feature data and the middle section error label data, train the middle section error prediction model; Based on the lower section error training feature data and the lower section error label data, train the lower section error prediction model; Step 3: Collect the dielectric loss meter model of the voltage transformer to be tested, and read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to the model; Step 4: Collect the upper section error characteristics, middle section error characteristics, and lower section error characteristics corresponding to the voltage transformer to be tested, and use the high-voltage lead test method without removing the high-voltage lead to obtain the upper section test dielectric loss value, middle section test dielectric loss value, and lower section test dielectric loss value of the voltage transformer to be tested respectively; Step 5: Based on the error characteristics of the upper section, the dielectric loss value tested in the upper section, and the error prediction model of the upper section, obtain the corrected dielectric loss of the upper section; based on the error characteristics of the middle section, the dielectric loss value tested in the middle section, and the error prediction model of the middle section, obtain the corrected dielectric loss of the middle section; based on the error characteristics of the lower section, the dielectric loss value tested in the lower section, and the error prediction model of the lower section, obtain the corrected dielectric loss of the lower section; Step 6: Calculate the comprehensive dielectric loss based on the corrected dielectric loss in the previous section, the corrected dielectric loss in the middle section, and the corrected dielectric loss in the next section.

2. The method for preventive testing of a capacitor voltage transformer according to claim 1, characterized in that: The method of collecting the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data for each dielectric loss meter is as follows: Collect several groups of test voltage transformers for each type of dielectric loss meter, and collect the upper section error training feature set, middle section error training feature set, and lower section error training feature set corresponding to each group of test voltage transformers; Use this type of dielectric loss meter to test the upper, middle, and lower capacitance of each set of test voltage transformers without removing the high-voltage leads to obtain the upper, middle, and lower capacitance values. The number of test voltage transformers collected by each type of dielectric loss meter is determined based on the actual number collected. Then, a high-precision dielectric loss meter is used to measure the dielectric loss value of each test voltage transformer to obtain the upper section accurate dielectric loss value, the middle section accurate dielectric loss value and the lower section accurate dielectric loss value respectively; For each set of tested voltage transformers, calculate the upper section test dielectric loss value divided by the upper section accurate dielectric loss value to obtain the upper section error label; calculate the middle section test dielectric loss value divided by the middle section accurate dielectric loss value to obtain the middle section error label; calculate the lower section test dielectric loss value divided by the lower section accurate dielectric loss value to obtain the lower section error label; For each dielectric loss meter, all of its upper-section error training feature sets constitute the upper-section error training feature data, and all of its upper-section error labels constitute the upper-section error label data; for each dielectric loss meter, all of its middle-section error training feature sets constitute the middle-section error training feature data, and all of its middle-section error labels constitute the middle-section error label data; for each dielectric loss meter, all of its lower-section error training feature sets constitute the lower-section error training feature data, and all of its lower-section error labels constitute the lower-section error label data.

3. The method for preventive testing of a capacitor voltage transformer according to claim 2, characterized in that: The feature elements in the upper section error training feature set include the temperature, humidity, stray capacitance, frequency mean, frequency variance, ground resistance and aging degree of the upper section capacitor of the test voltage transformer; The feature elements in the mid-section error training feature set include the temperature, humidity, frequency mean, frequency variance, ground resistance and aging degree of the upper section capacitor of the test voltage transformer; The feature elements in the lower section error training feature set include the temperature, humidity, frequency mean, frequency variance value, external electric field strength, grounding resistance and aging degree of the lower section capacitor of the test voltage transformer.

4. The method for preventive testing of a capacitor voltage transformer according to claim 3, wherein: The method of testing the high voltage lead without removing it is as follows: For the upper capacitor of the capacitive voltage transformer, ground its top end and then measure the dielectric loss by using the reverse connection method. For the middle section, the conventional connection method is adopted for measurement, and the measured data is the same as the disconnection data; When measuring the capacitance of the lower section, the self-excitation method is used.

5. The method for preventive testing of a capacitor voltage transformer according to claim 4, characterized in that: The method of training the upper section error prediction model based on the upper section error training feature data and the upper section error label data; training the middle section error prediction model based on the middle section error training feature data and the middle section error label data; and training the lower section error prediction model based on the lower section error training feature data and the lower section error label data is as follows: For each type of dielectric loss meter, the upper capacitance is: Each group of upper section error training feature sets is used as input of an upper section error prediction model, the upper section error prediction model uses the predicted value of the upper section error corresponding to the upper section error training feature set as output, the upper section error label of the test voltage transformer corresponding to the upper section error training feature set is used as a prediction target, the difference between the predicted value of the upper section error and the upper section error label is used as a first prediction error, and minimizing the sum of squares of the first prediction errors is used as a training target; the upper section error prediction model is trained until the sum of squares of the first prediction errors reaches convergence and the training is stopped; For each type of dielectric loss meter, the capacitance of the section is: Each group of middle-section error training feature sets is used as input to a middle-section error prediction model, the middle-section error prediction model uses the predicted value of the middle-section error corresponding to the middle-section error training feature set as output, uses the middle-section error label of the test voltage transformer corresponding to the middle-section error training feature set as a prediction target, uses the difference between the predicted value of the middle-section error and the middle-section error label as a second prediction error, and uses minimizing the sum of squares of the second prediction errors as a training target; the middle-section error prediction model is trained until the sum of squares of the second prediction errors reaches convergence and the training is stopped; For each type of dielectric loss meter lower section capacitance: Each group of next-section error training feature sets is used as the input of the next-section error prediction model, the next-section error prediction model uses the predicted value of the next-section error corresponding to the next-section error training feature set as the output, the next-section error label of the test voltage transformer corresponding to the next-section error training feature set as the prediction target, the difference between the predicted value of the next-section error and the next-section error label as the third prediction error, and minimizing the sum of squares of the third prediction error as the training target; the next-section error prediction model is trained until the sum of squares of the third prediction error reaches convergence and the training is stopped.

6. The method for preventive testing of a capacitor voltage transformer according to claim 5, characterized in that: The method of collecting the upper section error characteristics, the middle section error characteristics and the lower section error characteristics corresponding to the voltage transformer to be tested is: Before conducting a preventive test on the voltage transformer to be tested, the characteristic elements of the upper section capacitance of the voltage transformer to be tested are collected to form an upper section error feature, the characteristic elements of the middle section capacitance of the voltage transformer to be tested are collected to form a middle section error feature, and the characteristic elements of the lower section capacitance of the voltage transformer to be tested are collected to form a lower section error feature.

7. The method for preventive testing of a capacitor voltage transformer according to claim 6, characterized in that: The method of obtaining the corrected dielectric loss of the upper section based on the error characteristics of the upper section, the dielectric loss value tested in the upper section, and the error prediction model of the upper section; obtaining the corrected dielectric loss of the middle section based on the error characteristics of the middle section, the dielectric loss value tested in the middle section, and the error prediction model of the middle section; and obtaining the corrected dielectric loss of the lower section based on the error characteristics of the lower section, the dielectric loss value tested in the lower section, and the error prediction model of the lower section is as follows: Input the error characteristics of the previous section into the error prediction model of the previous section to obtain the predicted value of the error of the previous section output by the error prediction model of the previous section; mark the predicted value of the error of the previous section as A1, and mark the dielectric loss value of the previous section as B1, then the calculation formula of the corrected dielectric loss X1 of the previous section is: ; Input the middle section error characteristics into the middle section error prediction model to obtain the predicted value of the middle section error output by the middle section error prediction model; mark the predicted value of the middle section error as A2, and mark the middle section test dielectric loss value as B2, then the calculation formula of the middle section corrected dielectric loss X2 is: ; Input the error characteristics of the next section into the error prediction model of the next section to obtain the predicted value of the next section error output by the error prediction model of the next section; mark the predicted value of the next section error as A3, and mark the dielectric loss value of the next section test as B3, then the calculation formula of the corrected dielectric loss X3 of the next section is: .

8. The method for preventive testing of a capacitor voltage transformer according to claim 7, characterized in that: The method for calculating the comprehensive dielectric loss based on the corrected dielectric loss in the upper section, the corrected dielectric loss in the middle section, and the corrected dielectric loss in the lower section is: The comprehensive dielectric loss is marked as X, and the calculation formula of the comprehensive dielectric loss X is: , where C1, C2 and C3 are preset proportional coefficients.

9. A device for preventive testing of a capacitor voltage transformer, used to implement a method for preventive testing of a capacitor voltage transformer as claimed in any one of claims 1 to 8, characterized in that: It includes a training data collection module, a model training module and a preventive testing module; wherein each module is electrically connected to each other; A training data collection module is used to collect the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data for each dielectric loss meter in advance, and send the upper section error training feature data and the upper section error label data, the middle section error training feature data and the middle section error label data, and the lower section error training feature data and the lower section error label data to the module; Model training module, for each type of dielectric loss meter: Based on the error training feature data and the error label data of the previous section, the error prediction model of the previous section is trained; based on the error training feature data and the error label data of the middle section, the error prediction model of the middle section is trained; based on the error training feature data and the error label data of the next section, the error prediction model of the next section is trained, and the error prediction model of the previous section, the error prediction model of the middle section and the error prediction model of the next section are sent to the preventive testing module; A preventive testing module is used to collect the model of the dielectric loss meter used for testing the voltage transformer to be tested, and read the upper section error prediction model, middle section error prediction model and lower section error prediction model corresponding to the model, collect the upper section error characteristics, middle section error characteristics and lower section error characteristics corresponding to the voltage transformer to be tested, and use the high-voltage lead test method without removing the high-voltage lead to respectively obtain the upper section test dielectric loss value, middle section test dielectric loss value and lower section test dielectric loss value of the voltage transformer to be tested, based on the upper section error characteristics, upper section test dielectric loss value and upper section error prediction model, obtain the upper section corrected dielectric loss; based on the middle section error characteristics, middle section test dielectric loss value and middle section error prediction model, obtain the middle section corrected dielectric loss; based on the lower section error characteristics, lower section test dielectric loss value and lower section error prediction model, obtain the lower section corrected dielectric loss, and based on the upper section corrected dielectric loss, middle section corrected dielectric loss and lower section corrected dielectric loss, calculate the comprehensive dielectric loss.

10. An electronic device, characterized in that: include: processor and memory, wherein The memory stores a computer program that can be called by the processor; The processor executes the method for preventive testing of a capacitor voltage transformer according to any one of claims 1 to 8 in the background by calling the computer program stored in the memory.

11. A computer-readable storage medium, characterized in that A rewritable computer program is stored thereon; When the computer program is run on a computer device, the computer device is enabled to execute the method for preventive testing of a capacitor voltage transformer according to any one of claims 1 to 8 in the background.

Images