Tree-line fault impedance measurement method

Abstract

The present invention discloses a tree-line fault impedance measurement method, which relates to the field of high voltage test technology. The method comprises: measuring the leakage current flowing through the entire live tree by suspending a conductor in the front section of the fault point, collecting voltage signals at both ends of the line to obtain the fault point voltage; performing noise reduction on the collected leakage current and fault point voltage through wavelet transformation; interpolating shorter signal data through quadratic polynomial fitting to align the two signals; and calculating the equivalent impedance change of the tree. The present invention is used for fault mechanism analysis and fault identification, and is of great significance for ensuring the safety of distribution lines and preventing and controlling wildfires caused by tree-line faults.

Description

Technical Field

[0001] The invention relates to the technical field of high voltage testing, in particular to a method for measuring tree-line fault impedance. Background Art

[0002] A tree-line fault in a single-phase high-resistance grounding fault in a power system line refers to a tree contacting a conductor, thereby causing the conductor to be single-phase grounded. When analyzing the tree-line fault mechanism and fault identification, it is necessary to study how the impedance of the tree changes as the fault develops. The calculation is performed by measuring the leakage current flowing through the tree and the line phase voltage at the fault point. However, there are several problems. First, the leakage current flowing through the tree is difficult to measure; second, the leakage current needs to be measured at the fault point, and measuring the voltage at the fault point requires a power voltage transformer, but this device is not necessarily next to the fault point. Usually, the line voltage value can be measured at the beginning and end of the line, but the voltage at the fault point cannot be directly measured. voltage; thirdly, there may be other feeders in the line. If two fault recorders are needed to respectively collect the current waveform signal of the fault point and the voltage waveform signal at both ends of the line, the sampling rates of the signals collected by the two different devices may be different, and even the lengths of the fault signals collected by the two devices may be different due to the difference in equipment; fourthly, the current signal collected in the tree-line fault is not only affected by environmental noise, but also by the arc current caused by the gap between the conductor and the tree; the arc resistance is between 1-100Ω and the tree impedance is between 1-1000kΩ. The influence of the arc resistance on the entire transition resistance can be ignored, so the influence of the arc current must be excluded.

[0003] Line faults are high-resistance faults in high voltage and insulation technology. Tree-line faults are one of the important issues facing power grid security, which can cause transmission line tripping and wildfires. Studying the mechanism of tree-line faults and prevention measures is of great significance for preventing wildfires and power grid accidents. Currently, most scholars identify faults by analyzing and finding the characteristic quantities of electrical quantities such as three-phase voltage, three-phase current, zero-sequence voltage, and zero-sequence current, or by establishing arc models. Few scholars study the mechanism of tree-line faults and fault identification from the perspective of tree equivalent impedance. Studying the characteristics of tree impedance changes is also important for protecting lines and preventing wildfires. In the real experiment, wind caused the separation of trees and lines, generating arc current, which will affect the calculation of equivalent impedance. Summary of the Invention

[0004] To solve the problems existing in the prior art, the purpose of the present invention is to provide a tree-line fault impedance measurement method. The present invention is used for fault mechanism analysis and fault identification, which is of great significance for ensuring the safety of distribution lines and preventing and controlling wildfires caused by tree-line faults.

[0005] To achieve the above object, the present invention adopts a technical solution: a tree-line fault impedance measurement method, comprising the following steps:

[0006] Step 1: Measure the leakage current flowing through the entire live tree by suspending the wire in front of the fault point, and collect voltage signals at both ends of the line to obtain the fault point voltage;

[0007] Step 2: De-noise the collected leakage current and fault point voltage through wavelet transform;

[0008] Step 3: Interpolate the shorter signal data by fitting a quadratic polynomial to align the two signals;

[0009] Step 4: Calculate the change in tree equivalent impedance.

[0010] As a further improvement of the present invention, the step 1 specifically includes the following steps:

[0011] Step 1.1: Use a three-phase transformer to increase the 380V voltage of the power distribution cabinet to the required voltage level;

[0012] Step 1.2: One phase of the high-voltage three-phase line is in contact with the entire living tree;

[0013] Step 1.3: The end of the high-voltage three-phase line is unloaded, and the ends of the three-phase conductors are suspended. A through-hole current transformer is installed before the fault point. The fault recorder S2 uses the through-hole current transformer to collect the current waveform flowing through the fault phase conductor as the leakage current waveform I0; the fault recorder S1 uses the through-hole current transformer to collect the fault phase voltage waveform U0.

[0014] As a further improvement of the present invention, the step 2 specifically includes the following steps:

[0015] Step 2.1, estimate the noise standard deviation σ of the leakage current and the fault point voltage;

[0016] Step 2.2: Calculate the soft threshold T by multiplying the noise standard deviation by an empirical constant C to determine the threshold: T = σ × C;

[0017] Step 2.3, using wavelet basis to perform multi-layer decomposition on the fault waveform in the current waveform and voltage waveform and extract coefficients;

[0018] Step 2.4: Use the soft threshold T calculated based on the background noise to perform denoising and construct the denoised fault waveform;

[0019] Step 2.5: Calculate the root mean square value (RMS) of the entire fault waveform within a preset time period to obtain the effective values ​​I1 and U1 of the current signal and voltage signal.

[0020] As a further improvement of the present invention, the step 2.1 is specifically as follows:

[0021] Calculate the average value μ of the noise waveform of the leakage current and fault point voltage: Where μ represents the average value of the noise waveform, N represents the total number of data points of the noise waveform signal, and x i Represents each data point of the noise waveform;

[0022] Compute the square of the difference between each data point and the mean and take the average: Where, σ 2 represents the square of the noise standard deviation;

[0023] Calculate the noise standard deviation σ:

[0024] As a further improvement of the present invention, in step 2.5, the effective values ​​I1 and U1 of the current signal and the voltage signal are as follows:

[0025] As a further improvement of the present invention, the present invention further includes: using the moving average of the smooth function in matlab to smooth the effective value I1 of the current signal, removing the sudden change in current caused by the arc current, and obtaining the effective value I2 of the current signal after smoothing.

[0026] As a further improvement of the present invention, the step 3 specifically includes the following steps:

[0027] Step 3.1. Compare the lengths of the effective value signal arrays and calculate the number of points that need to be inserted.

[0028] Step 3.2: Evenly distribute the points to be inserted into the shorter signal array.

[0029] Step 3.3: Use polynomial fitting to interpolate: Take the data in multiple data windows before and after the interpolation point and perform polynomial fitting: y = a n x n +a n-1 x n-1 +…+a1x+1; solve the polynomial coefficients a0, a1, …, a by the least squares method n Generate a fitting curve, use the fitted polynomial to generate the curve, and calculate the values ​​of the missing points in the middle; insert the missing data points in the fitting curve and align the data to obtain signals U2 and I3.

[0030] As a further improvement of the present invention, step 4 specifically includes the following steps:

[0031] Step 4.1. Calculate the voltage of the fault phase at the fault point:

[0032] Calculate the line resistance: Where R L=I3R represents the resistance of the line, ρ represents the resistivity of the line material, L represents the length of the line, and S is the cross-sectional area of ​​the conductor; calculate the voltage at the fault point: U3 = U2 - I3R L Where U3 represents the effective value of the fault phase voltage at the fault point, U2 represents the effective value of the voltage signal after alignment, and I3 represents the effective value of the current signal after alignment, i.e., the effective value of the leakage current.

[0033] Step 4.2, calculate the tree equivalent impedance change R:

[0034] The beneficial effects of the present invention are:

[0035] The present invention can be used to collect and process electrical signals in tree-line fault experiments, realizing the collection of leakage current of an entire living tree, reducing the influence of arc current on the calculated impedance, solving the pain point that dual-fault recorders have difficulty calculating impedance, and improving the accuracy of the tree's equivalent impedance by calculating the phase voltage at the fault point. The present invention is used for fault mechanism analysis and fault identification, and is of great significance for ensuring the safety of distribution lines and preventing and controlling wildfires caused by tree-line faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flow chart of an embodiment of the present invention;

[0037] Figure 2 This is an example diagram of a method for measuring leakage current flowing through an entire living tree according to an embodiment of the present invention;

[0038] Figure 3 This is a complete waveform diagram of the current signal in an embodiment of the present invention;

[0039] Figure 4 The waveform diagram of the current signal in the fault stage is obtained by removing the waveforms of the current signal in the non-fault stage before and after the current signal according to the embodiment of the present invention;

[0040] Figure 5 This is a waveform diagram after threshold denoising obtained through background noise calculation in an embodiment of the present invention;

[0041] Figure 6 is a curve diagram of the effective value change of the current signal after calculation in an embodiment of the present invention;

[0042] Figure 7 3 is a curve diagram of the effective value change of the current signal after smoothing in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0044] Example

[0045] like Figure 1As shown, a tree-line fault impedance measurement method includes measuring the leakage current flowing through the entire live tree, current signal noise reduction, multiple signal alignment, and calculating the tree equivalent impedance; the details are as follows:

[0046] 1. A method for measuring leakage current flowing through a whole living tree comprises the following steps:

[0047] Step 1: Use a three-phase transformer to increase the 380V voltage of the distribution incoming cabinet to the required voltage level. Take 10kV as an example. It is equipped with other supporting facilities, such as transformers, circuit breakers, mutual inductors, fault recorders S1 and S2. Figure 2 shown.

[0048] Step 2: One phase of the high-voltage three-phase line is in contact with the entire living tree.

[0049] Step 3: The end of the high-voltage three-phase line is not loaded, and the ends of the three-phase conductors are suspended. A through-hole current transformer is installed before the fault point. The fault recorder S2 collects the current waveform flowing through the fault phase conductor through the through-hole current transformer as the leakage current waveform I0, as shown in Figure 3 As shown in Figure 1, the fault recorder S1 collects the fault phase voltage waveform U0.

[0050] 2. Signal noise reduction method takes current signal as an example, including the following steps:

[0051] Step 1: Estimate the noise standard deviation. The fault phase current waveform I0 collected by the fault recorder S2 includes the noise waveform collected before the fault occurs and the fault current waveform collected after the fault occurs. The standard deviation of the noise is calculated for the noise waveform part.

[0052] Step 1.1: Calculate the mean value μ of the noise waveform:

[0053]

[0054] Where μ represents the average value of the noise waveform, N represents the total number of data points of the noise waveform signal, and x i Represents each data point of the noise waveform.

[0055] Step 1.2: Calculate the square of the difference between each data point and the mean and average them:

[0056]

[0057] Where: σ 2 Represents the square of the standard deviation.

[0058] Step 1.3: Calculate the standard deviation:

[0059]

[0060] Where: σ represents the standard deviation.

[0061] Step 2: Calculate the soft threshold T and use the noise standard deviation multiplied by an empirical constant C to determine the threshold.

[0062] T=σ×C

[0063] Step 3: Use wavelet base DB2 to perform 3-layer decomposition on the fault waveform in the current waveform and extract the coefficients.

[0064] Step 4: Use the soft threshold T calculated based on the background noise to perform denoising and construct the denoised fault waveform.

[0065] Step 5: Calculate the root mean square value (RMS) of the entire fault waveform with a period of 0.02 seconds, which is the effective value I1. Similarly, calculate the effective value U1 of the voltage signal after processing.

[0066]

[0067] Where N is the total number of data points per 0.02s in the fault waveform, x i Represents each data point every 0.02s in the noise waveform.

[0068] Step 6: At this point, the processed current waveform still has the sudden current caused by the arc current, but it does not affect the overall trend of the data points. By smoothing the data, the sudden current value can be removed.

[0069] Step 6.1: Use the Moving Average of the smoothing function in Matlab, and the sliding window is set to 100 based on experience.

[0070] Step 6.2: After smoothing, the influence of arc current is removed to obtain the effective value I2 of the current signal.

[0071] 3. A method for aligning multiple signals, comprising the following steps:

[0072] Step 1: Compare the lengths of the two valid value signal arrays and calculate the number of points that need to be interpolated.

[0073] Step 2: Evenly distribute the points to be inserted into the shorter signal array.

[0074] Step 3: Interpolate using polynomial fitting.

[0075] Step 3.1: Take the data in the 200 data windows before and after the interpolation point and perform polynomial fitting, where the data points are (x1, y1), (x2, y2), and so on. Fit according to the formula. This data window can be adjusted as needed:

[0076] y=an x n +a n-1 x n-1 +…+a1x+1

[0077] Step 3.2: Solve the polynomial coefficients a0, a1, …, a by the least squares method n .

[0078] Step 3.3: Generate a fitting curve, use the fitted polynomial to generate a curve, and calculate the values ​​of the missing points in the middle.

[0079] Step 3.4: Insert the missing data points in the fitting curve and align the data to obtain signals U2 and I3. In this process, the current signal is processed as follows Figure 4-Figure 7 shown.

[0080] 4. The method for calculating the equivalent impedance of trees includes the following steps:

[0081] Step 1: You need to calculate the voltage of the fault phase at the fault point.

[0082] Step 1.1: The distribution network line is short, and the voltage drop is mainly caused by the line's own resistance. Calculate the line resistance:

[0083]

[0084] Where R L represents the resistance of the line, ρ represents the resistivity of the line material, L represents the length of the line, and S is the cross-sectional area of ​​the conductor.

[0085] Step: 1.2: Calculate the voltage at the fault point:

[0086] U3=U2-I3R L

[0087] Where U3 represents the effective value of the fault phase voltage at the fault point, U2 represents the effective value of the voltage signal after alignment, and I3 represents the effective value of the current signal after alignment, that is, the effective value of the leakage current.

[0088] Step 2: Calculate the tree equivalent impedance change R using the formula:

[0089]

[0090] The above-described embodiments merely represent specific implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, and all such variations and improvements fall within the scope of protection of the present invention.

Claims

1. A method for measuring tree-line fault impedance, characterized in that: The following steps are involved: Step 1: Measure the leakage current flowing through the entire live tree by suspending the wire in front of the fault point, and collect voltage signals at both ends of the line to obtain the fault point voltage; Step 2: De-noise the collected leakage current and fault point voltage through wavelet transform; The step 2 specifically includes the following steps: Step 2.1, estimate the noise standard deviation σ of the leakage current and the fault point voltage; The step 2.1 is as follows: Calculate the average value μ of the noise waveform of the leakage current and fault point voltage: Where μ represents the average value of the noise waveform, N represents the total number of data points of the noise waveform signal, and x i Represents each data point of the noise waveform; Compute the square of the difference between each data point and the mean and take the average: Where, σ 2 represents the square of the noise standard deviation; Calculate the noise standard deviation σ: Step 2.2: Calculate the soft threshold T by multiplying the noise standard deviation by an empirical constant C to determine the threshold: T = σ × C; Step 2.3, using wavelet basis to perform multi-layer decomposition on the fault waveform in the current waveform and voltage waveform and extract coefficients; Step 2.4: Use the soft threshold T calculated based on the background noise to perform denoising and construct the denoised fault waveform; Step 2.5: Calculate the root mean square value (RMS) of the entire fault waveform within a preset time period to obtain the effective values ​​I1 and U1 of the current signal and voltage signal; Step 3: Interpolate the shorter signal data by fitting a quadratic polynomial to align the two signals; The step 3 specifically includes the following steps: Step 3.

1. Compare the lengths of the effective value signal arrays and calculate the number of points that need to be inserted. Step 3.2: Evenly distribute the points to be inserted into the shorter signal array. Step 3.3: Use polynomial fitting to interpolate: Take the data in multiple data windows before and after the interpolation point and perform polynomial fitting: y = a n x n +a n-1 x n-1 +…+a1x+1; solve the polynomial coefficients a0, a1, …, a by the least squares method n Generate a fitting curve, use the fitted polynomial to generate the curve, calculate the value of the missing point in the middle; insert the missing data points in the fitting curve and align the data to obtain signals U2 and I3; Step 4: Calculate the change in tree equivalent impedance.

2. The tree-line fault impedance measurement method according to claim 1, characterized in that: The step 1 specifically includes the following steps: Step 1.1: Use a three-phase transformer to increase the 380V voltage of the power distribution cabinet to the required voltage level; Step 1.2: One phase of the high-voltage three-phase line is in contact with the entire living tree; Step 1.3: The end of the high-voltage three-phase line is unloaded, and the ends of the three-phase conductors are suspended. A through-hole current transformer is installed before the fault point. The fault recorder S2 uses the through-hole current transformer to collect the current waveform flowing through the fault phase conductor as the leakage current waveform I0; the fault recorder S1 uses the through-hole current transformer to collect the fault phase voltage waveform U0.

3. The tree-line fault impedance measurement method according to claim 1, characterized in that: In step 2.5, the effective values ​​of the current signal and voltage signal I1 and U1 are as follows:

4. The tree-line fault impedance measurement method according to claim 2, characterized in that: Also includes: In MATLAB, the moving average of the smooth function is used to smooth the effective value of the current signal I1 to remove the sudden change caused by the arc current. After smoothing, the effective value of the current signal I2 is obtained.

5. The tree-line fault impedance measurement method according to claim 1, characterized in that: The step 4 specifically includes the following steps: Step 4.

1. Calculate the voltage of the fault phase at the fault point: Calculate the line resistance: Where R L =I3R represents the resistance of the line, ρ represents the resistivity of the line material, L represents the length of the line, and S is the cross-sectional area of ​​the conductor; calculate the voltage at the fault point: U3 = U2 - I3R L ; Where, U3 represents the effective value of the fault phase voltage at the fault point, U2 represents the effective value of the voltage signal after alignment, and I3 represents the effective value of the current signal after alignment, that is, the effective value of the leakage current; Step 4.2, calculate the tree equivalent impedance change R:

Images