A method and system for calculating power transformer losses

By selecting a voltage data window in the power transformer, analyzing the potential loss anomaly degree and correcting the similarity tolerance, the problem of inaccurate voltage data anomaly detection in the fuzzy entropy method is solved, and a more accurate loss calculation is achieved.

CN119986467BActive Publication Date: 2025-09-05XD JINAN TRANSFORMER +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510479415.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-04-17
Publication Date
2025-09-05
Estimated Expiration
2045-04-17

AI Technical Summary

Technical Problem

In the prior art, the fuzzy entropy method has a low accuracy in detecting abnormalities in power transformer voltage data, resulting in inaccurate calculation results of power transformer losses.

Method used

By selecting a voltage data window within a predetermined time period, analyzing the potential loss anomaly of the voltage data, correcting the similarity tolerance, and calculating the fuzzy entropy value using the corrected similarity tolerance, the voltage data anomaly is determined and the loss is calculated.

Benefits of technology

The accuracy of abnormal detection of power transformer voltage data is improved, the false detection rate is reduced, and the accuracy of loss calculation results is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119986467B_ABST
    Figure CN119986467B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of transformer loss detection, and more specifically to a method and system for calculating power transformer loss. The method comprises: selecting a voltage data window for voltage data at each moment by fitting a voltage curve during operation of the power transformer within a predetermined time period; determining the potential loss anomaly degree of the voltage data window at each moment; correcting the initial similarity tolerance of a subwindow in the voltage data window according to the potential loss anomaly degree to obtain a corrected similarity tolerance of the subwindow in the voltage data window; calculating a fuzzy entropy value of the voltage data window for the voltage data at each moment according to the corrected similarity tolerance; and determining that the voltage data in the voltage data window is abnormal and that the anomaly is caused by power transformer loss when the fuzzy entropy value is greater than a threshold value; and calculating the power transformer loss. Thus, the present invention improves the accuracy of the calculated power transformer loss calculation result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of transformer loss detection, and in particular to a method and system for calculating power transformer loss. Background Art

[0002] Power transformers are electrical devices used to exchange alternating voltage and current within power systems, transmitting and distributing electrical energy. As core equipment in power systems, they are widely used for power transmission and distribution. As power systems continue to grow in complexity, the safe and stable operation of power transformers is crucial to their stability.

[0003] In some scenarios, it is necessary to calculate the loss of a power transformer to understand its status. Currently, fuzzy entropy methods are often used to detect anomalies in the voltage data of a power transformer during operation. Abnormal voltage data is identified within the data, and the power transformer loss is calculated using this abnormal voltage data. However, when calculating the fuzzy entropy values ​​within the corresponding windows of all voltage data points, the fuzzy entropy method provides the same similarity tolerance for the calculation of the fuzzy membership between each subwindow within the corresponding window of each voltage data point, thereby calculating the power transformer loss. In actual scenarios, the changes in voltage data caused by a power transformer fault often have similar characteristics to the changes in voltage data caused by the power transformer's own losses, meaning that both will cause changes in voltage data. Therefore, the accuracy of voltage data anomaly detection using this method is low, resulting in inaccurate calculations of the power transformer loss. Summary of the Invention

[0004] In order to solve the technical problem that the accuracy of voltage data anomaly detection is low, which leads to inaccurate calculation results of power transformer losses, the purpose of the present invention is to provide a power transformer loss calculation method and system. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for calculating the loss of a power transformer, comprising: selecting a voltage data window for voltage data at each moment by fitting a voltage curve during the operation of the power transformer within a predetermined time period; determining the potential loss abnormality degree of the voltage data window at each moment based on the voltage data window at each moment and the adjacent data segments of the voltage data window at each moment, wherein the adjacent data segments are multiple data segments continuously adjacent to each other before the first voltage data in the voltage data window of the voltage data at each moment; correcting the initial similarity tolerance of the subwindow in the voltage data window according to the potential loss abnormality degree to obtain a corrected similarity tolerance of the subwindow in the voltage data window; calculating the fuzzy entropy value of the voltage data window of the voltage data at each moment according to the corrected similarity tolerance, and when the fuzzy entropy value is greater than a threshold, determining that the voltage data in the voltage data window is abnormal and the abnormality is caused by the loss of the power transformer; and calculating the loss of the power transformer using the time period corresponding to the voltage data window where the abnormal voltage data is located and the rated power of the power transformer.

[0006] Optionally, determining the potential loss abnormality level of the voltage data window at each moment based on the voltage data window at each moment and the adjacent data segments of the voltage data window at each moment includes: determining the right-skewed distribution factor of the voltage data window based on the voltage data in the voltage data window at each moment; determining the right-skewed distribution reliability of the voltage data window at each moment based on the first minimum voltage value of the voltage data in the voltage data window at each moment and the second minimum voltage value of the voltage data in the adjacent data segments adjacent to the voltage data window at each moment; determining the downward fluctuation level of the voltage data in the voltage data window at each moment based on the right-skewed distribution factor and the right-skewed distribution reliability; determining the smooth characteristic factor of the voltage data window at each moment based on the time interval between the maximum voltage value and the first minimum voltage value in the voltage data window at each moment and the variance of the change in voltage data at each moment in the voltage data window; and determining the potential loss abnormality level of the voltage data window at each moment using the downward fluctuation level, the smooth characteristic factor, the first minimum voltage value in the voltage data window at each moment, and the voltage data at the final neighborhood comparison moment of each moment.

[0007] Optionally, determining the right-skewed distribution factor of the voltage data window based on the voltage data in the voltage data window at each moment includes: calculating a first difference between the median of the voltage data in the voltage data window and the average voltage data value of the voltage data in the voltage data window; and normalizing the first difference to obtain the right-skewed distribution factor.

[0008] Optionally, determining the right-biased arrangement reliability of the voltage data window at each moment based on the first minimum voltage value of the voltage data in the voltage data window at each moment and the second minimum voltage value of the voltage data in the adjacent data segments adjacent to the voltage data window at each moment includes: calculating the second difference between each second minimum voltage value and the first minimum voltage value, and superimposing each second difference to obtain a first superimposed value; and normalizing the first superimposed value to obtain the right-biased arrangement reliability.

[0009] Optionally, determining the downward fluctuation degree of the voltage data in the voltage data window at each moment according to the right-skewed distribution factor and the right-skewed distribution reliability includes: determining the sum of the right-skewed distribution factor and the right-skewed distribution reliability as the downward fluctuation degree.

[0010] Optionally, determining the smooth characteristic factor of the voltage data window at each moment based on the time interval between the maximum voltage value and the first minimum voltage value in the voltage data window at each moment and the variance of the change in voltage data at each moment in the voltage data window includes: determining the ratio between the time interval and the variance as the smooth characteristic factor.

[0011] Optionally, the potential loss abnormality degree of the voltage data window at each moment is determined by using the degree of downward fluctuation, the smooth characteristic factor, the first minimum voltage value in the voltage data window at each moment and the voltage data at the final neighborhood comparison moment at each moment, including: calculating the third difference between the voltage data at each final neighborhood comparison moment and the first minimum voltage value, and superimposing each third difference to obtain a second superimposed value; performing inverse proportional normalization processing on the second superimposed value to obtain a first normalized value; determining the first product of the first normalized value, the degree of downward fluctuation and the smooth characteristic factor as the potential loss abnormality degree.

[0012] Optionally, the initial similarity tolerance of the sub-window in the voltage data window is corrected according to the potential loss abnormality level, and the corrected similarity tolerance of the sub-window in the voltage data window includes: the initial similarity tolerance of the sub-window in the voltage data window in the preset fuzzy entropy algorithm and the length and number of the sub-windows; the potential loss abnormality level is inversely normalized to obtain a second normalized value; and the second product between the second normalized value and the initial similarity tolerance is determined as the corrected similarity tolerance of the sub-window in the voltage data window.

[0013] Optionally, calculating the loss of the power transformer using the time period corresponding to the voltage data window where the abnormal voltage data occurs and the rated power of the power transformer includes: superimposing the time periods corresponding to all voltage data windows where the abnormal voltage data occurs to obtain the superposition time; and determining the third product between the superposition time and the rated power as the loss of the power transformer.

[0014] In a second aspect, an embodiment of the present invention provides a power transformer loss calculation system, comprising: a processor and a memory; wherein the memory is used to store a computer program that can be run on the processor; and the processor is used to execute the program stored in the memory to implement the steps of the power transformer loss calculation method mentioned in the first aspect.

[0015] The present invention has the following beneficial effects: first, a voltage data window is selected for voltage data at each moment by means of a voltage fitting curve during operation of a power transformer within a predetermined time period; then, a potential loss abnormality degree of the voltage data window at each moment is determined according to the voltage data window at each moment and adjacent data segments of the voltage data window at each moment, wherein the adjacent data segments are a plurality of data segments continuously adjacent to the first voltage data in the voltage data window of the voltage data at each moment; and an initial similarity tolerance of a subwindow in the voltage data window is corrected according to the potential loss abnormality degree to obtain a corrected similarity tolerance of the subwindow in the voltage data window; secondly, a fuzzy entropy value of the voltage data window of the voltage data at each moment is calculated according to the corrected similarity tolerance, and when the fuzzy entropy value is greater than a threshold value, it is determined that the voltage data in the voltage data window is abnormal and the abnormality is caused by the loss of the power transformer; finally, the loss of the power transformer is calculated using the time period corresponding to the voltage data window where the abnormal voltage data is located and the rated power of the power transformer.

[0016] In this way, the embodiment of the present invention analyzes the changing characteristics of the voltage data of the power transformer and introduces the potential loss anomaly degree to specifically adjust the similarity tolerance of the sub-window in the voltage data window of each voltage data in the fuzzy entropy algorithm. This method can more accurately reflect the true characteristics of the voltage change of the power transformer, effectively distinguish the abnormal voltage change caused by the fault from the abnormal voltage change caused by the power transformer's own loss, reduce the occurrence of false detection, improve the accuracy of the abnormal detection of the voltage data in the power transformer, reduce the false detection rate of the abnormal voltage data, and thus improve the accuracy of the calculated loss calculation results of the power transformer. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a method for calculating power transformer losses provided by one embodiment of the present invention;

[0019] Figure 2 A schematic structural diagram of a power transformer loss calculation system provided by one embodiment of the present invention;

[0020] Figure 3 A schematic structural diagram of a power transformer loss calculation system provided in another embodiment of the present invention. DETAILED DESCRIPTION

[0021] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method and system for calculating power transformer losses according to the present invention, including its specific implementation, structure, features, and effectiveness. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0022] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0023] The specific scheme of the power transformer loss calculation method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0024] Example 1:

[0025] See also Figure 1 , which shows a flow chart of a method for calculating power transformer losses provided by one embodiment of the present invention, including:

[0026] S101 , selecting a voltage data window for voltage data at each moment by using a voltage fitting curve during operation of a power transformer within a predetermined time period.

[0027] Specifically, the embodiment of the present invention installs a voltage sensor at the input terminal of the power transformer to record voltage data during the operation of the power transformer. The predetermined time period can be determined based on actual conditions. In the embodiment of the present invention, the value is 2 hours, and the collection frequency is five times per second.

[0028] Furthermore, embodiments of the present invention use an analog-to-digital conversion device to digitize voltage data to obtain a digital representation of the voltage data. A least squares curve fitting technique is then used to fit the digitized voltage data at various moments during the operation of the power transformer within a predetermined time period to obtain a voltage fitting curve. The voltage fitting curve includes a plurality of voltage data points, and a voltage data window is then selected from the voltage fitting curve for each voltage data point at each moment.

[0029] Furthermore, the embodiment of the present invention presets the size of the voltage data window to be , the voltage data before the sampling time The voltage data at each moment is used as a window of each voltage data. It should be noted that the size of the voltage data window can also be determined according to the actual scenario, and the embodiment of the present invention does not limit this.

[0030] S102 , determining the potential loss abnormality degree of the voltage data window at each moment according to the voltage data window at each moment and the adjacent data segments of the voltage data window at each moment.

[0031] The adjacent data segments are a plurality of data segments that are continuous and adjacent to the first voltage data in the voltage data window of the voltage data at each moment.

[0032] Specifically, the embodiment of the present invention can select multiple continuous adjacent data segments for the voltage data window of the voltage data at each moment. The number of adjacent data segments can be determined according to actual conditions, and in the embodiment of the present invention, the value is 6. For example: for any voltage data window, the 300 voltage data before the sampling moment of the first voltage data in the voltage data window are used as the first adjacent data segment of the voltage data window. The voltage data at the 300 sampling moments before the first voltage data in the first adjacent data segment are used as the second adjacent data segment of the voltage data window, and so on, until the 6th adjacent data segment of the voltage data window is obtained. Among them, the voltage data in each adjacent data segment is different, but the number is the same. In addition, if the amount of data before the sampling moment of the voltage data window is insufficient to construct its adjacent data segments, six minutes of voltage data can be collected in advance to ensure that the adjacent data segments of the voltage data window can be constructed.

[0033] Furthermore, since when the power transformer itself has losses, the corresponding change in the voltage passing through the power transformer is a decrease. Therefore, when analyzing the change characteristics of the voltage data during the operation of the power transformer, if the numerical distribution characteristics of the voltage data in the voltage data window of each voltage data are more in line with the right-skewed distribution, and the minimum value of the voltage data in each voltage data window is smaller than the minimum value of its adjacent data segment, it means that the voltage data value in the voltage data window is lower, and the power transformer is more likely to have potential abnormal losses in the time period corresponding to the voltage data window, and the corresponding voltage data will also have a greater degree of downward fluctuation. Therefore, the embodiment of the present invention determines the potential loss abnormality degree of the voltage data window based on the downward fluctuation degree of the voltage data window.

[0034] Furthermore, when determining the degree of downward fluctuation, embodiments of the present invention determine the degree of downward fluctuation of the voltage data by analyzing the numerical distribution of the voltage data within the voltage data window. The closer the voltage data within the voltage data window approaches a right-skewed distribution, the lower the numerical value of the voltage data within the voltage data window, indicating that the corresponding change in the voltage passing through the power transformer is decreasing. This increases the likelihood of potential abnormal losses in the power transformer during the time period corresponding to the voltage data window, and the corresponding degree of downward fluctuation will also be greater.

[0035] Further, as an optional embodiment of the present invention, determining the potential loss abnormality degree of the voltage data window at each moment based on the voltage data window at each moment and the adjacent data segments of the voltage data window at each moment includes: determining the right-skewed distribution factor of the voltage data window based on the voltage data in the voltage data window at each moment; determining the right-skewed distribution reliability of the voltage data window at each moment based on the first minimum voltage value of the voltage data in the voltage data window at each moment and the second minimum voltage value of the voltage data in the adjacent data segments adjacent to the voltage data window at each moment; determining the downward fluctuation degree of the voltage data in the voltage data window at each moment based on the right-skewed distribution factor and the right-skewed distribution reliability; determining the smooth characteristic factor of the voltage data window at each moment based on the time interval between the maximum voltage value and the first minimum voltage value in the voltage data window at each moment and the variance of the change in the voltage data at each moment in the voltage data window; and determining the potential loss abnormality degree of the voltage data window at each moment using the downward fluctuation degree, the smooth characteristic factor, the first minimum voltage value in the voltage data window at each moment, and the voltage data at the final neighborhood comparison moment of each moment.

[0036] Specifically, as an optional embodiment of the present invention, determining the right-skewed distribution factor of the voltage data window based on the voltage data in the voltage data window at each moment includes: calculating a first difference between the median of the voltage data in the voltage data window and the average voltage data value of the voltage data in the voltage data window; and normalizing the first difference to obtain the right-skewed distribution factor.

[0037] The embodiment of the present invention uses the following formula to calculate the right-skewed distribution factor:

[0038]

[0039] In the above formula, Indicates the The right-skewed distribution factor of a voltage data window. Indicates the The median of the voltage data within a voltage data window. Indicates the The average voltage data value of the voltage data in a voltage data window. Represents the normalization function, which is used to Perform normalization processing.

[0040] Furthermore, since when there is loss in the power transformer itself, the corresponding change in the voltage passing through the power transformer is a decrease, the smaller the minimum value of the voltage data in the voltage data window is compared with the minimum value of the voltage data in its adjacent data segment, the greater the reliability of the right-skew distribution of the voltage data in the voltage data window, and the more likely the power transformer is to have potential abnormal loss in the time period corresponding to the voltage data window, and the greater the corresponding downward fluctuation degree will be.

[0041] Further, as an optional embodiment of the present invention, determining the right-biased arrangement reliability of the voltage data window at each moment based on the first minimum voltage value of the voltage data in the voltage data window at each moment and the second minimum voltage value of the voltage data in the adjacent data segments adjacent to the voltage data window at each moment includes: calculating the second difference between each second minimum voltage value and the first minimum voltage value, and superimposing each second difference to obtain a first superimposed value; and normalizing the first superimposed value to obtain the right-biased arrangement reliability.

[0042] Specifically, the embodiment of the present invention uses the following formula to calculate the right-biased distribution reliability:

[0043]

[0044] In the above formula, Indicates the The right-biased distribution of voltage data within a voltage data window is used to determine the reliability of the voltage data. Indicates the The number of adjacent data segments in a voltage data window. Indicates the The first voltage data window The minimum value of the voltage data in two adjacent data segments is the second minimum voltage value. Indicates the The minimum value of the voltage data in a voltage data window is the first minimum voltage value. Represents the normalization function, which is used to Perform normalization processing.

[0045] Furthermore, when the numerical distribution of the voltage data within the voltage data window approaches a right-skewed distribution, it indicates that the power transformer is more likely to experience potential abnormal losses during the time period corresponding to the voltage data window, and the corresponding degree of downward fluctuation will also be greater. As an optional embodiment of the present invention, determining the degree of downward fluctuation of the voltage data in the voltage data window at each moment based on the right-skewed distribution factor and the right-skewed distribution reliability includes: determining the sum of the right-skewed distribution factor and the right-skewed distribution reliability as the degree of downward fluctuation.

[0046] Specifically, the embodiment of the present invention uses the following formula to calculate the degree of downward fluctuation:

[0047]

[0048] In the above formula, Indicates the The degree of voltage fluctuation within a voltage data window. Indicates the The right-biased distribution of voltage data within a voltage data window is used to determine the reliability of the voltage data. Indicates the The right-skewed distribution factor of a voltage data window.

[0049] Furthermore, the degree of downward fluctuation of the voltage data in each voltage data window is determined by analyzing whether there is a fluctuation of voltage value reduction in the time period corresponding to a voltage data window. However, the voltage change caused by the aging failure of the load equipment usually has similar characteristics to the change of voltage data caused by the loss of the power transformer itself, that is, both will cause the voltage data in the pipeline to decrease. Therefore, it is not possible to distinguish the voltage value reduction caused by the above two situations by only analyzing the numerical performance of the voltage data in the voltage data window, which will eventually lead to errors in the abnormal detection results of the voltage data. Through analysis, the embodiment of the present invention finds that the change trend characteristics of the voltage data caused by the aging failure of the load equipment are different from the change trend of the voltage data caused by the loss of the power transformer itself. The change of voltage data caused by the aging failure of the load equipment is more drastic, but will return to stability in a short time, while the change of voltage data caused by the loss of the power transformer itself is relatively gentle and continuous.

[0050] Therefore, the embodiment of the present invention analyzes the numerical variation characteristics of the voltage data in each voltage data window, and optimizes the degree of downward fluctuation of the voltage data in each voltage data window to obtain the potential loss anomaly degree of each voltage data window. Among them, the lower the relative size of the voltage data at the last sampling moment in each voltage data window in the time period corresponding to the voltage data window, and the smoother the numerical variation between the maximum voltage value and the minimum voltage value in each voltage data window, it can be explained that the value of the voltage data in the voltage data window is continuous and the greater the possibility of a gradual decrease, and then it can be explained that the more likely the power transformer itself is to have losses in the time period corresponding to this voltage data window, the greater the potential loss anomaly degree in the voltage data window will be.

[0051] Furthermore, as an optional embodiment of the present invention, determining the smooth characteristic factor of the voltage data window at each moment based on the time interval between the maximum voltage value and the first minimum voltage value in the voltage data window at each moment and the variance of the change in voltage data at each moment in the voltage data window includes: determining the ratio between the time interval and the variance as the smooth characteristic factor.

[0052] Specifically, the embodiment of the present invention is In the voltage data window The voltage data at the first sampling moment is The absolute value of the difference between the voltage data at the sampling moment is recorded as In the voltage data window The change in voltage data at each sampling moment. In the embodiment of the present invention, the flat characteristic factor of the voltage data window at each moment is calculated by the following formula:

[0053]

[0054] In the above formula, Indicates the The flat characteristic factor within a voltage data window. Indicates the The time interval between the maximum voltage value and the minimum value (the first minimum voltage value) in a voltage data window. Indicates the The variance of the voltage data changes at all sampling moments within a voltage data window.

[0055] In the above formula, The larger the The longer it takes for the maximum voltage value to change to the minimum voltage value in a voltage data window, the The smoother the change in value between the maximum voltage value and the minimum voltage value in a voltage data window, the greater the smooth characteristic factor. The smaller the The more consistent the voltage reduction between each group of adjacent sampling moments in the interval from the maximum voltage value to the minimum voltage value in the voltage data window, the more consistent the voltage reduction between each group of adjacent sampling moments is ... The smoother the change in value between the maximum voltage and the minimum voltage in a voltage data window, the greater the credibility, and the corresponding smooth characteristic factor will also be greater.

[0056] Furthermore, the greater the degree of downward fluctuation of the voltage data within the voltage data window, the more likely the power transformer is to have an abnormal risk during the time period corresponding to the voltage data window, and the greater the potential abnormality of its own loss should be. The smaller the value of the voltage data at the last sampling moment in the voltage data window, the more likely the minimum voltage in the voltage data window is to occur at the last sampling moment. The more the voltage variation characteristics within the voltage data window conform to the characteristics of the voltage value continuously decreasing due to the loss of the power transformer itself, the more likely the power transformer is to have a loss risk during the time period corresponding to the voltage data window, and the greater the potential abnormality of its own loss should be. In addition, the larger the flat characteristic factor within the voltage data window, the more likely the voltage value within the voltage data window is to have a continuous and gradual decrease. The more the voltage variation characteristics within the voltage data window conform to the characteristics of the voltage value continuously decreasing due to the loss of the power transformer itself, the more likely the power transformer is to have a loss risk during the time period corresponding to the voltage data window, and the greater the potential abnormality of its own loss should be.

[0057] Furthermore, as an optional embodiment of the present invention, the potential loss abnormality degree of the voltage data window at each moment is determined by using the downward fluctuation degree, the smooth characteristic factor, the first minimum voltage value in the voltage data window at each moment and the voltage data at the final neighborhood comparison moment at each moment, including: calculating the third difference between the voltage data at each final neighborhood comparison moment and the first minimum voltage value, and superimposing each third difference to obtain a second superimposed value; performing inverse proportional normalization processing on the second superimposed value to obtain a first normalized value; determining the first product of the first normalized value, the downward fluctuation degree and the smooth characteristic factor as the potential loss abnormality degree.

[0058] Specifically, the embodiment of the present invention is The five sampling moments before the last sampling moment in the voltage data window are used as the final neighborhood comparison moment, where the final neighborhood comparison moment includes the last sampling moment. Furthermore, the embodiment of the present invention uses the following formula to calculate the potential loss abnormality degree:

[0059]

[0060] Where, Indicates the The potential loss anomaly level for each voltage data window. Indicates the The minimum value of the voltage data in a voltage data window is the first minimum voltage value. Indicates the The number of all final neighborhood comparison moments within a voltage data window. Indicates the In the voltage data window The voltage data at the final neighborhood comparison moment. Indicates the The flat characteristic factor within a voltage data window. Indicates the The degree of voltage fluctuation within a voltage data window. (-) indicates the inverse normalization function, which is used to Perform inverse proportional normalization.

[0061] Thus, the embodiment of the present invention obtains the potential loss abnormality degree of the voltage data window of each voltage data.

[0062] S103 , correcting the initial similarity tolerance of the sub-window in the voltage data window according to the potential loss abnormality degree to obtain a corrected similarity tolerance of the sub-window in the voltage data window.

[0063] Specifically, the greater the potential loss anomaly in each voltage data window, the greater the possibility that the voltage data in the voltage data window is caused by the loss of the power transformer itself. Therefore, in order to more accurately and conveniently identify the loss risk of the voltage data itself, the similarity tolerance between the sub-windows should be smaller when calculating the fuzzy entropy value in the voltage data window to ensure that the final calculated fuzzy entropy value is larger.

[0064] Furthermore, as an optional embodiment of the present invention, the initial similarity tolerance of the sub-window in the voltage data window is corrected according to the potential loss abnormality degree, and the corrected similarity tolerance of the sub-window in the voltage data window includes: the initial similarity tolerance of the sub-window in the voltage data window in the preset fuzzy entropy algorithm and the length and number of the sub-windows; the potential loss abnormality degree is inversely normalized to obtain a second normalized value; and the second product between the second normalized value and the initial similarity tolerance is determined as the corrected similarity tolerance of the sub-window in the voltage data window.

[0065] Specifically, in the embodiment of the present invention, the initial similarity tolerance, the length and number of sub-windows can be determined according to actual conditions. In the embodiment of the present invention, the initial similarity tolerance of the sub-windows in the preset voltage data window is The length of the sub-window is set to 10, and the number is set to the length of the voltage data window - the length of each sub-window + 1.

[0066] Furthermore, the embodiment of the present invention uses the following formula to calculate the modified similarity tolerance:

[0067]

[0068] In the above formula, Indicates calculation of The fuzzy entropy value within a voltage data window is the modified similarity tolerance between sub-windows. Indicates the The potential loss anomaly level for each voltage data window. Represents the initial similarity tolerance. (-) indicates the inverse normalization function, which is used to Perform inverse proportional normalization.

[0069] At this point, the modified similarity tolerance between sub-windows when calculating the fuzzy entropy value in each voltage data window is obtained. The modified similarity tolerance is a parameter set when calculating the fuzzy entropy value in the voltage data window using the fuzzy entropy algorithm.

[0070] S104, calculating the fuzzy entropy value of the voltage data window of the voltage data at each moment according to the modified similarity tolerance, and when the fuzzy entropy value is greater than a threshold, determining that the voltage data in the voltage data window is abnormal and the abnormality is caused by the loss of the power transformer.

[0071] Specifically, in the embodiments of the present invention, a known technique is employed to calculate an initial fuzzy entropy value within a voltage data window for each voltage data using a modified similarity tolerance, and then the initial fuzzy entropy value is normalized to obtain a fuzzy entropy value for each voltage data window. The use of the modified similarity tolerance to calculate the initial fuzzy entropy value within the voltage data window for each voltage data can be referenced to known techniques. It is understood that the initial fuzzy entropy value can be obtained by using the modified similarity tolerance as the original similarity tolerance in the fuzzy entropy calculation process, and this embodiment of the present invention will not be further described herein.

[0072] Furthermore, the threshold value can be set according to the actual situation. In the embodiment of the present invention, the threshold value If the fuzzy entropy value is greater than 0.9, it can be said that the voltage data within this voltage data window is abnormal. The abnormal voltage data within the time period corresponding to this voltage data window is caused by the loss of the power transformer itself. When the embodiment of the present invention determines that the power transformer itself has lost power, it issues an early warning and activates an automatic alarm program to notify the operator to take timely measures. The threshold value can be determined according to actual conditions and is not limited in this embodiment of the present invention.

[0073] S105 , calculating the loss of the power transformer using the time period corresponding to the voltage data window where the abnormal voltage data occurs and the rated power of the power transformer.

[0074] Specifically, after determining that the abnormal voltage data is caused by the loss of the power transformer itself, as an optional embodiment of the present invention, the loss of the power transformer is calculated using the time period corresponding to the voltage data window where the abnormal voltage data is located and the rated power of the power transformer, including: superimposing the time periods corresponding to the voltage data windows where all abnormal voltage data are located to obtain the superposition time; and determining the third product between the superposition time and the rated power as the loss of the power transformer.

[0075] More specifically, the embodiment of the present invention superimposes the time periods corresponding to the voltage data windows where all abnormal voltage data occur to obtain the superimposed time. The rated power of the power transformer is then obtained from the power transformer's manual. The superimposed time and the rated power are multiplied to obtain the power transformer loss.

[0076] The embodiment of the present invention analyzes the variation characteristics of the voltage data of the power transformer and introduces the potential loss anomaly degree to specifically adjust the similarity tolerance of the sub-window in the voltage data window of each voltage data in the fuzzy entropy algorithm. This method can more accurately reflect the true characteristics of the voltage variation of the power transformer, effectively distinguish the abnormal voltage variation caused by faults from the abnormal voltage variation caused by the power transformer itself, improve the accuracy of the abnormal detection of voltage data in the power transformer, reduce the false detection rate of abnormal voltage data, and thus improve the accuracy of the calculated loss calculation results of the power transformer.

[0077] Example 2:

[0078] Corresponding to the power transformer loss calculation method provided in the above embodiment, based on the same technical concept, an embodiment of the present invention further provides a power transformer loss calculation system, which is used to execute the above power transformer loss calculation method. Figure 2 A schematic diagram of a power transformer loss calculation system provided by an embodiment of the present invention is shown in FIG. Figure 2As shown. The power transformer loss calculation system 200 includes: a selection module 201, configured to select a voltage data window for voltage data at each moment by using a voltage fitting curve during the operation of the power transformer within a predetermined time period; a determination module 202, configured to determine the potential loss anomaly degree of the voltage data window at each moment based on the voltage data window at each moment and adjacent data segments of the voltage data window at each moment, where the adjacent data segments are multiple data segments that are consecutively adjacent to the first voltage data in the voltage data window at each moment; a correction module 203, configured to correct the initial similarity tolerance of a subwindow in the voltage data window according to the potential loss anomaly degree to obtain a corrected similarity tolerance of the subwindow in the voltage data window; the determination module 202 is further configured to calculate a fuzzy entropy value of the voltage data window of the voltage data at each moment based on the corrected similarity tolerance, and when the fuzzy entropy value is greater than a threshold, determine that the voltage data in the voltage data window is abnormal and the anomaly is caused by the loss of the power transformer; and a calculation module 204, configured to calculate the loss of the power transformer using the time period corresponding to the voltage data window where the abnormal voltage data is located and the rated power of the power transformer.

[0079] The embodiment of the present invention analyzes the variation characteristics of the voltage data of the power transformer and introduces the potential loss anomaly degree to specifically adjust the similarity tolerance of the sub-window in the voltage data window of each voltage data in the fuzzy entropy algorithm. This method can more accurately reflect the true characteristics of the voltage variation of the power transformer, effectively distinguish the abnormal voltage variation caused by faults from the abnormal voltage variation caused by the power transformer itself, improve the accuracy of the abnormal detection of voltage data in the power transformer, reduce the false detection rate of abnormal voltage data, and thus improve the accuracy of the calculated loss calculation results of the power transformer.

[0080] Example 3:

[0081] Corresponding to the power transformer loss calculation method provided in the above embodiment, based on the same technical concept, an embodiment of the present invention further provides a power transformer loss calculation system, which is used to execute the above power transformer loss calculation method. Figure 3 A schematic diagram of a power transformer loss calculation system provided in another embodiment of the present invention is shown in FIG. Figure 3 The power transformer loss calculation system may have relatively large differences due to different configurations or performances, and may include one or more processors 301 and memory 302, the memory 302 is used to store computer programs that can be run on the processor 301, and the processor 301 is used to execute the program stored in the memory 302 to achieve the above Figure 1The various steps in the method embodiment are described above. Memory 302 may be either a temporary storage device or a persistent storage device. The application stored in memory 302 may include one or more modules (not shown), each of which may include a series of computer-executable instructions for the power transformer loss calculation system.

[0082] Furthermore, the processor 301 can be configured to communicate with the memory 302 and execute a series of computer-executable instructions in the memory 302 on the power transformer loss calculation system. The power transformer loss calculation system can also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, and one or more keyboards 306.

[0083] Specifically in this embodiment, the power transformer loss calculation system includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to achieve the above Figure 1 The various steps in the method embodiment have the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described again here.

[0084] It should be noted that the power transformer loss calculation system provided in the embodiment of the present invention and the power transformer loss calculation method provided in the embodiment of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned power transformer loss calculation method, and has the same or similar beneficial effects, and the repeated parts will not be repeated.

[0085] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0086] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for calculating power transformer losses, characterized in that: The power transformer loss calculation method includes: Selecting a voltage data window for voltage data at each moment by using a voltage fitting curve during operation of the power transformer within a predetermined time period; Determining the potential loss anomaly degree of the voltage data window at each moment based on the voltage data window at each moment and adjacent data segments of the voltage data window at each moment, wherein the adjacent data segments are a plurality of consecutive adjacent data segments preceding the first voltage data in the voltage data window at each moment; Correcting the initial similarity tolerance of the sub-window in the voltage data window according to the potential loss anomaly degree to obtain a corrected similarity tolerance of the sub-window in the voltage data window; Calculating a fuzzy entropy value of a voltage data window of the voltage data at each moment according to the modified similarity tolerance, and determining that the voltage data in the voltage data window is abnormal and that the abnormality is caused by a loss of the power transformer when the fuzzy entropy value is greater than a threshold value; Calculating the loss of the power transformer using a time period corresponding to a voltage data window in which the abnormal voltage data occurs and the rated power of the power transformer; The step of correcting the initial similarity tolerance of the sub-window in the voltage data window according to the potential loss anomaly degree to obtain the corrected similarity tolerance of the sub-window in the voltage data window includes: Presetting the initial similarity tolerance of sub-windows in the voltage data window in the fuzzy entropy algorithm and the length and number of sub-windows; Performing inverse proportional normalization processing on the potential loss abnormality degree to obtain a second normalized value; A second product of the second normalized value and the initial similarity margin is determined as a modified similarity margin of the sub-window in the voltage data window.

2. The power transformer loss calculation method according to claim 1, characterized in that: Determining the potential loss abnormality degree of the voltage data window at each moment based on the voltage data window at each moment and the adjacent data segments of the voltage data window at each moment includes: determining a right-skew distribution factor of the voltage data window according to voltage data in the voltage data window at each moment; Determining the right-biased distribution reliability of the voltage data window at each moment according to a first minimum voltage value of the voltage data in the voltage data window at each moment and a second minimum voltage value of the voltage data in an adjacent data segment adjacent to the voltage data window at each moment; determining a downward fluctuation degree of the voltage data in the voltage data window at each moment according to the right-skewed distribution factor and the right-skewed distribution reliability; Determining a smooth characteristic factor of the voltage data window at each moment according to a time interval between a maximum voltage value and a first minimum voltage value in the voltage data window at each moment and a variance of a change in voltage data at each moment in the voltage data window; The potential loss abnormality degree of the voltage data window at each moment is determined by using the downward fluctuation degree, the flat characteristic factor, the first minimum voltage value in the voltage data window at each moment and the voltage data at the final neighborhood comparison moment at each moment.

3. The method for calculating power transformer loss according to claim 2, characterized in that: Determining the right-skew distribution factor of the voltage data window according to the voltage data in the voltage data window at each moment includes: Calculate a first difference between a median of the voltage data in the voltage data window and an average voltage data value of the voltage data in the voltage data window; Normalizing the first difference to obtain the right-skewed distribution factor.

4. The method for calculating power transformer loss according to claim 2, wherein: Determining the right-biased arrangement reliability of the voltage data window at each moment according to the first minimum voltage value of the voltage data in the voltage data window at each moment and the second minimum voltage value of the voltage data in the adjacent data segments adjacent to the voltage data window at each moment includes: calculating a second difference between each of the second minimum voltage values ​​and the first minimum voltage value, and superimposing each of the second differences to obtain a first superimposed value; Normalization is performed on the first superposition value to obtain the right-biased distribution reliability.

5. The method for calculating power transformer loss according to claim 2, characterized in that: The determining of the downward fluctuation degree of the voltage data in the voltage data window at each moment according to the right-skewed distribution factor and the right-skewed distribution reliability comprises: The sum of the right-skewed distribution factor and the right-skewed distribution confidence is determined to be the degree of downward fluctuation.

6. The method for calculating power transformer loss according to claim 2, characterized in that: Determining the smooth characteristic factor of the voltage data window at each moment according to the time interval between the maximum voltage value and the first minimum voltage value in the voltage data window at each moment and the variance of the voltage data change at each moment in the voltage data window includes: The ratio between the time interval and the variance is determined as the flat characteristic factor.

7. The method for calculating power transformer loss according to claim 2, characterized in that: Determining the potential loss abnormality degree of the voltage data window at each moment by using the downward fluctuation degree, the smooth characteristic factor, the first minimum voltage value in the voltage data window at each moment, and the voltage data at the final neighborhood comparison moment at each moment includes: Calculating a third difference between the voltage data at each of the final neighborhood comparison moments and the first minimum voltage value, and superimposing the third differences to obtain a second superimposed value; performing an inverse proportional normalization process on the second superposition value to obtain a first normalized value; A first product of the first normalized value, the downward fluctuation degree, and the flat characteristic factor is determined as the potential loss abnormality degree.

8. The method for calculating power transformer loss according to claim 1, wherein: The calculating of the loss of the power transformer by using the time period corresponding to the voltage data window where the abnormal voltage data is located and the rated power of the power transformer includes: Superimpose the time periods corresponding to the voltage data windows where all abnormal voltage data occur to obtain the superposition time; A third product of the superposition time and the rated power is determined to be the loss of the power transformer.

9. A power transformer loss calculation system, characterized in that: include: A processor and a memory; wherein the memory is used to store a computer program that can be run on the processor; A processor is configured to execute a program stored in a memory to implement the steps of the power transformer loss calculation method according to any one of claims 1 to 8.

Citation Information

Patent Citations

  • Power battery pack fault on-line diagnosis method and system

    CN115097319A

  • Data processing method and system for flexible regulation and control terminal

    CN119474831A