Winding change degree monitoring method and device, medium and equipment

By acquiring and clustering analysis of vibration data under different operating modes of the transformer, the problem of inaccurate judgment of the degree of winding changes in the existing technology is solved, accurate monitoring and fault warning of the health status of the transformer is achieved, and the safety of the equipment is improved. reliability.

CN119044836APending Publication Date: 2024-11-29YUNNAN POWER GRID CO LTD ELECTRIC POWER RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411144466.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The prior art cannot accurately determine the degree of change in the transformer winding, resulting in missed fault detection and may even cause major accidents.

Method used

By obtaining sample vibration data under different operating modes of the transformer, calculating characteristic values ​​and performing cluster analysis, drawing reference diagrams, obtaining monitoring vibration data under the target operating mode, reclustering and drawing monitoring diagrams, and according to reference diagrams and monitoring diagrams The degree of overlap determines the degree of change of the winding.

Benefits of technology

Accurate monitoring of the degree of change in the transformer windings, timely identification of deterioration, reduce equipment failures, and improve operational safety and reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119044836A_ABST
    Figure CN119044836A_ABST
Patent Text Reader

Abstract

The invention discloses a winding change degree monitoring method and device, a medium and equipment, and the method comprises the steps: firstly obtaining K groups of sample vibration data of each mode, calculating a feature value, and obtaining a plurality of reference diagrams of each operation mode through clustering analysis; and in an actual target operation mode, acquiring K groups of monitoring vibration data, calculating characteristic values of the monitoring vibration data, and performing re-clustering based on the clustering center of the reference graph to obtain a monitoring graph. The change degree of the winding can be determined by comparing the coincidence degree of the reference diagram and the monitoring diagram, and the higher the coincidence degree is, the smaller the change of the winding is. According to the method, the health conditions of the transformer in different operation states can be accurately monitored by analyzing the vibration data in detail and performing clustering comparison. The deterioration condition of the winding can be recognized in time, potential equipment faults are reduced, and the operation safety and reliability of the transformer are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of high voltage testing of electrical equipment, and in particular to a method, device, medium and equipment for monitoring the degree of winding changes. Background Art

[0002] Transformer windings will constantly change during operation, especially for transformers whose windings are already in poor condition. For such transformers, vibration severity detection is often required, but in the existing technology, there is no diagnostic algorithm that can accurately determine the extent of further changes in the windings, which can cause faults to be missed and even lead to major accidents. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, medium and equipment for monitoring the degree of winding changes to solve the problem in the prior art that there is no diagnostic algorithm that can accurately determine the degree of further changes in the windings.

[0004] A method for monitoring winding variation degree, the method comprising:

[0005] Under various operation modes preset by the transformer, K groups of sample vibration data under each operation mode are obtained respectively; wherein K≥2;

[0006] In each operation mode, the characteristic value of each group of sample vibration data is calculated respectively to obtain K sample characteristic values, and the K sample characteristic values ​​are clustered to obtain N first clustering clusters; wherein N≥2;

[0007] Draw corresponding reference graphs based on the N first clusters respectively to obtain multiple reference graphs under the multiple operation modes; wherein, in the reference graphs, the cluster centers and sample quantities of the N first clusters are used as coordinate axis values;

[0008] Under the target operation mode, K groups of monitoring vibration data are obtained; wherein the target operation mode is any one of the multiple operation modes;

[0009] Calculating the characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values, and clustering the K monitoring characteristic values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters;

[0010] Drawing a monitoring graph based on the N second clusters; wherein, in the monitoring graph, the cluster centers and sample quantities of the N second clusters are used as coordinate axis values;

[0011] The degree of change of the winding is determined according to the degree of overlap between the reference diagram and the monitoring diagram under the target operation mode; wherein the degree of overlap is negatively correlated with the degree of change of the winding.

[0012] In one embodiment, respectively calculating the characteristic value of each group of sample vibration data to obtain K sample characteristic values ​​includes:

[0013] Calculate the signal sharpness of each group of sample vibration data respectively, and calculate the root mean square average value of each group of sample vibration data to obtain K signal sharpness and root mean square average values; wherein the signal sharpness is used to indicate the sharpness of the signal;

[0014] The signal sharpness and the root mean square average value of each group of sample vibration data form a sample array to obtain K sample arrays, and the K sample arrays are used as the K sample feature values;

[0015] The calculating the characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values ​​includes:

[0016] Calculate the signal sharpness of each group of monitoring vibration data respectively, and calculate the root mean square average value of each group of monitoring vibration data to obtain K signal sharpness and root mean square average values;

[0017] The signal sharpness and the root mean square average value of each group of monitoring vibration data form a monitoring array to obtain K monitoring arrays, and the K monitoring arrays are used as the K monitoring characteristic values.

[0018] In one embodiment, the step of calculating the signal sharpness of each set of sample vibration data comprises:

[0019] Calculate the peak factor of each group of sample vibration data, and use the peak factor as the signal sharpness of each group of sample vibration data; wherein the peak factor is the ratio between the maximum value of the data and the average value of the root mean square value;

[0020] The step of calculating the signal sharpness of each set of monitored vibration data comprises:

[0021] The peak factor of each set of monitoring vibration data is calculated, and the peak factor is used as the signal sharpness of each set of monitoring vibration data.

[0022] In one embodiment, clustering the K sample feature values ​​to obtain N first clusters includes:

[0023] Randomly selecting N sample feature values ​​from the K sample feature values ​​as current first cluster centers;

[0024] Calculate the distance between each sample feature value and each current first cluster center respectively, and divide the K sample feature values ​​into clusters to which the corresponding first cluster centers belong based on the minimum distance principle to obtain N clusters;

[0025] In each cluster, the total distance between each sample feature value and the remaining sample feature values ​​is calculated respectively, and the target sample feature value is updated as the current first cluster center; wherein the target sample feature value is the sample feature value with the smallest total distance in the corresponding cluster;

[0026] If any updated first cluster center is inconsistent with the first cluster center before updating, return to execute the step of respectively calculating the distance between each sample feature value and each current first cluster center and subsequent steps until all updated first cluster centers are consistent with the first cluster center before updating to obtain N first cluster clusters.

[0027] In one embodiment, clustering the K monitoring feature values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters includes:

[0028] Taking the cluster centers of the N first clusters under the target operation mode as the current second cluster centers;

[0029] The distance between each monitoring characteristic value and each current second cluster center is calculated respectively, and the K monitoring characteristic values ​​are divided into clusters to which the corresponding second cluster centers belong based on the minimum distance principle to obtain N second clusters.

[0030] In one embodiment, the step of drawing corresponding reference graphs based on the N first clusters respectively includes:

[0031] Draw a first histogram, and use the first histogram as the reference graph; wherein, in the first histogram, the cluster center of each first cluster is the horizontal axis value of a columnar bar, and the number of samples of each first cluster corresponds to the vertical axis value of the columnar bar;

[0032] The drawing of a monitoring diagram based on the N second clusters includes:

[0033] Draw a second histogram, and use the second histogram as the monitoring chart; wherein, in the second histogram, the cluster center of each second cluster is the horizontal axis value of a columnar bar, and the number of samples of each second cluster corresponds to the vertical axis value of the columnar bar.

[0034] In one embodiment, the step of drawing corresponding reference graphs based on the N first clusters respectively includes:

[0035] Draw a first line graph, and use the first line graph as the reference graph; wherein, in the first line graph, the cluster center of each first cluster is the horizontal axis value of a data point in the line, and the number of samples of each first cluster corresponds to the vertical axis value of the data point;

[0036] The drawing of a monitoring diagram based on the N second clusters includes:

[0037] Draw a second line graph and use the second line graph as the monitoring graph; wherein, in the second line graph, the cluster center of each second cluster is the horizontal axis value of a data point in the line graph, and the number of samples of each second cluster corresponds to the vertical axis value of the data point.

[0038] A winding variation degree monitoring device, the winding variation degree monitoring device comprising:

[0039] A reference graph drawing module is used to obtain K groups of sample vibration data under each operation mode preset by the transformer, wherein K≥2; under each operation mode, the characteristic value of each group of sample vibration data is calculated to obtain K sample characteristic values, and the K sample characteristic values ​​are clustered to obtain N first clustering clusters; wherein N≥2; corresponding reference graphs are drawn based on the N first clustering clusters to obtain multiple reference graphs under the multiple operation modes; wherein, in the reference graph, the cluster centers and sample quantities of the N first clustering clusters are used as coordinate axis values;

[0040] A monitoring diagram drawing module, used for obtaining K groups of monitoring vibration data under a target operation mode; wherein the target operation mode is any one of the multiple operation modes; calculating characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values, and clustering the K monitoring characteristic values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters; drawing a monitoring diagram based on the N second clusters; wherein in the monitoring diagram, the cluster centers of the N second clusters and the number of samples are used as coordinate axis values;

[0041] The winding change degree determination module is used to determine the winding change degree according to the overlap degree between the reference diagram under the target operation mode and the monitoring diagram; wherein the overlap degree is negatively correlated with the winding change degree.

[0042] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the steps of the above-mentioned winding variation degree monitoring method.

[0043] A winding variation degree monitoring device comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the winding variation degree monitoring method.

[0044] The present invention provides a method, device, medium and equipment for monitoring the degree of winding change. Under different operating modes of the transformer, K groups of sample vibration data of each mode are first obtained, and the characteristic values ​​are calculated, and multiple reference graphs of each operating mode are obtained through cluster analysis. Under the actual target operating mode, K groups of monitoring vibration data are obtained, their characteristic values ​​are calculated, and re-clustering is performed based on the cluster center of the reference graph to obtain a monitoring graph. By comparing the degree of overlap between the reference graph and the monitoring graph, the degree of change of the winding can be determined, wherein the higher the degree of overlap, the smaller the winding change. This method can accurately monitor the health status of the transformer under different operating conditions by analyzing the vibration data in detail and performing cluster comparison. It can timely identify the deterioration of the winding, reduce potential equipment failures, and improve the operating safety and reliability of the transformer. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.

[0046] in:

[0047] Figure 1 It is a schematic flow chart of a method for monitoring winding variation degree in one embodiment;

[0048] Figure 2 It is a structural schematic diagram of a winding variation degree monitoring device;

[0049] Figure 3 This is the structural block diagram of the winding variation degree monitoring device. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0052] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0053] like Figure 1 As shown, Figure 1 : is a flow chart of a method for monitoring winding variation degree in one embodiment. The steps provided by the method for monitoring winding variation degree in this embodiment include:

[0054] S101, under multiple operating modes preset for the transformer, respectively obtain K groups of sample vibration data under each operating mode.

[0055] Among them, the multiple operating modes preset by the transformer refer to the multiple different operating states or modes set for different load conditions, ambient temperatures or other operating parameters during the design and operation of the transformer. These operating modes may include different load rates (such as light load, medium load, full load), different cooling methods (such as natural cooling, forced cooling), or seasonal adjustments (such as winter mode, summer mode), etc. K≥2, these sample vibration data are used to reflect the vibration conditions of the transformer in a specific time period

[0056] Optionally, we set the transformer to operate in three modes: full load (100% load), medium load (50% load), and light load (20% load). In each operating mode, we collected 24 hours of vibration data and divided it into units of 10 seconds, and a total of 8640 sets of sample vibration data were obtained.

[0057] S102 , in each operation mode, respectively calculating the characteristic value of each group of sample vibration data to obtain K sample characteristic values, and clustering the K sample characteristic values ​​to obtain N first clusters.

[0058] The characteristic values ​​of the vibration data may include a root mean square value, a peak value, a spectrum characteristic, etc. Here, N≥2, and each first cluster represents a set of sample vibration data with similar vibration characteristics.

[0059] Optionally, in each operation mode, the vibration signal is subjected to spectrum analysis to extract spectrum features such as characteristic values ​​of main frequency, harmonics, etc. These spectrum features are then used for cluster analysis (such as k-means algorithm) to form N first clusters.

[0060] S103: Draw corresponding reference graphs based on the N first clusters respectively to obtain multiple reference graphs under multiple operation modes.

[0061] The reference diagram shows the cluster distribution of sample vibration data under different operation modes. In the reference diagram of this embodiment, the cluster centers and sample quantities of N first clusters are used as coordinate axis values.

[0062] Optionally, the reference graph may be in the form of a two-dimensional scatter plot, a histogram, etc., which may be set according to actual needs.

[0063] S104, obtaining K groups of monitoring vibration data under the target operation mode.

[0064] The target operation mode refers to the specific operation state or mode of the transformer in actual operation, and the target operation mode can only be any one of the multiple operation modes set in S101. In this step, K groups of data are also obtained, and the unit time length and sampling frequency of the K groups of monitoring vibration data are preferably consistent with the sample vibration data.

[0065] Optionally, under full load operation, vibration data is also collected continuously for 24 hours, with each set of data covering a time period of 10 seconds, and a total of 8640 sets of monitoring vibration data can be obtained.

[0066] S105 , calculating characteristic values ​​of K groups of monitoring vibration data to obtain K monitoring characteristic values, and clustering the K monitoring characteristic values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters.

[0067] The principle of this step is basically the same as S102, but the difference is that here the cluster centers of the N first clusters are used as references and combined with the cluster analysis of the current monitoring characteristic values ​​to obtain N second clusters, which facilitates the subsequent comparison of the matching degree between the actual operating status and the historical data.

[0068] Optionally, in the target operation mode, the root mean square value and peak value of K groups of monitoring data are calculated to obtain K monitoring characteristic values. These characteristic values ​​are subjected to k-means cluster analysis with the centers of the aforementioned N first clusters to obtain N second clusters.

[0069] S106: Draw a monitoring map based on the N second clusters.

[0070] The monitoring diagram shows the cluster distribution of sample vibration data under the target operation mode. In the monitoring diagram, the cluster centers of the N second clusters and the number of samples are used as coordinate axis values.

[0071] It can be understood that the monitoring map drawn in this step should be consistent with the type of the reference map drawn in S103.

[0072] S107, determining the degree of change of the winding according to the degree of overlap between the reference diagram and the monitoring diagram under the target operation mode.

[0073] The overlap degree is used to measure the similarity between the reference diagram and the monitoring diagram. In this embodiment, the overlap degree is set to be negatively correlated with the winding change degree.

[0074] Optionally, the overlapping area of ​​the two figures is used as the degree of overlap, and the degree of change of the winding is determined as follows: if the overlapping area accounts for more than 90% of the area of ​​the reference figure, it is considered that the winding state has not changed; if the overlapping area accounts for 70-90% of the area of ​​the reference figure, it is considered that the winding state has changed slightly, and further inspection is required in conjunction with the next transformer power outage; if the overlapping area accounts for less than 70% of the area of ​​the reference figure, it is considered that the winding state has changed significantly, and the transformer needs to be immediately powered off for further inspection.

[0075] The above-mentioned method for monitoring the degree of winding change, under different operating modes of the transformer, first obtains K groups of sample vibration data for each mode, calculates the characteristic values, and obtains multiple reference graphs for each operating mode through cluster analysis. Under the actual target operating mode, K groups of monitoring vibration data are obtained, their characteristic values ​​are calculated, and re-clustering is performed based on the cluster center of the reference graph to obtain a monitoring graph. By comparing the degree of overlap between the reference graph and the monitoring graph, the degree of change of the winding can be determined, where the higher the degree of overlap, the smaller the winding change. This method can accurately monitor the health status of the transformer under different operating conditions by analyzing the vibration data in detail and performing cluster comparison. It can timely identify the deterioration of the winding, reduce potential equipment failures, and improve the operating safety and reliability of the transformer.

[0076] In a specific embodiment, calculating the characteristic value of each group of sample vibration data in S102 to obtain K sample characteristic values ​​specifically includes the following steps:

[0077] A1. Calculate the signal sharpness of each group of sample vibration data respectively, and calculate the root mean square average value of each group of sample vibration data to obtain K signal sharpness and root mean square average values.

[0078] Among them, signal sharpness is used to indicate the sharpness of the signal, which reflects the transient characteristics of the signal.

[0079] Optionally, the ratio of the maximum value to the minimum value of the sample vibration data is taken as the signal sharpness.

[0080] Optionally, the peak factor of each group of sample vibration data is calculated, and the peak factor is used as the signal sharpness of each group of sample vibration data; wherein the peak factor is the ratio between the maximum value of the data and the average value of the root mean square value, and the calculation formula of the kth group peak factor is:

[0081]

[0082] where v kmax Refers to the kth group of vibration data v k The maximum value of v kRms Refers to the first set of vibration data v k The average root mean square value of .

[0083] A2. The signal sharpness and the root mean square average value of each group of sample vibration data form a sample array to obtain K sample arrays, and the K sample arrays are used as K sample eigenvalues.

[0084] That is to form the following sample array Q and use it as the sample eigenvalue:

[0085] Q={q1,q2,···,q k}={(p1,v 1Rms ),(p2,v 2Rms ),···,(p k , v kRms )}.

[0086] Correspondingly, calculating the characteristic values ​​of K groups of monitoring vibration data in S105 to obtain K monitoring characteristic values ​​specifically includes the following steps:

[0087] A3. Calculate the signal sharpness of each group of monitoring vibration data respectively, and calculate the root mean square average value of each group of monitoring vibration data to obtain K signal sharpness and root mean square average values.

[0088] Optionally, calculating the signal sharpness of each set of monitoring vibration data includes: calculating a peak factor of each set of monitoring vibration data, and using the peak factor as the signal sharpness of each set of monitoring vibration data.

[0089] A4. The signal sharpness and the root mean square average value of each group of monitoring vibration data form a monitoring array to obtain K monitoring arrays, and the K monitoring arrays are used as K monitoring characteristic values.

[0090] That is to form the following monitoring array Q and use it as the monitoring characteristic value:

[0091] Q 监测 = {q 1监测 ,q 2监测 ,···,q k监测}={(p 1监测 , v 1监测Rms ), (p 2监测 , v 2监测Rms ), ···,

[0092] (p k监测 , v k监测Rms )}.

[0093] The principles of the above steps A3 and A4 are basically the same as those of steps A1 and A2. The only difference is that the processing object is changed from K groups of sample vibration data to monitoring vibration data, so they will not be described in detail.

[0094] In a specific embodiment, clustering the K sample feature values ​​in S102 to obtain N first clusters specifically includes the following steps:

[0095] B1. Randomly select N sample feature values ​​from the K sample feature values ​​as the current first cluster center.

[0096] That is, N samples are randomly selected from the array Q as the current first clustering center c1, c2, ···, c N The clustering result at this time is C k ={c1, c2, ···, c N}, that is, each cluster contains only the cluster center.

[0097] B2. Calculate the distance between each sample feature value and each current first cluster center respectively, and divide the K sample feature values ​​into the clusters to which the corresponding first cluster centers belong based on the minimum distance principle to obtain N clusters.

[0098] Here, Euclidean distance, Manhattan distance, etc. can be used as the distance calculation method. Optionally, the distance calculation formula is:

[0099] d ik =|q i -c k |

[0100] Then sort the distances, and according to the principle of the smallest distance from the cluster center, put qi into the clusters to which the corresponding first cluster center belongs to form a new clustering result.

[0101] B3. In each cluster, the total distance between each sample feature value and the remaining sample feature values ​​is calculated respectively, and the target sample feature value is updated to the current first cluster center.

[0102] Among them, the target sample feature value is the sample feature value with the smallest total distance in the corresponding cluster.

[0103] That is to say, in the new clustering results Calculate the new cluster centers Update, the calculation method is in each cluster, each sample q ik The sum of the distances to other samples in the same cluster D ik , assuming that there are Ik samples in a cluster, then the first sample q 1k The sum of the distances to other samples D 1k The calculation is as follows:

[0104]

[0105] Calculate D ik Then, take D ik The smallest sample is updated as the current first cluster center

[0106] In this way, the updated N clusters can be obtained And the current first cluster center

[0107] B4. If any updated first cluster center is inconsistent with the first cluster center before updating, return to execute the step of calculating the distance between each sample feature value and each current first cluster center and subsequent steps, until all updated first cluster centers are consistent with the first cluster center before updating, and N first cluster clusters are obtained.

[0108] That is, when the cluster center no longer changes, the final clustering result C is obtained. fink ={C fin1 , C fin2 ,···,C finN}, and the corresponding final cluster center c fink ={c fin1 , c fin2 ,···,c finN}.

[0109] In a specific embodiment, clustering the K monitoring feature values ​​based on the cluster centers of the N first clusters under the target operation mode in S105 to obtain N second clusters specifically includes the following steps:

[0110] C1. The cluster centers of the N first clusters under the target operation mode are used as the current second cluster centers.

[0111] That is to say, in the target operation mode, it is not necessary to redetermine the cluster center from the beginning. The cluster centers of the N first clusters in the target operation mode previously calculated in the reference data can be directly used as the second cluster center of the current monitoring data.

[0112] Doing so can reduce computational complexity and maintain consistency, so that the clustering process of the monitoring data directly corresponds to the clustering process of the reference data, making it easier to compare and analyze.

[0113] C2. Calculate the distance between each monitoring feature value and each current second cluster center respectively, and divide the K monitoring feature values ​​into the clusters to which the corresponding second cluster centers belong based on the minimum distance principle to obtain N second clusters.

[0114] Optionally, the distance calculation formula is:

[0115]

[0116] Then, each monitoring characteristic value is divided into the cluster to which the second cluster center closest to it belongs. In this way, K monitoring characteristic values ​​are divided into N second clusters.

[0117] In this specific embodiment, since the cluster centers are derived from known reference data, the clustering results of the monitoring data are more reliable and referenceable, which helps to accurately evaluate the state changes of the equipment under the target operation mode.

[0118] In a specific embodiment, S103 draws corresponding reference graphs based on N first clusters respectively, including: drawing a first histogram and using the first histogram as a reference graph; wherein, in the first histogram, the cluster center of each first cluster is the horizontal axis value of a bar, and the number of samples of each first cluster corresponds to the vertical axis value of the bar.

[0119] Correspondingly, drawing a monitoring chart based on N second clusters in S106 includes: drawing a second histogram and using the second histogram as the monitoring chart; wherein, in the second histogram, the cluster center of each second cluster is the horizontal axis value of a column bar, and the number of samples of each second cluster corresponds to the vertical axis value of the column bar.

[0120] Then, when calculating the overlap degree in S107, the bars can be matched in pairs based on the horizontal axis value, and then the area overlap degree is calculated according to the height ratio between the two bars, so as to obtain the overlap degree between the reference graph and the monitoring graph, and then determine the degree of change of the winding.

[0121] In this way, by drawing the first histogram and the second histogram, the difference between the reference state and the actual operating state can be intuitively compared.

[0122] In a specific embodiment, drawing corresponding reference graphs based on the N first clusters in S103 includes: drawing a first line graph and using the first line graph as a reference graph; wherein, in the first line graph, the cluster center of each first cluster is the horizontal axis value of a data point in the line graph, and the number of samples of each first cluster corresponds to the vertical axis value of the data point;

[0123] Correspondingly, drawing a monitoring chart based on N second clusters in S106 includes: drawing a second line chart, and using the second line chart as a monitoring chart; wherein, in the second line chart, the cluster center of each second cluster is the horizontal axis value of a data point in the line, and the number of samples in each second cluster corresponds to the vertical axis value of the data point.

[0124] Then, when calculating the overlap degree in S107, the coverage area between the broken line and the two coordinate axes can be used as the area for calculating the overlap degree, and then the ratio of the two coverage areas can be calculated to obtain the overlap degree between the reference map and the monitoring map, thereby determining the degree of change of the winding. Alternatively, the similarity between the two broken lines can be calculated, and the similarity can be used as the overlap degree between the reference map and the monitoring map, thereby determining the degree of change of the winding.

[0125] In this way, by drawing the first line graph and the second line graph, the difference between the reference state and the actual operating state can be intuitively compared.

[0126] In one embodiment, Figure 2 As shown, a winding variation degree monitoring device is proposed, the device comprising:

[0127] The reference graph drawing module 201 is used to obtain K groups of sample vibration data under each operation mode under multiple operation modes preset by the transformer, wherein K≥2; under each operation mode, the characteristic value of each group of sample vibration data is calculated to obtain K sample characteristic values, and the K sample characteristic values ​​are clustered to obtain N first clustering clusters; wherein N≥2; corresponding reference graphs are drawn based on the N first clustering clusters to obtain multiple reference graphs under multiple operation modes; wherein, in the reference graph, the cluster centers and sample quantities of the N first clustering clusters are used as coordinate axis values;

[0128] The monitoring diagram drawing module 202 is used to obtain K groups of monitoring vibration data under a target operation mode; wherein the target operation mode is any one of a plurality of operation modes; calculate the characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values, and cluster the K monitoring characteristic values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters; draw a monitoring diagram based on the N second clusters; wherein in the monitoring diagram, the cluster centers of the N second clusters and the number of samples are used as coordinate axis values;

[0129] The winding change degree determination module 203 is used to determine the winding change degree according to the overlap degree between the reference diagram and the monitoring diagram under the target operation mode; wherein the overlap degree is negatively correlated with the winding change degree.

[0130] Figure 3 FIG. 2 shows an internal structure diagram of a winding variation degree monitoring device in one embodiment. Figure 3 As shown, the winding variation degree monitoring device includes a processor, a memory and a network interface connected via a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the winding variation degree monitoring device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the winding variation degree monitoring method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the winding variation degree monitoring method. Those skilled in the art can understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the winding variation degree monitoring device to which the scheme of the present application is applied. The specific winding variation degree monitoring device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0131] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented: in a plurality of operation modes preset by a transformer, K groups of sample vibration data in each operation mode are obtained respectively; wherein K≥2; in each operation mode, characteristic values ​​of each group of sample vibration data are calculated respectively to obtain K sample characteristic values, and the K sample characteristic values ​​are clustered to obtain N first clustering clusters; wherein N≥2; corresponding reference graphs are drawn based on the N first clustering clusters respectively to obtain multiple reference graphs in a plurality of operation modes; wherein, in the reference graphs, the cluster centers and sample values ​​of the N first clustering clusters are used as the reference graphs. This quantity is the coordinate axis value; under the target operating mode, K groups of monitoring vibration data are obtained; wherein the target operating mode is any one of multiple operating modes; the characteristic values ​​of the K groups of monitoring vibration data are calculated to obtain K monitoring characteristic values, and the K monitoring characteristic values ​​are clustered based on the cluster centers of the N first clusters under the target operating mode to obtain N second clusters; a monitoring diagram is drawn based on the N second clusters; wherein, in the monitoring diagram, the cluster centers and sample quantities of the N second clusters are used as coordinate axis values; the degree of change of the winding is determined according to the degree of overlap between the reference diagram and the monitoring diagram under the target operating mode; wherein the degree of overlap is negatively correlated with the degree of change of the winding.

[0132] A winding variation degree monitoring device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: under multiple operation modes preset by the transformer, respectively obtain K groups of sample vibration data under each operation mode; wherein K≥2; under each operation mode, respectively calculate the characteristic value of each group of sample vibration data to obtain K sample characteristic values, and cluster the K sample characteristic values ​​to obtain N first clustering clusters; wherein N≥2; respectively draw corresponding reference graphs based on the N first clustering clusters to obtain multiple reference graphs under multiple operation modes; wherein, in the reference graph, the N first clusters are used as the reference graphs; The cluster centers and sample numbers of the clusters are coordinate axis values; under the target operating mode, K groups of monitoring vibration data are obtained; wherein the target operating mode is any one of a plurality of operating modes; characteristic values ​​of the K groups of monitoring vibration data are calculated to obtain K monitoring characteristic values, and the K monitoring characteristic values ​​are clustered based on the cluster centers of the N first clusters under the target operating mode to obtain N second clusters; a monitoring diagram is drawn based on the N second clusters; wherein, in the monitoring diagram, the cluster centers and sample numbers of the N second clusters are used as coordinate axis values; the degree of change of the winding is determined according to the degree of overlap between the reference diagram and the monitoring diagram under the target operating mode; wherein the degree of overlap is negatively correlated with the degree of change of the winding.

[0133] It should be noted that the above-mentioned winding variation degree monitoring method, device, equipment and computer-readable storage medium belong to a general inventive concept, and the contents in the embodiments of the winding variation degree monitoring method, device, equipment and computer-readable storage medium are applicable to each other.

[0134] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0135] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for monitoring winding variation, characterized in that: The method comprises: Under various operation modes preset by the transformer, K groups of sample vibration data under each operation mode are obtained respectively; wherein K≥2; In each operation mode, the characteristic value of each group of sample vibration data is calculated respectively to obtain K sample characteristic values, and the K sample characteristic values ​​are clustered to obtain N first clustering clusters; wherein N≥2; Draw corresponding reference graphs based on the N first clusters respectively to obtain multiple reference graphs under the multiple operation modes; wherein, in the reference graphs, the cluster centers and sample quantities of the N first clusters are used as coordinate axis values; Under the target operation mode, K groups of monitoring vibration data are obtained; wherein the target operation mode is any one of the multiple operation modes; Calculating the characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values, and clustering the K monitoring characteristic values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters; Drawing a monitoring graph based on the N second clusters; wherein, in the monitoring graph, the cluster centers and sample quantities of the N second clusters are used as coordinate axis values; The degree of change of the winding is determined according to the degree of overlap between the reference diagram and the monitoring diagram under the target operation mode; wherein the degree of overlap is negatively correlated with the degree of change of the winding.

2. The method according to claim 1, characterized in that The eigenvalues ​​of each group of sample vibration data are calculated respectively to obtain K sample eigenvalues, including: Calculate the signal sharpness of each group of sample vibration data respectively, and calculate the root mean square average value of each group of sample vibration data to obtain K signal sharpness and root mean square average values; wherein the signal sharpness is used to indicate the sharpness of the signal; The signal sharpness and the root mean square average value of each group of sample vibration data form a sample array to obtain K sample arrays, and the K sample arrays are used as the K sample feature values; The calculating the characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values ​​includes: Calculate the signal sharpness of each group of monitoring vibration data respectively, and calculate the root mean square average value of each group of monitoring vibration data to obtain K signal sharpness and root mean square average values; The signal sharpness and the root mean square average value of each group of monitoring vibration data form a monitoring array to obtain K monitoring arrays, and the K monitoring arrays are used as the K monitoring characteristic values.

3. The method according to claim 2, characterized in that The step of calculating the signal sharpness of each group of sample vibration data comprises: Calculate the peak factor of each group of sample vibration data, and use the peak factor as the signal sharpness of each group of sample vibration data; wherein the peak factor is the ratio between the maximum value of the data and the average value of the root mean square value; The step of calculating the signal sharpness of each set of monitored vibration data comprises: The peak factor of each set of monitoring vibration data is calculated, and the peak factor is used as the signal sharpness of each set of monitoring vibration data.

4. The method according to claim 1, characterized in that: The clustering of the K sample feature values ​​to obtain N first clusters includes: Randomly selecting N sample feature values ​​from the K sample feature values ​​as current first cluster centers; Calculate the distance between each sample feature value and each current first cluster center respectively, and divide the K sample feature values ​​into clusters to which the corresponding first cluster centers belong based on the minimum distance principle to obtain N clusters; In each cluster, the total distance between each sample feature value and the remaining sample feature values ​​is calculated respectively, and the target sample feature value is updated as the current first cluster center; wherein the target sample feature value is the sample feature value with the smallest total distance in the corresponding cluster; If any updated first cluster center is inconsistent with the first cluster center before updating, return to execute the step of respectively calculating the distance between each sample feature value and each current first cluster center and subsequent steps until all updated first cluster centers are consistent with the first cluster center before updating to obtain N first cluster clusters.

5. The method according to claim 1, characterized in that The clustering of the K monitoring feature values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters includes: Taking the cluster centers of the N first clusters under the target operation mode as the current second cluster centers; The distance between each monitoring characteristic value and each current second cluster center is calculated respectively, and the K monitoring characteristic values ​​are divided into clusters to which the corresponding second cluster centers belong based on the minimum distance principle to obtain N second clusters.

6. The method according to claim 1, characterized in that The drawing of corresponding reference graphs based on the N first clusters respectively includes: Draw a first histogram, and use the first histogram as the reference graph; wherein, in the first histogram, the cluster center of each first cluster is the horizontal axis value of a columnar bar, and the number of samples of each first cluster corresponds to the vertical axis value of the columnar bar; The drawing of a monitoring diagram based on the N second clusters includes: Draw a second histogram, and use the second histogram as the monitoring chart; wherein, in the second histogram, the cluster center of each second cluster is the horizontal axis value of a columnar bar, and the number of samples of each second cluster corresponds to the vertical axis value of the columnar bar.

7. The method according to claim 1, characterized in that The drawing of corresponding reference graphs based on the N first clusters respectively includes: Draw a first line graph, and use the first line graph as the reference graph; wherein, in the first line graph, the cluster center of each first cluster is the horizontal axis value of a data point in the line, and the number of samples of each first cluster corresponds to the vertical axis value of the data point; The drawing of a monitoring diagram based on the N second clusters includes: Draw a second line graph and use the second line graph as the monitoring graph; wherein, in the second line graph, the cluster center of each second cluster is the horizontal axis value of a data point in the line graph, and the number of samples of each second cluster corresponds to the vertical axis value of the data point.

8. A winding variation degree monitoring device, characterized in that: The winding variation degree monitoring device comprises: A reference graph drawing module is used to obtain K groups of sample vibration data under each operation mode preset by the transformer, wherein K≥2; under each operation mode, the characteristic value of each group of sample vibration data is calculated to obtain K sample characteristic values, and the K sample characteristic values ​​are clustered to obtain N first clustering clusters; wherein N≥2; corresponding reference graphs are drawn based on the N first clustering clusters to obtain multiple reference graphs under the multiple operation modes; wherein, in the reference graph, the cluster centers and sample quantities of the N first clustering clusters are used as coordinate axis values; A monitoring diagram drawing module, used for obtaining K groups of monitoring vibration data under a target operation mode; wherein the target operation mode is any one of the multiple operation modes; calculating characteristic values ​​of the K groups of monitoring vibration data to obtain K monitoring characteristic values, and clustering the K monitoring characteristic values ​​based on the cluster centers of the N first clusters under the target operation mode to obtain N second clusters; drawing a monitoring diagram based on the N second clusters; wherein in the monitoring diagram, the cluster centers of the N second clusters and the number of samples are used as coordinate axis values; The winding change degree determination module is used to determine the winding change degree according to the overlap degree between the reference diagram under the target operation mode and the monitoring diagram; wherein the overlap degree is negatively correlated with the winding change degree.

9. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

10. A winding variation degree monitoring device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 7.