A method for correlation analysis of vibration envelope of on-load tap changer
By performing segmentation processing of the envelope curve and calculation of the overlapping degree of vibration signal of the on-load tap-off switch, the problem of peak segment dislocation of vibration signal switching is solved, the accuracy and reliability of fault diagnosis is improved, and the vibration envelope is realized Accuracy and reliability of correlation abnormality diagnosis.
Patent Information
- Application Number
- CN202210141871.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-16
- Publication Date
- 2025-01-14
- Estimated Expiration
- 2042-02-16
AI Technical Summary
In the prior art, the time domain envelope correlation diagnosis method based on vibration signals has a vibration signal shift peak segment dislocation phenomenon in the on-load tap-off fault diagnosis, resulting in deviation of calculation results and affecting the accuracy of diagnostic analysis.
By segmenting the standard envelope curve and the measured envelope curve, select the starting point of the first switching segment as the reference point, calculate the overlap degree step by step, adjust the shift of the measured envelope curve, and make it coincide with the standard envelope curve. , calculate the overlap as the alarm coefficient for abnormal diagnosis and analysis, and use the mutual correlation coefficient algorithm to calculate the overlap.
It realizes automatic alignment of vibration signals, corrects the switching peak segment dislocation phenomenon, improves the accuracy and reliability of fault diagnosis, and provides accurate data guarantee.
Smart Images

Figure CN114547550B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of power equipment monitoring, and in particular to a vibration envelope correlation analysis method for an on-load tap changer. Background Art
[0002] On-load tap changer (OLTC) is a key component of on-load voltage regulating transformer, and plays an important role in stabilizing grid voltage, regulating reactive power flow and increasing grid flexibility in the power system. Mechanical failure of on-load tap changer often causes serious transformer accidents and power outages, resulting in significant social and economic losses. Therefore, the detection and diagnosis of on-load tap changer faults is of great significance to ensure grid safety and power supply reliability.
[0003] Time domain envelope correlation diagnosis based on vibration signals is an important method for OLTC fault diagnosis. However, the sampling starting point of the current and vibration data of the on-load tap changer is realized by the current threshold trigger method. This trigger method has slight differences due to the random waveform of the current. This slight difference will not cause problems when judging the overall operating status of the on-load tap changer. However, when applying the vibration signal envelope correlation method to analyze the vibration signals of different actions, it will cause the vibration signal switching peak segment to be misplaced, resulting in obvious deviations in the calculation results, which ultimately affects the diagnostic analysis results. Summary of the invention
[0004] The invention provides a vibration envelope correlation analysis method for an on-load tap changer, the main purpose of which is to solve the problems existing in the prior art.
[0005] The present invention adopts the following technical solution:
[0006] A method for analyzing the correlation of vibration envelope of an on-load tap changer comprises the following steps:
[0007] S1. Segment processing is performed on a preset standard envelope curve and a measured envelope curve obtained based on real-time monitoring, so as to divide the vibration signal in the OLTC switch switching process into an energy storage segment and a switching segment with nonlinear characteristics;
[0008] S2, selecting the starting point of the first switching segment in the standard envelope curve and the measured envelope curve as the reference point, shifting the measured envelope curve after the reference point in a set shift interval according to a set step length, and calculating the coincidence degree between the measured envelope curve after the reference point and the standard envelope curve after the reference point;
[0009] S3, recording the coordinates of the reference point of the measured envelope curve when the overlap is maximum, and shifting the entire measured envelope curve so that the coordinates of the starting point of the first switching segment coincide with the coordinates of the reference point;
[0010] S4. Calculate the degree of overlap between the overall measured envelope curve after displacement and the overall standard envelope curve, and use the degree of overlap as an alarm coefficient to perform vibration envelope correlation abnormality diagnosis and analysis.
[0011] Furthermore, in steps S2 and S4, the calculation method of the overlap degree adopts the mutual correlation coefficient algorithm, with the standard sample data as the array x and the measured data as the array y, and the calculation formula of the mutual correlation coefficient is:
[0012]
[0013] Where: r xy Represents the sample correlation coefficient, that is, the alarm coefficient; S xy represents the sample covariance; S x Represents the sample standard deviation of array x; S y Represents the sample standard deviation of array y.
[0014] Furthermore, in formula (1), the sample covariance S xy The calculation formula is:
[0015]
[0016] Sample standard deviation S x The calculation formula is:
[0017]
[0018] Sample standard deviation S y The calculation formula is:
[0019]
[0020] Where: and Respectively represent the average values of arrays x and y, and n represents the number of data points in the waveform.
[0021] Further, in step S2, the shift interval is set to -0.1*saRate to 0.1*saRate, where saRate is the system sampling rate.
[0022] Further, in step S2, the fixed shift step is set to 50 data points.
[0023] Further, in step S1, vibration signals when the OLTC switch is switched are collected through several channels, and the measured envelope curve of each channel is plotted; the measured envelope curve of each channel is processed in steps S2, S3 and S4 to obtain the alarm coefficient of each channel.
[0024] Furthermore, let the number of channels be j, and the alarm coefficient of each channel be [V1, V2, ..., V j ]; then the vibration envelope correlation abnormality diagnosis method is: if the alarm coefficients of each channel [V1, V2, ..., V j ] is less than 0.7, or the alarm coefficient of any single channel is less than 0.5, the vibration envelope correlation diagnosis is abnormal.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] Based on the formation characteristics of vibration signals, the invention takes the standard envelope curve as a reference, shifts the measured envelope curve, and determines the shift distance by calculating the overlap between the measured envelope curve and the standard envelope curve, thereby realizing automatic alignment of the OLTC vibration envelope curve, thereby correcting the phenomenon of misalignment of the switching peak segment of the vibration signal, providing accurate data guarantee for the abnormal diagnosis and analysis of the vibration envelope correlation, fundamentally improving the accuracy and reliability of the OLTC fault diagnosis results, and having the advantages of simple algorithm and strong versatility. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a schematic diagram of the standard envelope curve before shifting and the measured envelope curve in the present invention.
[0028] Figure 2 It is a schematic diagram of the first switching section of the standard envelope curve and the measured envelope curve before shifting in the present invention.
[0029] Figure 3 It is a schematic diagram of the first switching section of the standard envelope curve and the measured envelope curve after shifting in the present invention.
[0030] Figure 4 It is a schematic diagram of the standard envelope curve and the measured envelope curve after shifting in the present invention. DETAILED DESCRIPTION
[0031] The specific implementation of the present invention is described below with reference to the accompanying drawings. In order to fully understand the present invention, many details are described below, but for those skilled in the art, the present invention can be implemented without these details.
[0032] Reference Figures 1 to 4 The present invention provides a method for analyzing the vibration envelope correlation of an on-load tap changer, comprising the following steps:
[0033] S1. The preset standard envelope curve and the measured envelope curve obtained based on real-time monitoring are processed in sections, so as to divide the vibration signal in the OLTC switch switching process into an energy storage section and a switching section with nonlinear characteristics.
[0034] In this step, the standard sample data of the standard envelope curve can be selected by the tester based on field data and experience. The measured data of the measured envelope curve comes from the vibration signal collected by the vibration acceleration sensor when the OLTC switch is switched. Specifically, the present invention collects the vibration signal of the vibration acceleration sensor through several channels, and calculates the real-time data of the vibration envelope correlation of each channel respectively, thereby drawing the measured envelope curve of each channel. Assume that the number of vibration acceleration sensors is j, that is, the number of channels is j, and j≥1, the installation location of the vibration acceleration sensor is determined according to actual needs, and is generally installed on the side wall, top and gear box of the OLTC.
[0035] The segmented processing method of the standard envelope curve and the measured envelope curve can be implemented by referring to the curve segmentation method of the vibration signal provided by the Chinese patent application number 201910823825.6, which will not be elaborated here.
[0036] S2. Select the starting point of the first switching segment in the standard envelope curve and the measured envelope curve as the reference point, shift the measured envelope curve after the reference point in a set shift interval according to the set step size, and calculate the degree of overlap between the measured envelope curve after the reference point and the standard envelope curve after the reference point.
[0037] Preferably, the shift interval set in this embodiment is -0.1*saRate to 0.1*saRate, where saRate is the system sampling rate; and the fixed shift step is set to 50 data points.
[0038] Preferably, in this step, the calculation method of the overlap degree adopts a mutual correlation coefficient algorithm, that is, the overlap degree is evaluated by evaluating the correlation between the measured data of the vibration envelope correlation and the standard sample data. Specifically, the standard sample data is used as the array x, and the measured data is used as the array y, then the calculation formula of the mutual correlation coefficient is:
[0039]
[0040] Where: r xy Represents the sample correlation coefficient, that is, the alarm coefficient; S xy represents the sample covariance; S x Represents the sample standard deviation of array x; S y Represents the sample standard deviation of array y.
[0041] In formula (1), the sample covariance S xy The calculation formula is:
[0042]
[0043] Sample standard deviation S x The calculation formula is:
[0044]
[0045] Sample standard deviation S y The calculation formula is:
[0046]
[0047] Where: and Respectively represent the average values of arrays x and y, and n represents the number of data points in the waveform.
[0048] S3. Record the coordinates of the reference point of the measured envelope curve when the overlap is the largest, and shift the entire measured envelope curve so that the coordinates of the starting point of the first switching segment coincide with the coordinates of the reference point. Figure 2 and 3 It can be seen that the OLTC vibration envelope curve can be automatically aligned through shifting, thereby correcting the phenomenon of misalignment of the vibration signal switching peak segment and providing accurate data guarantee for the next step of calculating the overall overlap.
[0049] S4. Calculate the degree of overlap between the overall measured envelope curve after displacement and the overall standard envelope curve, and use the degree of overlap as an alarm coefficient to perform vibration envelope correlation abnormality diagnosis and analysis.
[0050] The calculation method of the overlap degree in this process is the same as step S2, so it will not be repeated. However, it should be noted that when calculating the overlap degree, the length of the data should be kept consistent. The alarm coefficient of each channel obtained by the algorithm is recorded as [V1, V2, ..., V j ]; then the abnormal diagnosis method of vibration envelope correlation is: if the alarm coefficients of each channel [V1, V2, ..., V j ] is less than 0.7, or the alarm coefficient of any single channel is less than 0.5, the vibration envelope correlation diagnosis is abnormal. This diagnostic method can accurately and sensitively reflect the abnormal situation of the vibration signal, fully ensuring the accuracy and reliability of the vibration envelope correlation abnormality diagnosis analysis.
[0051] The above is only a specific implementation of the present invention, but the design concept of the present invention is not limited to this. Any non-substantial changes to the present invention using this concept shall be deemed as an infringement of the protection scope of the present invention.
Claims
1. A method for analyzing the vibration envelope correlation of an on-load tap changer, characterized in that: The steps include: S1. Segment processing is performed on a preset standard envelope curve and a measured envelope curve obtained based on real-time monitoring, so as to divide the vibration signal in the OLTC switch switching process into an energy storage segment and a switching segment with nonlinear characteristics; S2, selecting the starting point of the first switching segment in the standard envelope curve and the measured envelope curve as the reference point, shifting the measured envelope curve after the reference point in a set shift interval according to a set step length, and calculating the coincidence degree between the measured envelope curve after the reference point and the standard envelope curve after the reference point; S3, recording the coordinates of the reference point of the measured envelope curve when the overlap is maximum, and shifting the entire measured envelope curve so that the coordinates of the starting point of the first switching segment coincide with the coordinates of the reference point; S4. Calculate the degree of overlap between the overall measured envelope curve after displacement and the overall standard envelope curve, and use the degree of overlap as an alarm coefficient to perform vibration envelope correlation abnormality diagnosis and analysis.
2. The on-load tap changer vibration envelope correlation analysis method according to claim 1, characterized in that: In steps S2 and S4, the calculation method of the overlap degree adopts the mutual correlation coefficient algorithm, with the standard sample data as the array x and the measured data as the array y. The calculation formula of the mutual correlation coefficient is: Where: r xy Represents the sample correlation coefficient, i.e., the alarm coefficient; S xy represents the sample covariance; S x Represents the sample standard deviation of array x; S y Represents the sample standard deviation of array y.
3. The on-load tap changer vibration envelope correlation analysis method according to claim 2, characterized in that: In formula (1), the sample covariance S xy The calculation formula is: Sample standard deviation S x The calculation formula is: Sample standard deviation S y The calculation formula is: Where: and Respectively represent the average values of arrays x and y, and n represents the number of data points in the waveform.
4. The on-load tap changer vibration envelope correlation analysis method according to claim 1, characterized in that: In step S2, the shift interval is set to -0.1*saRate to 0.1*saRate. saRate is the system sampling rate.
5. The on-load tap changer vibration envelope correlation analysis method according to claim 1, characterized in that: In step S2, the fixed shift step is set to 50 data points.
6. The on-load tap changer vibration envelope correlation analysis method according to claim 1, characterized in that: In step S1, vibration signals when the OLTC switch is switched are collected through several channels, and the measured envelope curve of each channel is plotted; the measured envelope curve of each channel is processed in steps S2, S3 and S4 to obtain the alarm coefficient of each channel.
7. The method for analyzing the vibration envelope correlation of an on-load tap changer according to claim 6, characterized in that: Assume that the number of channels is j, and the alarm coefficient of each channel is [V1, V2, ..., V j ]; then the vibration envelope correlation abnormality diagnosis method is: if the alarm coefficients of each channel [V1, V2, ..., V j ] is less than 0.7, or the alarm coefficient of any single channel is less than 0.5, the vibration envelope correlation diagnosis is abnormal.
Citation Information
Patent Citations
A method for analyzing and diagnosing mechanical faults in on-load tap changers
CN110658445B
Transformer on-load tap-changer fault diagnosis method based on vibration characteristics
CN102998618A
On-load tap-changer switching program online detection and diagnosis method
CN109507579A