Voltage transformer measurement method based on multi-source data
By constructing a voltage transformer correction model based on multi-source data, the problem of insufficient measurement accuracy of traditional voltage transformers in high voltage and load fluctuations is solved, and higher measurement accuracy and adaptability are achieved.
Patent Information
- Application Number
- CN202510541135.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-27
- Publication Date
- 2025-08-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional voltage transformers have limited measurement accuracy in environments with high voltage or high load fluctuations, and are sensitive to ambient temperature and have slow response speed, so they cannot quickly adapt to grid voltage fluctuations.
The first correction model and the second correction model of the voltage transformer are constructed based on multi-source data. Combined with the historical data of the voltage transformer measurement points and load, the voltage correction evaluation value P1 and the environmental correction value C2 are generated, and the current measurement data is corrected through the correction model.
It improves the accuracy and adaptability of voltage transformer measurement, reduces measurement errors, and can effectively correct under different load conditions and environmental conditions.
Smart Images
Figure CN120446852A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of voltage transformer measurement, and in particular to a voltage transformer measurement method based on multi-source data. Background Art
[0002] Voltage transformers are widely used in power systems, mainly for voltage measurement and protection of high-voltage power lines. Their output signals can be used for energy metering, equipment monitoring and fault diagnosis.
[0003] However, traditional voltage transformers have several shortcomings, including: measurement accuracy is limited by their nonlinear characteristics, especially in environments with high voltage or large load fluctuations, where large amplitude and phase errors often occur; traditional voltage transformers are sensitive to ambient temperature and are prone to performance degradation after long-term operation, and have a slow response speed and are unable to quickly adapt to voltage fluctuations in the power grid. Summary of the Invention
[0004] The purpose of the present invention is to more comprehensively consider various factors affecting measurement based on a multi-source data processing method, thereby effectively correcting measurement data, reducing measurement errors, and improving measurement accuracy.
[0005] In order to achieve the above object, the present invention provides a voltage transformer measurement method based on multi-source data, comprising: Generate measurement history data and corresponding environmental history data based on the acquired historical operation data of the voltage transformer; constructing a first correction model of the voltage transformer based on the measurement history data; constructing a second correction model of the voltage transformer based on environmental historical data; The obtained measurement data of the voltage transformer at the current time node is corrected in combination with the first correction model and the second correction model to generate corrected voltage measurement data.
[0006] In some embodiments of the present invention, the constructing of the first modified model of the voltage transformer further includes: Obtain historical voltage data of a voltage transformer measurement point and historical voltage data of a load to generate historical measurement data; Generate a measurement point feature set V1 based on historical voltage data of the voltage transformer measurement point; Generate a load feature set V2 based on the historical voltage data of the load; Combine the measurement point feature set V1 and the load feature set V2 to generate a voltage correction evaluation value P1; Performing equal gradient division on the voltage correction evaluation value P1 to generate a first-level gradient data set of the voltage correction evaluation value P1, and obtaining a first deviation value B between the actual value and the measured value of the voltage transformer measurement point corresponding to each first-level gradient data; Based on the first deviation value B, the voltage correction evaluation value P1 is gradient-divided again to generate a secondary gradient data set of the voltage correction evaluation value P1; The first correction value C1 is generated by combining the secondary gradient data set of the voltage correction evaluation value P1 and the corresponding deviation value.
[0007] In some embodiments of the present invention, the generating of historical measurement data further includes: Obtain corresponding historical voltage data from the voltage transformer monitoring system and the load monitoring device respectively; Obtain historical voltage data of the voltage transformer measurement point and historical voltage data of the load at the same test time node; Combine the historical voltage data of the voltage transformer measurement point and the historical voltage data of the load to generate a sample data set D of historical measurement data, D = {D1, D2…Di…Dn}; Where Di=(U pi ,U li ), Di represents the sample point data at the i-th value-taking moment, n represents the total number of values taken at the current test time node, U pi Represents the historical voltage data reference value of the voltage transformer measurement point at the i-th value taking moment, U li Indicates the historical voltage data reference value of the load at the i-th value-taking moment.
[0008] In some embodiments of the present invention, the generating of the voltage correction evaluation value P1 further includes: Obtaining a historical voltage data reference value of a voltage transformer measurement point at the current test time node and a historical voltage data reference value of a load at the current test time node based on a sample data set D of historical measurement data at the current test time node; Based on the historical voltage data reference value of the voltage transformer measurement point, the measurement point feature set V1 is obtained, V1={v 11 ,v 12 …v 1j …v 1m}; The load characteristic set V2 is obtained based on the historical voltage data reference value of the load, V2={v 21 ,v 22 …v 2j …v 2m}; Among them, v 1j represents the jth type of eigenvalue in the measurement point feature set V1, 1m represents the total number of eigenvalue types in the measurement point feature set V1, and v 2j represents the j-th type of eigenvalue in the load feature set V2, and 2m represents the total number of eigenvalue types in the load feature set V2; Combine the measurement point feature set V1 and the load feature set V2 to generate a voltage correction evaluation value P1;
[0009] in, is the weight of the j-th type of eigenvalue in the measurement point feature set V1, is the weight of the j-th type of eigenvalue in the load feature set V2, is the weight of the evaluation value of the measurement point feature set, The weight of the load feature set evaluation value.
[0010] In some embodiments of the present invention, the generating of the secondary gradient data set of the voltage correction evaluation value P1 further includes: Obtaining a value range of the voltage correction evaluation value P1 based on the historical voltage correction evaluation value P1; Divide the value range of the voltage correction evaluation value P1 into intervals to generate k gradient intervals to generate a first-level gradient data set of the voltage correction evaluation value P1; For each first-level gradient data interval, obtain the actual value of the voltage transformer measurement point corresponding to the xth voltage correction evaluation value P1 in the current first-level gradient data interval With measured value ; Calculate a first deviation value B between the actual value and the measured value of the voltage transformer measurement point corresponding to each first-level gradient data;
[0011] Wherein, xn represents the total number of voltage correction evaluation values P1 in the current first-level gradient data interval.
[0012] In some embodiments of the present invention, the generating of the first correction value C1 further includes: Based on the first deviation value B, the voltage correction evaluation value P1 is gradient-divided again, and a comparison preset value Ba of the first deviation value B is set; If B>Ba, the corresponding first-level gradient data interval is removed from the first-level gradient data set, and the current first-level gradient data interval is divided into k intervals again to generate a temporary gradient data set; Combining the first-level gradient dataset and the temporary gradient dataset to generate a second-level gradient dataset; The first correction value C1 is generated by combining the secondary gradient data set of the voltage correction evaluation value P1 and the corresponding deviation value.
[0013] In some embodiments of the present invention, when constructing the second correction model of the voltage transformer based on the environmental historical data, the method further includes: Construct an environmental assessment value model based on historical environmental data; Combining historical environmental data and an environmental evaluation value model to generate a first environmental evaluation value H; Generate an environmental feature set based on a change trend of a first environmental evaluation value H in historical environmental data; The second deviation value corresponding to the environmental feature sample in each environmental feature set is calculated in combination with historical data to generate a second correction value C2.
[0014] In some embodiments of the present invention, the generating of the first environmental evaluation value H further includes:
[0015] Among them, h y Indicates the reference value of environmental data of type y, a y represents the weight of the reference value of the yth type of environmental data, represents the correction value of the first environmental evaluation value H, Indicates the total number of environment data types.
[0016] In some embodiments of the present invention, the generating of the second correction value C2 further includes: Obtain a curve of the change of the first environmental evaluation value H over time; Generate an environmental feature set based on the change curve, the environmental feature set includes multiple types of environmental feature samples; Obtain a second deviation value R between an actual value and a measured value of a voltage transformer measurement point corresponding to each environmental characteristic sample; The environment preset value is set based on the second deviation value R corresponding to all environmental feature samples; The second deviation is clustered by comparing the second deviation value R with the environmental preset value to generate a second correction value C2.
[0017] In some embodiments of the present invention, the generating of the corrected voltage measurement data includes: Obtain the measured value L1 of the voltage transformer, the load voltage value, and environmental data at the current monitoring time node; Generate a first correction value c1 and a second correction value c2 by combining the measured value L1 of the voltage transformer, the load voltage value and the environmental data; Combining the voltage transformer's measurement value L1 and environmental data to generate a first correction value c1 and a second correction value c2 to generate a corrected voltage measurement L; L=L1+c1+c2.
[0018] Compared with the prior art, the voltage transformer measurement method based on multi-source data provided by the embodiment of the present invention has the following advantages: The first correction model is constructed based on the historical measurement data, taking into account multiple factors such as the historical voltage data of the voltage transformer measurement point and the load. The processing method based on multi-source data can more comprehensively consider the various factors affecting the measurement, thereby effectively correcting the measurement data, reducing measurement errors, and improving measurement accuracy.
[0019] The second correction model is constructed based on the environmental historical data. Taking into account the impact of environmental factors on the measurement, the measurement deviation caused by environmental factors can be corrected, thereby further improving the accuracy of the measurement.
[0020] By analyzing and processing these deviation values, the deviation in the measurement can be accurately corrected, so that the final corrected voltage measurement value is closer to the true value.
[0021] The second correction model based on the construction of environmental historical data fully considers the impact of environmental factors on voltage transformer measurement; it can adapt the measurement method to different environmental conditions and effectively perform measurement corrections regardless of high or low temperature, humid or dry environment.
[0022] The measurement is corrected according to the change of the load, so that the measurement method can adapt to the voltage transformer measurement under different load conditions, thereby improving the versatility and reliability of the measurement method.
[0023] By constructing a sample data set, the acquired historical data can be systematically organized and utilized, making data processing more orderly and efficient in the subsequent model construction and calculation process, which helps to improve the efficiency and accuracy of the entire measurement method. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 This is a flow chart of a voltage transformer measurement method based on multi-source data provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0025] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.
[0026] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.
[0027] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.
[0028] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.
[0029] Example 1: The embodiment of the present invention provides a voltage transformer measurement method based on multi-source data, such as Figure 1 As shown, including: Generate measurement history data and corresponding environmental history data based on the acquired historical operation data of the voltage transformer; constructing a first correction model of the voltage transformer based on the measurement history data; constructing a second correction model of the voltage transformer based on environmental historical data; The obtained measurement data of the voltage transformer at the current time node is corrected in combination with the first correction model and the second correction model to generate corrected voltage measurement data.
[0030] Example 2: The step of constructing the first modified model of the voltage transformer further includes: Obtain historical voltage data of a voltage transformer measurement point and historical voltage data of a load to generate historical measurement data; Generate a measurement point feature set V1 based on historical voltage data of the voltage transformer measurement point; Generate a load feature set V2 based on the historical voltage data of the load; Combine the measurement point feature set V1 and the load feature set V2 to generate a voltage correction evaluation value P1; Performing equal gradient division on the voltage correction evaluation value P1 to generate a first-level gradient data set of the voltage correction evaluation value P1, and obtaining a first deviation value B between the actual value and the measured value of the voltage transformer measurement point corresponding to each first-level gradient data; Based on the first deviation value B, the voltage correction evaluation value P1 is gradient-divided again to generate a secondary gradient data set of the voltage correction evaluation value P1; The first correction value C1 is generated by combining the secondary gradient data set of the voltage correction evaluation value P1 and the corresponding deviation value.
[0031] In this embodiment, to construct the first correction model, it is necessary to first obtain historical voltage data from the voltage transformer's measurement point and the load's voltage data. The voltage transformer's historical voltage data reflects the transformer's voltage conditions at different times in the past, while the load's historical voltage data reflects the voltage changes of the load connected to the transformer. These two types of data are crucial for fully understanding the voltage transformer's operating status and performance changes.
[0032] These two data sources are different, obtained from voltage transformer measurement points and load monitoring equipment respectively. Combining them together generates historical measurement data, which is the basis for building the correction model.
[0033] Example 3: The generating of historical measurement data further includes: Obtain corresponding historical voltage data from the voltage transformer monitoring system and the load monitoring device respectively; Obtain historical voltage data of the voltage transformer measurement point and historical voltage data of the load at the same test time node; Combine the historical voltage data of the voltage transformer measurement point and the historical voltage data of the load to generate a sample data set D of historical measurement data, D = {D1, D2…Di…Dn}; Where Di=(U pi ,U li ), Di represents the sample point data at the i-th value-taking moment, n represents the total number of values taken at the current test time node, U pi Represents the historical voltage data reference value of the voltage transformer measurement point at the i-th value taking moment, U li Indicates the historical voltage data reference value of the load at the i-th value-taking moment.
[0034] Example 4: When generating the voltage correction evaluation value P1, the method further includes: Obtaining a historical voltage data reference value of a voltage transformer measurement point at the current test time node and a historical voltage data reference value of a load at the current test time node based on a sample data set D of historical measurement data at the current test time node; Based on the historical voltage data reference value of the voltage transformer measurement point, the measurement point feature set V1 is obtained, V1={v 11 ,v 12 …v 1j …v 1m}; The load characteristic set V2 is obtained based on the historical voltage data reference value of the load, V2={v 21 ,v 22 …v 2j …v 2m}; Among them, v 1j represents the jth type of eigenvalue in the measurement point feature set V1, 1m represents the total number of eigenvalue types in the measurement point feature set V1, and v 2j represents the j-th type of eigenvalue in the load feature set V2, and 2m represents the total number of eigenvalue types in the load feature set V2; Combine the measurement point feature set V1 and the load feature set V2 to generate a voltage correction evaluation value P1;
[0035] in, is the weight of the j-th type of eigenvalue in the measurement point feature set V1, is the weight of the j-th type of eigenvalue in the load feature set V2, is the weight of the evaluation value of the measurement point feature set, The weight of the load feature set evaluation value.
[0036] In this embodiment, the element v in the feature set V1 1j Represents the jth type of eigenvalue in the measurement point feature set V1. For example, these eigenvalues may include the maximum, minimum, average, and fluctuation frequency of the voltage at the measurement point. By analyzing historical voltage data and extracting these eigenvalues, the characteristics of the voltage transformer measurement point can be described from multiple perspectives.
[0037] Similarly, the load feature set V2 is generated based on the historical voltage data of the load. The element v 2j Represents the jth type of eigenvalue in the load characteristic set V2. These eigenvalues may include load-related characteristics such as the magnitude of the load voltage change and the stability of the load voltage. The load characteristic set V2 can reflect the impact of the load on the operating state of the voltage transformer.
[0038] Combine the measurement point feature set V1 and the load feature set V2 to generate the voltage correction evaluation value P1. In this process, the weights of different types of feature values are involved. For example, the j-th type of feature value in the measurement point feature set V1 has its corresponding weight ω 1j , the j-th type of characteristic value in the load characteristic set V2 has its corresponding weight, as well as the weight of the evaluation value of the measurement point characteristic set and the weight of the evaluation value of the load characteristic set.
[0039] By setting these weights, the importance of the measurement point characteristics and load characteristics in generating the voltage correction evaluation value P1 can be adjusted according to actual conditions. For example, if the voltage fluctuation frequency at the measurement point has a greater impact on the voltage transformer measurement, the corresponding weight will be set relatively large.
[0040] Based on the set weights, the characteristic values in the measurement point characteristic set V1 and the load characteristic set V2 are comprehensively calculated to obtain the voltage correction evaluation value P1. This voltage correction evaluation value P1 can comprehensively reflect the combined impact of the measurement point and load on the voltage transformer measurement.
[0041] Example 5: The generating of the secondary gradient data set of the voltage correction evaluation value P1 further includes: Obtaining a value range of the voltage correction evaluation value P1 based on the historical voltage correction evaluation value P1; Divide the value range of the voltage correction evaluation value P1 into intervals to generate k gradient intervals to generate a first-level gradient data set of the voltage correction evaluation value P1; For each first-level gradient data interval, obtain the actual value of the voltage transformer measurement point corresponding to the xth voltage correction evaluation value P1 in the current first-level gradient data interval With measured value ; Calculate a first deviation value B between the actual value and the measured value of the voltage transformer measurement point corresponding to each first-level gradient data;
[0042] Wherein, xn represents the total number of voltage correction evaluation values P1 in the current first-level gradient data interval.
[0043] In this embodiment, the value range of the corrected historical voltage evaluation value P1 is first determined. This corrected historical voltage evaluation value P1 is calculated by integrating the measurement point feature set V1 and the load feature set V2 during the construction of the first correction model. This historical data reflects the comprehensive evaluation of the voltage transformer under different operating conditions.
[0044] For example, if the voltage correction evaluation value P1 has been taken between [a, b] in multiple measurements and analyses in the past, this interval [a, b] is the value range to be processed currently.
[0045] The value range [a, b] of the voltage correction evaluation value P1 is partitioned into k gradient intervals, thus obtaining the first-level gradient dataset of the voltage correction evaluation value P1. The length of each gradient interval can be calculated as (b − a) / k (assuming uniform partitioning).
[0046] By calculating the first deviation value B within each primary gradient data interval, we can analyze the actual measurement deviation of the measurement point within different voltage correction evaluation value P1 intervals. These first deviation values B serve as an important basis for further dividing the secondary gradient data set, allowing for more accurate correction of voltage transformer measurements.
[0047] Example 6: The generating of the first correction value C1 further includes: Based on the first deviation value B, the voltage correction evaluation value P1 is gradient-divided again, and a comparison preset value Ba of the first deviation value B is set; If B>Ba, the corresponding first-level gradient data interval is removed from the first-level gradient data set, and the current first-level gradient data interval is divided into k intervals again to generate a temporary gradient data set; Combining the first-level gradient dataset and the temporary gradient dataset to generate a second-level gradient dataset; The first correction value C1 is generated by combining the secondary gradient data set of the voltage correction evaluation value P1 and the corresponding deviation value.
[0048] In this embodiment, the voltage correction evaluation value P1 is further gradient-partitioned based on the first deviation value B. The voltage correction evaluation value P1 has previously been gradient-partitioned to obtain a first-level gradient data set, and the deviation value B corresponding to each first-level gradient data set has been calculated. This further partitioning is performed because analysis of the deviation value B reveals that the first-level gradient data set alone may not accurately reflect the actual voltage transformer measurement situation.
[0049] For example, in some first-level gradient data intervals, the deviation value B may be large, indicating that there are large fluctuations in the measurement accuracy in this interval, and further subdivision is needed for more accurate analysis and correction.
[0050] A comparison preset value Ba is set for the first deviation value B. This preset value Ba serves as a criterion for determining which first-level gradient data intervals require special processing. Its setting may be based on the voltage transformer measurement accuracy requirements, previous empirical data, or an understanding of the overall measurement system characteristics.
[0051] For example, if the measurement accuracy requirement is high, Ba may be set to be relatively small, so that the interval with a large deviation value B will be processed more strictly.
[0052] The first-level gradient dataset and the temporary gradient dataset are combined to generate the second-level gradient dataset. The first-level gradient dataset contains most of the interval data with deviation values within the acceptable range, while the temporary gradient dataset is the data after the interval with larger deviation values is divided again.
[0053] By combining the two, the resulting secondary gradient dataset can more comprehensively and accurately reflect the relationship between the voltage correction evaluation value P1 and the deviation value B. It not only includes interval data with good overall conditions, but also provides more detailed processing for intervals with problems.
[0054] The first correction value C1 is generated by combining the secondary gradient dataset of the voltage correction evaluation value P1 and the corresponding deviation value. Based on the secondary gradient dataset, the deviation value corresponding to each interval is considered. The deviation values are weighted averaged based on factors such as the width of each interval, the size of the deviation value within the interval, and its importance within the entire secondary gradient dataset. The result is the first correction value C1. This first correction value C1 is used to correct the voltage transformer measurement data to improve measurement accuracy.
[0055] Example 7: When the second correction model of the voltage transformer is constructed based on the environmental historical data, the method further includes: Construct an environmental assessment value model based on historical environmental data; Combining historical environmental data and an environmental evaluation value model to generate a first environmental evaluation value H; Generate an environmental feature set based on a change trend of a first environmental evaluation value H in historical environmental data; The second deviation value corresponding to the environmental feature sample in each environmental feature set is calculated in combination with historical data to generate a second correction value C2.
[0056] In this embodiment, an environmental evaluation value model is constructed based on historical environmental data. The historical environmental data includes various information related to the voltage transformer's operating environment, such as temperature, humidity, air pressure, and other factors. This data reflects the voltage transformer's operating background under different environmental conditions.
[0057] The purpose of building an environmental assessment value model is to comprehensively evaluate the impact of environmental factors on voltage transformer measurements. This model can be based on a mathematical formula or algorithm, such as a multivariate linear regression model. This model uses different environmental factors as independent variables and analyzes historical data to determine the relationship between these factors and voltage transformer measurement results, thereby constructing a model capable of calculating the environmental assessment value.
[0058] The environmental feature set is generated based on the change trend of the first environmental evaluation value H in the historical environmental data. The change trend of the first environmental evaluation value H reflects the dynamic change of environmental factors over time or under different working conditions.
[0059] If H exhibits periodic fluctuations over time, this periodic fluctuation can be used as an element in the environmental feature set. This set includes features related to H's changing trend, such as its maximum and minimum values, fluctuation period, and amplitude. These features can describe the impact of environmental factors on voltage transformer measurements from different perspectives.
[0060] The second deviation value corresponding to each environmental feature sample in the environmental feature set is calculated in combination with the historical data. For each environmental feature sample in the environmental feature set, the deviation value is calculated by comparing the actual value of the actual voltage transformer measurement point with the measured value.
[0061] For example, if the environmental feature sample is the maximum value of the first environmental evaluation value H, then the actual value and the measured value of the voltage transformer measurement point when the maximum value occurs are found, and the deviation between them is calculated.
[0062] Example 8: The step of generating the first environmental evaluation value H further includes:
[0063] Among them, h y Indicates the reference value of environmental data of type y, a y represents the weight of the reference value of the yth type of environmental data, represents the correction value of the first environmental evaluation value H, Indicates the total number of environment data types.
[0064] Example 9: The generating of the second correction value C2 further includes: Obtain a curve of the change of the first environmental evaluation value H over time; Generate an environmental feature set based on the change curve, the environmental feature set includes multiple types of environmental feature samples; Obtain a second deviation value R between an actual value and a measured value of a voltage transformer measurement point corresponding to each environmental characteristic sample; The environment preset value is set based on the second deviation value R corresponding to all environmental feature samples; The second deviation is clustered by comparing the second deviation value R with the environmental preset value to generate a second correction value C2.
[0065] Example 10: The generating of the corrected voltage measurement data includes: Obtain the measured value L1 of the voltage transformer, the load voltage value, and environmental data at the current monitoring time node; Generate a first correction value c1 and a second correction value c2 by combining the measured value L1 of the voltage transformer, the load voltage value and the environmental data; Combining the voltage transformer's measurement value L1 and environmental data to generate a first correction value c1 and a second correction value c2 to generate a corrected voltage measurement L; L=L1+c1+c2.
[0066] In this embodiment, the corrected voltage measurement value L comprehensively considers the voltage transformer's own measurement value, the effect of the load on the measurement, and the effect of environmental factors on the measurement, and more accurately reflects the actual voltage conditions compared to the original measurement value L1. For example, if the measured value L1 has a certain deviation due to load changes or environmental factors before correction, after correcting it using the calculated c1 and c2, L can be closer to the actual voltage value.
[0067] Finally, it should be noted that it is apparent that those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, to the extent such modifications and variations fall within the scope of the present invention and its equivalents, the present invention is intended to include such modifications and variations.
[0068] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.
Claims
1. A voltage transformer measurement method based on multi-source data, characterized in that: include: Generate measurement history data and corresponding environmental history data based on the acquired historical operation data of the voltage transformer; constructing a first correction model of the voltage transformer based on the measurement history data; constructing a second correction model of the voltage transformer based on environmental historical data; The obtained measurement data of the voltage transformer at the current time node is corrected in combination with the first correction model and the second correction model to generate corrected voltage measurement data.
2. The voltage transformer measurement method based on multi-source data according to claim 1, characterized in that: The step of constructing the first modified model of the voltage transformer further includes: Obtain historical voltage data of a voltage transformer measurement point and historical voltage data of a load to generate historical measurement data; Generate a measurement point feature set V1 based on historical voltage data of the voltage transformer measurement point; Generate a load feature set V2 based on the historical voltage data of the load; Combine the measurement point feature set V1 and the load feature set V2 to generate a voltage correction evaluation value P1; Performing equal gradient division on the voltage correction evaluation value P1 to generate a first-level gradient data set of the voltage correction evaluation value P1, and obtaining a first deviation value B between the actual value and the measured value of the voltage transformer measurement point corresponding to each first-level gradient data set; Based on the first deviation value B, the voltage correction evaluation value P1 is gradient-divided again to generate a secondary gradient data set of the voltage correction evaluation value P1; The first correction value C1 is generated by combining the secondary gradient data set of the voltage correction evaluation value P1 and the corresponding deviation value.
3. The voltage transformer measurement method based on multi-source data according to claim 2, characterized in that: The generating of historical measurement data further includes: Obtain corresponding historical voltage data from the voltage transformer monitoring system and the load monitoring device respectively; Obtain historical voltage data of the voltage transformer measurement point and historical voltage data of the load at the same test time node; Combine the historical voltage data of the voltage transformer measurement point and the historical voltage data of the load to generate a sample data set D of historical measurement data, D = {D1, D2…Di…Dn}; Where Di=(U pi ,U li ), Di represents the sample point data at the i-th value-taking moment, n represents the total number of values taken at the current test time node, U pi Represents the historical voltage data reference value of the voltage transformer measurement point at the i-th value taking moment, U li Indicates the historical voltage data reference value of the load at the i-th value-taking moment.
4. The voltage transformer measurement method based on multi-source data according to claim 3, characterized in that: When generating the voltage correction evaluation value P1, the method further includes: Obtaining a historical voltage data reference value of a voltage transformer measurement point at the current test time node and a historical voltage data reference value of a load at the current test time node based on a sample data set D of historical measurement data at the current test time node; Based on the historical voltage data reference value of the voltage transformer measurement point, the measurement point feature set V1 is obtained, V1={v 11 ,v 12 …v 1j …v 1m }; The load characteristic set V2 is obtained based on the historical voltage data reference value of the load, V2={v 21 ,v 22 …v 2j …v 2m }; Among them, v 1j represents the jth type of eigenvalue in the measurement point feature set V1, 1m represents the total number of eigenvalue types in the measurement point feature set V1, and v 2j represents the j-th type of eigenvalue in the load feature set V2, and 2m represents the total number of eigenvalue types in the load feature set V2; Combine the measurement point feature set V1 and the load feature set V2 to generate a voltage correction evaluation value P1; in, is the weight of the j-th type of eigenvalue in the measurement point feature set V1, is the weight of the j-th type of eigenvalue in the load feature set V2, is the weight of the evaluation value of the measurement point feature set, The weight of the load feature set evaluation value.
5. The voltage transformer measurement method based on multi-source data according to claim 4, characterized in that: The generating of the secondary gradient data set of the voltage correction evaluation value P1 further includes: Obtaining a value range of the voltage correction evaluation value P1 based on the historical voltage correction evaluation value P1; Divide the value range of the voltage correction evaluation value P1 into intervals to generate k gradient intervals to generate a first-level gradient data set of the voltage correction evaluation value P1; For each first-level gradient data interval, obtain the actual value of the voltage transformer measurement point corresponding to the xth voltage correction evaluation value P1 in the current first-level gradient data interval With measured value ; Calculate a first deviation value B between the actual value and the measured value of the voltage transformer measurement point corresponding to each first-level gradient data; Wherein, xn represents the total number of voltage correction evaluation values P1 in the current first-level gradient data interval.
6. The voltage transformer measurement method based on multi-source data according to claim 5, characterized in that: The generating of the first correction value C1 further includes: Based on the first deviation value B, the voltage correction evaluation value P1 is gradient-divided again, and a comparison preset value Ba of the first deviation value B is set; If B>Ba, the corresponding first-level gradient data interval is removed from the first-level gradient data set, and the current first-level gradient data interval is divided into k intervals again to generate a temporary gradient data set; Combining the first-level gradient dataset and the temporary gradient dataset to generate a second-level gradient dataset; The first correction value C1 is generated by combining the secondary gradient data set of the voltage correction evaluation value P1 and the corresponding deviation value.
7. The voltage transformer measurement method based on multi-source data according to claim 6, characterized in that: When the second correction model of the voltage transformer is constructed based on the environmental historical data, the method further includes: Construct an environmental assessment value model based on historical environmental data; Combining historical environmental data and an environmental evaluation value model to generate a first environmental evaluation value H; Generate an environmental feature set based on a change trend of a first environmental evaluation value H in historical environmental data; The second deviation value corresponding to the environmental feature sample in each environmental feature set is calculated in combination with historical data to generate a second correction value C2.
8. The voltage transformer measurement method based on multi-source data according to claim 7, characterized in that: The step of generating the first environmental evaluation value H further includes: Among them, h y Indicates the reference value of environmental data of type y, a y represents the weight of the reference value of the yth type of environmental data, represents the correction value of the first environmental evaluation value H, Indicates the total number of environment data types.
9. The voltage transformer measurement method based on multi-source data according to claim 8, characterized in that: The generating of the second correction value C2 further includes: Obtain a curve of the change of the first environmental evaluation value H over time; Generate an environmental feature set based on the change curve, the environmental feature set includes multiple types of environmental feature samples; Obtain a second deviation value R between an actual value and a measured value of a voltage transformer measurement point corresponding to each environmental characteristic sample; The environment preset value is set based on the second deviation value R corresponding to all environmental feature samples; The second deviation is clustered by comparing the second deviation value R with the environmental preset value to generate a second correction value C2.
10. The voltage transformer measurement method based on multi-source data according to claim 9, characterized in that: The generating of the corrected voltage measurement data includes: Obtain the measured value L1 of the voltage transformer, the load voltage value, and environmental data at the current monitoring time node; Generate a first correction value c1 and a second correction value c2 by combining the measured value L1 of the voltage transformer, the load voltage value and the environmental data; Combining the measured value L1 of the voltage transformer and the environmental data to generate a first correction value c1 and a second correction value c2 to generate a corrected voltage measurement value L; L=L1+c1+c2.