High-voltage switch cabinet partial discharge identification method based on time domain signal parameter analysis
Through the analysis of time domain signal parameters, combined with the sliding observation window and filter threshold, the pulse time flag position, charge quantity and energy difference value are calculated, and a three-dimensional map is drawn, which solves the false alarm and missed alarm problems of local discharge detection in the existing technology, and realizes the More accurate local discharge pattern recognition and evaluation.
Patent Information
- Application Number
- CN202411419089.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-12
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-10-12
AI Technical Summary
In the local discharge detection of high-voltage switch cabinets, methods based on the threshold parameters of local discharge signal size are easily disturbed, false alarms or missed reports are severe, and the PRPD map does not consider the parameter relationship between continuous pulses, resulting in loss of characteristic information. It is difficult to accurately identify the discharge mode.
Using a method based on time domain signal parameter analysis, the UHF or TEV pulse time domain waveform data is collected by building a localized test circuit, the sliding observation window and filtering threshold are determined, the pulse time flag position, charge quantity and energy difference are calculated, and a three-dimensional map is drawn. , analyze the map features to diagnose and identify local discharge patterns.
This method fully considers the correlation between continuous pulses, and uses the three-dimensional map to diagnose and evaluate local discharge more carefully, improving the accuracy and applicability of the detection, and can more accurately reflect the changing information during the discharge process.
Smart Images

Figure CN118937934A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of partial discharge identification, and in particular to a method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis. Background Art
[0002] High-voltage switchgear is a device responsible for receiving and distributing electric energy in the power system. It is the "transfer station" of the power system and has been widely used in the transmission and distribution network. In actual operation, partial discharge will occur in the switchgear due to electrical, thermal, chemical and other factors, which will lead to a decrease in insulation strength. In the long run, it will damage the insulation state of the switchgear and eventually cause the switchgear to fail. Partial discharge is closely related to the insulation state of the switchgear. Accurately identifying the type of partial discharge is of great significance to ensure the reliable operation of the switchgear.
[0003] At present, in the actual on-site partial discharge detection of switch cabinets, ultra-high frequency, ultrasound, and transient ground voltage methods are all commonly used partial discharge detection methods. By setting the threshold parameters of the partial discharge signal size, it is possible to determine whether partial discharge of the switch cabinet has occurred and the severity of the partial discharge. In addition, the spectrum that characterizes the partial discharge characteristics can also be used to analyze the partial discharge signal. The partial discharge phase distribution (PRPD) spectrum is currently the most mainstream partial discharge analysis mode. This method can clearly reflect the distribution characteristics of parameters such as partial discharge pulse phase, amplitude, and discharge frequency, and extract the characteristic parameters of various defects to perform partial discharge pattern recognition. The use of PRPD spectrum for partial discharge detection and pattern recognition has the advantages of small data volume and good stability, and has been widely used in practice.
[0004] The method of identifying partial discharge based on setting the threshold parameters of the partial discharge signal size is simple, but it extracts too few partial discharge information features, which is prone to unreasonable threshold settings, and serious false alarms or omissions due to interference. The PRPD spectrum is a method of evaluating and detecting partial discharge by extracting many characteristic parameters. It is a characteristic spectrum based on the phase and amplitude of the discharge pulse, but it does not consider the parameter relationship between continuous pulses. It only focuses on the characteristics of the phase and pulse amplitude, and displays very little information about changes in the discharge process, so some features will be lost. In addition, considering the actual situation on site, in most cases, a stable phase reference signal is difficult to obtain. Even if it can be obtained, it is limited by the on-site operation mode and economy, resulting in unstable or drifting phases of partial discharge pulses in the PRPD spectrum, resulting in distortion of the PRPD spectrum characteristics.
[0005] In fact, by analyzing the relationship between continuous discharge pulses and fully mining the information in the discharge process, such as calculating the time interval, the change in the amplitude of the partial discharge pulse and other parameters, the partial discharge can be better evaluated and detected. The spectral characteristics of the change in the parameters between pulses can be used to improve the discharge mode corresponding to different defects. At the same time, it is possible to gain a deep understanding of the correlation between continuous partial discharge pulses and different partial discharge sources without accessing the phase reference signal. Summary of the invention
[0006] In order to make up for the deficiencies in the prior art, the present invention provides a method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis.
[0007] A method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis, characterized in that it comprises the following steps: S1, build a partial discharge test circuit or conduct partial discharge detection of switch cabinet on site to collect partial discharge UHF or TEV pulse time domain waveform data; S2, determine the appropriate sliding observation window width and filtering threshold size; S3, select the signal starting point, slide the observation window, and use peak detection in each observation window to intercept the effective partial discharge pulse signal; the partial discharge pulse current signal is used as the excitation of the sensor, and the sensor will generate a corresponding output signal. There is a corresponding relationship between the characteristics of the output signal and the partial discharge pulse current signal. This relationship is determined by the response characteristics of the sensor. Therefore, the partial discharge pulse voltage signal obtained by the sensor output can represent the actual partial discharge current pulse signal; S4, calculate and analyze the pulse signal in the observation window to obtain the partial discharge pulse time mark t n , n is the ordinal value of the number of pulses; according to the order of n, the time mark difference Δt of adjacent partial discharge pulses is calculated in sequence n , charge difference ΔQ n and the partial discharge pulse energy difference ΔE n parameter; S5, draw a spectrum of relevant parameters, analyze the characteristics of the spectrum, and establish a spectrum pattern under a certain partial discharge condition; the distribution pattern of data points in the spectrum reflects the characteristics of the corresponding discharge pattern, and partial discharge diagnosis and pattern recognition are performed based on the distribution pattern characteristics.
[0008] Furthermore, in order to better realize the present invention, S1 is specifically to build a partial discharge test circuit using UHF or TEV, wherein UHF is ultra-high frequency, and TEV is transient ground voltage method, and a high sampling frequency and high bandwidth oscilloscope or acquisition card is used to collect the partial discharge time domain signal, wherein the recommended sampling rate and bandwidth are not less than 10MHz, and in terms of partial discharge defects, taking the needle plate electrode simulating corona discharge as an example, the original partial discharge pulse waveform obtained by experiment.
[0009] Furthermore, in order to better realize the present invention, S2 is specifically as follows: after completing data acquisition, the size of the signal filtering threshold and the width of the single pulse observation window must be determined before peak detection; due to the influence of on-site interference, the original partial discharge signal obtained has more small-amplitude peaks, which may affect the accurate judgment of the partial discharge pulse peak in the observation window, so a signal filtering threshold should be set to ensure the effectiveness of partial discharge pulse identification; signals with amplitudes less than the filtering threshold are set to zero, and only signals greater than the filtering threshold are analyzed; the signal filtering threshold value is determined according to the signal-to-noise ratio; after selecting the sensor, a pre-experiment will be conducted on the device under test, and the waveform of the pre-experiment will be observed to determine the signal filtering threshold Value selection; For the original pulse waveform, an observation window with an appropriate time width should be selected to capture a single valid pulse; Here, the parameter T is used to represent the average width corresponding to each time window. The selection of the time window should prevent the window from being too wide to record multiple pulses, and prevent the window from being too narrow to cause the same pulse signal to be cut. The window should contain a single complete pulse; Within a single time window T, the maximum amplitude of all pulses is taken as the amplitude of the effective pulse in this window; The value of T is determined based on the results of preliminary experiments; Select UHF or TEV sensors to conduct multiple preliminary tests on a certain partial discharge to determine the effective duration t of a single partial discharge pulse signal. The window width T is related to t, and T can be 3t~5t.
[0010] Further, in order to better implement the present invention, S3 is specifically, after setting various parameters, taking the measured initial point as the starting point, sliding the observation window and cooperating with the peak detection to intercept the effective partial discharge pulse signal, and the peak detection algorithm is specifically as follows: S31, obtaining a sampling point sequence within a certain observation window; F=[f1,f2,f3,...,f n ] S32, performing differential processing; Diff = f i+1 -f i S33, normalize the data to -1, 0, 1; S34, to prevent the peak detection failure caused by the point with a differential value of 0, the same point 0 is set to the previous non-zero value; S35, finally Diff is differentiated again, and the value -2 or +2 is the peak or trough, and a valid peak value can be obtained in each observation window.
[0011] Further, in order to better implement the present invention, the S4 is specifically to calculate and analyze the pulse signal in the observation window to obtain Δt n , ΔQ n and ΔE n : S41, calculation and analysis of pulse signals in the observation window including partial discharge pulse time marker t n Determination of the charge Q n , partial discharge pulse energy E n Calculation of numerical values, where n is the ordinal value of the number of pulses; S42, peak detection can measure the time mark of each effective partial discharge pulse t n , which is the time point of the peak value in the observation window; the charge Q n and partial discharge pulse energy E n The integral calculation should be performed in a separate observation window; S43, the initial point of the current integration is set to the initial time point of each time window, and the specific calculation formula is as follows Among them, Q n is the charge amount, T is the average width of each time window, and I(t) is the current pulse signal, which is approximated by a Gaussian pulse; Similarly, the partial discharge pulse energy E n The specific calculation formula is: Where T is the average width of each time window, V(t) is the voltage waveform corresponding to I(t), and Z0 is the impedance value corresponding to the impedance matching of the detection circuit; S44, in order to analyze the relationship between consecutive pulses, the time mark, charge and energy difference between pulses are constructed. In order to make full use of the data, the change between the second-order data is selected. The time mark difference can be expressed as where i=[2,...,N] ΔQ n It can be expressed as S45, discretize the calculation formula, divide the time window T into N parts, change the integral form to the sum form, then ΔQ n The corresponding discretization formula is: where i=[2,...,N], I n is the partial discharge pulse voltage signal data with sequence number n output by the obtained sensor; Similarly, the discretization calculation formula for energy change can be obtained: Where i=[2,...,N], Z0 is the impedance value corresponding to the impedance matching of the detection circuit; In addition, it should be noted that in order to calculate t n and t n-1 The time initialization is crucial because t0 = 0 acts as a pre-trigger for the time starting point, so the ΔQ1 and ΔE1 at the beginning should be excluded.
[0012] Further, in order to better implement the present invention, S5 is specifically, according to the Δt calculated in S4 n , ΔQ n and ΔE n The data are plotted with Δt n , ΔQ n and ΔE n The three-dimensional scatter plots of the X, Y, and Z axes are shown. It can be seen that the time mark difference is within 0.6~2.0 ms, the charge difference is within -10~10, and the energy difference is within -0.6~0.6. n and ΔE n The points near the smaller values are evenly distributed on both sides of the zero axis of Y and Z, the distribution ratio of the positive half axis is slightly higher, and the overall distribution is dispersed; the signals collected from other partial discharge defects can be processed as above to obtain the corresponding three-dimensional spectrum distribution.
[0013] The beneficial effects of the present invention are: The method of the present invention fully considers the correlation between continuous pulses. The time difference Δt, charge difference ΔQ and partial discharge pulse energy difference ΔE obtained by pulse discretization calculation n It can show the change information during the discharge process, which is composed of Δt, ΔQ and ΔE n The features displayed by the constructed three-dimensional map can diagnose and evaluate the partial discharge in more detail and have wider applicability. In addition, evaluating the relationship between the discharge amount and the time difference can more accurately reflect the influence of the discharge current, which helps to understand the correlation between the continuous pulse relationship and different partial discharge sources, so as to accurately characterize the partial discharge with high resolution. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 This is a pre-experimental voltage waveform diagram of the present invention; Figure 2 It is a schematic diagram of the continuous sliding window of the present invention; Figure 3 is the Δt of the present invention n -ΔQ n -ΔE n Three-dimensional scatter plot. DETAILED DESCRIPTION
[0015] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Generally, the components of the embodiments of the present invention described and shown in the accompanying drawings here can be arranged and designed in various different configurations.
[0016] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.
[0017] Figure 1-Figure 3 This is a specific embodiment of the present invention, which is a method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis: 1. Build a partial discharge test circuit or conduct partial discharge detection on the on-site switch cabinet to collect partial discharge UHF or TEV pulse time domain waveform data.
[0018] 2. Determine the appropriate sliding observation window width and filtering threshold size.
[0019] 3. Select the signal starting point, slide the observation window, and use peak detection in each observation window to intercept the effective partial discharge pulse signal. At the same time, it is believed that there is a corresponding relationship between the partial discharge pulse voltage signal output by the sensor and the actual partial discharge current pulse signal, so the partial discharge pulse voltage signal output by the sensor can represent the actual partial discharge current pulse signal.
[0020] 4. Calculate and analyze the pulse signal in the observation window to obtain the partial discharge pulse time mark t n , n is the ordinal value of the number of pulses; according to the order of n, the time mark difference Δt of adjacent partial discharge pulses is calculated in sequence n , charge difference ΔQ n and the partial discharge pulse energy difference ΔE n And other parameters.
[0021] 5. Draw the spectrum of relevant parameters, analyze the characteristics of the spectrum, and establish the spectrum mode under certain partial discharge conditions. The distribution pattern of data points in the spectrum reflects the characteristics of the corresponding discharge mode, and partial discharge diagnosis and pattern recognition are performed based on the characteristics of the distribution pattern.
[0022] The specific plan is as follows: 1. Time domain waveform data acquisition of partial discharge UHF or TEV pulse of switch cabinet Use ultra-high frequency (UHF) or transient earth voltage method (TEV) to build a partial discharge test circuit, and use a high sampling frequency and high bandwidth oscilloscope or acquisition card to collect partial discharge time domain signals (the recommended sampling rate and bandwidth are not less than 10MHz). In terms of partial discharge defects, take the needle-plate electrode simulation corona discharge form as an example, and the original partial discharge pulse waveform obtained in the experiment.
[0023] 2. Determine the size of the signal filtering threshold and the width of the single pulse observation window; After completing data acquisition, the size of the signal filtering threshold and the width of the single pulse observation window must be determined before peak detection.
[0024] Due to the influence of on-site interference, the original partial discharge signal obtained contains many small-amplitude spikes, which may affect the accurate judgment of the partial discharge pulse peak in the observation window. Therefore, a signal filtering threshold should be set to ensure the effectiveness of partial discharge pulse identification. Signals with amplitudes less than the filtering threshold are set to zero, and only signals greater than the filtering threshold are analyzed. The signal filtering threshold value is generally determined according to the signal-to-noise ratio. In practice, after the sensor is selected, a pre-experiment will be conducted on the device under test, and the waveform of the pre-experiment will be observed to determine the value of the signal filtering threshold. For example, the pre-experiment results Figure 1 It can be observed that the maximum amplitude of the waveform is about 15mV, so it is more reasonable to set the trigger level value to 5mV according to the ratio of 3:1.
[0025] For the original pulse waveform, an observation window with an appropriate time width should be selected to intercept a single valid pulse. Here, the parameter T is used to represent the average width corresponding to each time window. The selection of the time window should prevent the window from being too wide to record multiple pulses, and prevent the window from being too narrow to cause the same pulse signal to be cut. The window should contain a single complete pulse. Within a single time window T, the maximum amplitude of all pulses is taken as the amplitude of the effective pulse in this window. The value of T is determined based on the results of the preliminary experiment. Select UHF or TEV sensors to conduct multiple preliminary tests on a certain partial discharge to determine the effective duration t of a single partial discharge pulse signal. The window width T is related to t and is generally taken as 3t~5t. In this experiment, UHF sensors were selected to conduct multiple preliminary experiments on corona discharge. The duration t of a single effective partial discharge pulse signal was about 1.5ns, so the window width T can be taken as 4.5~7.5ns.
[0026] 3. After setting all parameters, take the measured initial point as the starting point, slide the observation window and cooperate with peak detection to intercept the effective partial discharge pulse signal. The peak detection algorithm is as follows: 1) Obtain a sequence of sampling points within a certain observation window; F=[f1,f2,f3,...,f n ] 2) Perform differential processing; Diff = f i+1 -f i 3) Normalize the data to -1, 0, 1; 4) To prevent the peak detection failure caused by the point with a differential value of 0, set the same point 0 to the previous non-zero value; 5) Finally, Diff is differentiated again, and the value of -2 or +2 is the peak or trough. A valid peak value can be obtained in each observation window.
[0027] 4. Calculate and analyze the pulse signal in the observation window to obtain Δt n , ΔQ n and ΔE n .
[0028] The pulse signal calculation and analysis in the observation window includes the partial discharge pulse time mark t n Determination of the charge Q n , partial discharge pulse energy E n Calculation of numerical values, where n is the ordinal value of the number of pulses; Peak detection can measure the time mark of each effective partial discharge pulse t n, which is the time point of the peak value in the observation window; the charge Q n and partial discharge pulse energy E n The integral calculation should be performed in a separate observation window; The initial point of current integration is set as the initial time point of each time window. The specific calculation formula is as follows Similarly, the partial discharge pulse energy E n The specific calculation formula is: In order to analyze the relationship between consecutive pulses, the time mark, charge and energy differences between pulses are constructed. In order to make full use of the data, the variation between the second-order data is selected. The time mark difference can be expressed as where i=[2,...,N] ΔQ n It can be expressed as Discretize the calculation formula, divide the time window T into N parts, and change the integral form to the sum form, then ΔQ n The corresponding discretization formula is: where i=[2,...,N], I n is the partial discharge pulse voltage signal data with sequence number n output by the obtained sensor; Similarly, the discretization calculation formula for energy change can be obtained: Where i=[2,...,N], Z0 is the impedance value corresponding to the impedance matching of the detection circuit; In addition, it should be noted that in order to calculate t n and t n-1 The time initialization is crucial because t0 = 0 acts as a pre-trigger for the time starting point, so the ΔQ1 and ΔE1 at the beginning should be excluded.
[0029] 5. Draw a three-dimensional map of relevant parameters, analyze the map characteristics, and establish a map pattern under certain partial discharge conditions. The distribution pattern of data points in the map reflects the characteristics of the corresponding discharge pattern, and partial discharge diagnosis and pattern recognition are performed based on the distribution pattern characteristics.
[0030] Δt obtained by the calculation formula in 4 n , ΔQ n and ΔE n The data are plotted with Δt n, ΔQ n and ΔE n The three-dimensional scatter plots of the X, Y, and Z axes show that the time mark difference is basically within 0.6~2.0 ms, the charge difference is basically within -10~10, and the energy difference is basically within -0.6~0.6. n and ΔE n The points near the smaller values are basically evenly distributed on both sides of the zero axis of Y and Z, with a slightly higher distribution ratio on the positive half axis, and are generally dispersed. The signals collected from other partial discharge defects can be processed as described above to obtain the corresponding three-dimensional spectrum distribution.
[0031] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.
Claims
1. A method for identifying partial discharge of high-voltage switchgear based on time domain signal parameter analysis, characterized in that: The following steps are involved: S1, build a partial discharge test circuit or conduct partial discharge detection of switch cabinet on site to collect partial discharge UHF or TEV pulse time domain waveform data; S2, determine the width of the sliding observation window and the size of the filtering threshold; S3, select the signal starting point, slide the observation window, and use peak detection in each observation window to intercept the effective partial discharge pulse signal; the partial discharge pulse current signal is used as the excitation of the sensor, and the sensor will generate a corresponding output signal. There is a corresponding relationship between the characteristics of the output signal and the partial discharge pulse current signal. This relationship is determined by the response characteristics of the sensor. Therefore, the partial discharge pulse voltage signal obtained by the sensor output can represent the actual partial discharge current pulse signal; S4, calculate and analyze the pulse signal in the observation window to obtain the partial discharge pulse time mark t n , n is the ordinal value of the number of pulses; according to the order of n, the time mark difference Δt of adjacent partial discharge pulses is calculated in sequence n , charge difference ΔQ n and the partial discharge pulse energy difference ΔE n parameter; S5, draw a spectrum of relevant parameters, analyze the characteristics of the spectrum, and establish a spectrum pattern under a certain partial discharge condition; the distribution pattern of data points in the spectrum reflects the characteristics of the corresponding discharge pattern, and partial discharge diagnosis and pattern recognition are performed based on the distribution pattern characteristics.
2. The method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis according to claim 1 is characterized in that: Specifically, S1 is to build a partial discharge test circuit using UHF or TEV, wherein UHF is ultra-high frequency, and TEV is transient earth voltage method, and to collect the partial discharge time domain signal using a high sampling frequency and high bandwidth oscilloscope or acquisition card, wherein the sampling rate and bandwidth are not less than 10 MHz.
3. The method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis according to claim 1 is characterized in that: Specifically, S2 is as follows: after completing data acquisition, first determine the size of the signal filtering threshold and the width of the single pulse observation window before peak detection; set a signal filtering threshold to ensure the effectiveness of partial discharge pulse identification; set signals with amplitudes less than the filtering threshold to zero, and only analyze signals greater than the filtering threshold; the value of the signal filtering threshold is determined according to the signal-to-noise ratio; after selecting the sensor, conduct a preliminary experiment on the device under test, and observe the waveform of the preliminary experiment to determine the value of the signal filtering threshold; for the original pulse waveform, select a time window with a time width to intercept a single valid pulse; use parameter T to represent the average width corresponding to each time window, and the time window should contain a single complete pulse; within a single time window T, take the maximum amplitude of all pulses as the amplitude of the effective pulse in this window; the value of T is determined according to the results of the preliminary experiment; select a UHF or TEV sensor to conduct multiple preliminary tests on a certain partial discharge to determine the effective duration t of a single partial discharge pulse signal, the window width T is related to t, and T can be 3t~5t.
4. The method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis according to claim 1 is characterized in that: Specifically, S3 is to set various parameters, take the measured initial point as the starting point, slide the observation window and cooperate with peak detection to intercept the effective partial discharge pulse signal. The peak detection algorithm is as follows: S31, obtaining a sampling point sequence within a certain observation window; <h2 style=";text-align:left;direction:ltr">F=[f1,f2,f3,...,f<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> ] S32, performing differential processing; Diff = f i+1 -f i S33, normalize the data to -1, 0, 1; S34, to prevent the peak detection failure caused by the point with a differential value of 0, the same point 0 is set to the previous non-zero value; S35, finally Diff is differentiated again, and the value -2 or +2 is the peak or trough, and a valid peak value can be obtained in each observation window.
5. The method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis according to claim 1 is characterized in that: Specifically, S4 is to calculate and analyze the pulse signal in the observation window to obtain the pulse time mark difference value Δt n , charge difference ΔQ n and the partial discharge pulse energy difference ΔE n : S41, calculation and analysis of pulse signals in the observation window including partial discharge pulse time marker t n Determination of the charge Q n , partial discharge pulse energy E n Calculation of numerical values, where n is the ordinal value of the number of pulses; S42, peak detection can measure the time mark of each effective partial discharge pulse t n , which is the time point of the peak value in the observation window; the charge Q n and partial discharge pulse energy E n The integral calculation should be performed in a separate observation window; S43, the initial point of the current integration is set to the initial time point of each time window, and the specific calculation formula is as follows Among them, Q n is the charge amount, T is the average width of each time window, and I(t) is the current pulse signal, which is approximated by a Gaussian pulse; Similarly, the partial discharge pulse energy E n The specific calculation formula is: Where T is the average width of each time window, V(t) is the voltage waveform corresponding to I(t), and Z0 is the impedance value corresponding to the impedance matching of the detection circuit; S44, in order to analyze the relationship between consecutive pulses, the time mark, charge and energy difference between pulses are constructed. In order to make full use of the data, the change between the second-order data is selected. The time mark difference can be expressed as where i=[2,...,N] ΔQ n It can be expressed as S45, discretize the calculation formula, divide the time window T into N parts, change the integral form to the sum form, then ΔQ n The corresponding discretization formula is: where i=[2,...,N], I n is the partial discharge pulse voltage signal data with sequence number n output by the obtained sensor; Similarly, the discretization calculation formula for energy change can be obtained: Where i=[2,...,N], Z0 is the impedance value corresponding to the impedance matching of the detection circuit; In addition, in order to calculate t n and t n-1 The continuous time difference between them needs to be initialized. Since t0=0 acts as a pre-trigger for the starting point of time, ΔQ1 and ΔE1 at the beginning should be excluded.
6. The method for identifying partial discharge of a high-voltage switch cabinet based on time domain signal parameter analysis according to claim 1 is characterized in that: Specifically, S5 is: according to the Δt calculated in S4 n , ΔQ n and ΔE n The data are plotted with Δt n , ΔQ n and ΔE n The three-dimensional scatter plots of the X, Y, and Z axes are shown. It can be seen that the time mark difference is within 0.6~2.0 ms, the charge difference is within -10~10, and the energy difference is within -0.6~0.
6. n and ΔE n The points near the zero axis and other points are evenly distributed on both sides of the zero axis of Y and Z, with a slightly higher distribution ratio on the positive half axis, and are generally dispersed; the signals collected from other partial discharge defects can be processed as above to obtain the corresponding three-dimensional spectrum distribution.
Citation Information
Patent Citations
Local discharge development process diagnosis method based on pulse-train analysis
CN104198898A
Signal and noise separation method for partial discharge signal and information data processing terminal
CN109901031A
Time domain reconstruction-based local discharging pulse counting method
CN110244199A
Method for detecting partial discharge of power distribution network switch cabinet by ultrahigh-frequency sensor
CN112098748A
Audio-based switch cabinet partial discharge detection method
CN114200260A
Cited By
Ring main unit intelligent sensing on-line monitoring system and method thereof
CN120103041A
Partial discharge monitoring operation and maintenance method, system and device based on intelligent agent
CN120805004A
GIS equipment partial discharge type classification method based on partial discharge time domain signals
CN120948990A
