Arc-type ground fault detection method for low voltage distribution lines based on parameter identification

Abstract

The invention discloses a method for detecting arc-type ground faults in low-voltage distribution lines based on parameter identification, which belongs to the field of fault protection of low-voltage distribution networks, including setting monitoring points on low-voltage distribution lines, synchronously sampling the voltage and current of the monitoring points, and extracting data for constructing overdetermined equations; based on the data, using overdetermined equations in units of half-waves, respectively, calculating the values ​​of unknown quantities under positive half-waves and negative half-waves; and determining the identification results of ground faults in units of half-waves. The method of the invention detects arc-type ground faults by identifying the waveform characteristics of the arc voltage at the fault point, and uses a method of solving the unknown quantity of arc voltage using overdetermined equations to solve the protection blind spot of the ground fault in the TN-S part of the TN-C-S system. The low-voltage distribution terminal can be used as an implementation carrier, and combined with the existing low-voltage ground fault protection algorithm to improve the protection capability of ground faults.

Description

Technical Field

[0001] The present invention relates to the field of low-voltage distribution network fault protection, and in particular to a method for detecting arc-type grounding faults in low-voltage distribution lines based on parameter identification. Background Art

[0002] Ground fault is a common fault in low-voltage power distribution system. The protection against ground fault is directly related to the wiring mode of low-voltage power distribution system. TN-CS system is the main wiring mode in urban systems in my country. For such system, the system from distribution transformer to user meter box is TN-C mode (neutral line and ground line are shared), and TN-S mode is adopted after entering the house. The ground fault on the user side is mainly caused by phase line touching PE line (ground line) or phase line touching equipment casing.

[0003] my country requires the use of three-level protection measures for overcurrent protection and residual current protection, that is, to install overcurrent and residual current protection devices at the transformer outlet, branch line outlet and user side (i.e. TN-S part of TN-CS system). Its actual implementation methods include low-voltage intelligent miniature circuit breakers, low-voltage leakage protectors, molded case circuit breakers, etc.

[0004] Nevertheless, for overcurrent protection, when arc-type grounding occurs, due to the strong randomness of the arc resistance at the fault point, when the arc resistance at the fault point is large, any level of overcurrent protection theoretically has a protection blind spot. As for residual current protection, the TN-CS system residual current protection has the problem that the total protection and the intermediate protection cannot be put into operation. The reason is that there are one or more grounding points on the neutral line, and part of the neutral line current returns to the transformer from the grounding point, resulting in the actual normal residual current at the total protection far exceeding the 500mA required by the national standard, and the normal residual current at the intermediate protection far exceeding the 300mA required by the national standard. The phenomenon that the total protection and the intermediate protection cannot be put into operation is common.

[0005] In summary, for ground faults in TN-CS systems, there are a large number of scenarios in actual sites where residual current protection is not installed or cannot be installed, and there is a problem of being unable to perceive arc-type grounding faults. Summary of the invention

[0006] In view of the above problems, the present invention provides a method for detecting arc-type grounding faults in low-voltage distribution lines based on parameter identification, focusing on solving the problem of detecting arc-type grounding faults in scenarios without residual current protection.

[0007] To achieve the above object, the technical solution of the present invention is:

[0008] A method for detecting arc-type ground faults in low-voltage distribution lines based on parameter identification comprises the following steps:

[0009] Step S1, setting a monitoring point on the low-voltage distribution line, synchronously sampling the voltage and current of the monitoring point, and extracting data for constructing an overdetermined equation;

[0010] Step S2, based on the data, using half-wave as a unit, respectively using overdetermined equations to calculate the line resistance under positive half-wave and negative half-wave R 1. Inductor L 1. Voltage U h The value of , the overdetermined equation is,

[0011]

[0012] In the above formula, i h ( t ) represents the sampling point current, u s ( t ) represents the sampling point voltage (t=1,2,3…), R 1 represents the line resistance between the fault point and the monitoring point, L 1 represents the inductance between the fault point and the monitoring point, U h Represents the voltage of the equivalent DC power supply at the fault point, , , The calculation formula is:

[0013]

[0014] In the formula, ΔT Indicates the sampling time interval;

[0015] Step S3, judging the identification result of the ground fault in units of half-wave, and the judgment criterion is:

[0016]

[0017] In the formula, α Indicates the judgment threshold of arc burning voltage identification, The voltage of the equivalent DC power supply at the fault point representing the positive half-wave, The voltage of the equivalent DC power supply at the fault point representing the negative half-wave.

[0018] In step S1 of the above-mentioned low-voltage distribution line arc-type ground fault detection method based on parameter identification, data for constructing an overdetermined equation is extracted, including the following steps:

[0019] Step S11, set the number of sampling points per cycle to N, and measure the voltage of the sampling point at the monitoring point in units of cycles. u s (t ) and sampling point current i h ( t ) for synchronous sampling, and the sampled discrete signal is recorded as the sampling point voltage discrete signal u s ( n ) and the discrete current signal of the sampling point i h ( n );

[0020] Step S12, using fast Fourier transform to obtain the discrete voltage signal of the sampling point u s ( n )'s real and imaginary parts of the fundamental component a , b , , and then through , calculate the initial phase angle θ 1;

[0021] Step S13, respectively calculating the interval ΔN1 between the initial sampling point of the current cycle and the positive half-wave and the interval ΔN2 between the initial sampling point of the current cycle and the negative half-wave, the calculation formula is:

[0022] Positive half wave:

[0023] Negative half wave:

[0024] In the formula, “|…|” means taking the absolute value, “mod |…| 2π " means taking the modulus of the angle;

[0025] Step S14: If the voltage discrete signal at the sampling point u s ( n ) is denoted as N 0, then the sampling points of the peak voltage points of the positive half-wave and the negative half-wave are:

[0026] Positive half wave:

[0027] Negative half wave:

[0028] In the formula, represents the sampling point of the peak voltage point of the positive half-wave, The sampling point representing the peak voltage point of the negative half-wave;

[0029] Step S15, assuming that the width of the data selection window is set to M+1 sampling points, the voltage discrete signals of the sampling points are respectively u s( n ) is the sampling point of the peak voltage point of the positive half wave , the sampling point of the peak voltage point of the negative half wave Select the center of the data window, select M / 2 sampling points on the left and right sides respectively, and complete the sampling point voltage discrete signal with a length of M+1 u s ( n ) in data extraction;

[0030] Step S16, obtaining the current discrete signal of the sampling point i h ( n ), and the current discrete signal of the sampling point i h ( n ) to filter out the high-frequency components in the current and obtain a low-frequency current discrete signal i' h ( n );

[0031] Step S17, respectively using discrete signals of the sampling point voltage u s ( n ) is the sampling point of the peak voltage point of the positive half wave , the sampling point of the peak voltage point of the negative half wave Corresponding synchronous current sampling points , As a low-frequency current discrete signal i' h ( n ) is the center of the positive half-wave data acquisition window and the negative half-wave data acquisition window, and M / 2 sampling points are selected on the left and right sides to complete the low-frequency current discrete signal with a length of M+1. i' h ( n ) in data extraction;

[0032] Step S18, according to the following formula, using i' h (1) i' h (2) , ..., the value of , ,……,

[0033] .

[0034] In a preferred embodiment of the present invention, in step S16, the filtering process adopts smoothing filtering or FIR low-pass filtering.

[0035] In a preferred embodiment of the present invention, if the filtering algorithm results ini' h ( n )and i h ( n ) has moved, the phase of the filtered signal needs to be restored.

[0036] In a preferred embodiment of the present invention, the total time length corresponding to the data selection window is preferably 3ms~6ms.

[0037] In a preferred embodiment of the present invention, the determination threshold of arc burning voltage identification is α Preferably 12V.

[0038] In a preferred embodiment of the present invention, the low-voltage power distribution line is a TN-CS system, and the monitoring point is arranged in the TN-S part of the TN-CS system.

[0039] Beneficial effect: The method of the present invention detects arc-type grounding faults by identifying the waveform characteristics of the arc voltage at the fault point, and uses a method of solving the unknown quantity of the arc voltage using an overdetermined equation to solve the protection blind spot of the grounding fault in the TN-S part of the TN-CS system. Low-voltage distribution terminals such as low-voltage intelligent circuit breakers, leakage protectors, molded case circuit breakers, and smart meters can be used as implementation carriers, and combined with existing low-voltage grounding fault protection algorithms to improve the protection capability of grounding faults.

[0040] In order to make the above features and advantages of the invention more obvious and easy to understand, embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of the arc voltage waveform at the fault point when an arc-type grounding fault occurs.

[0042] Figure 2 This is the equivalent circuit diagram of a ground fault.

[0043] Figure 3 The present invention is a flow chart of a method for detecting arc-type ground faults in low-voltage power distribution lines based on parameter identification.

[0044] Figure 4 Based on Figure 2 Schematic diagram of the equivalent circuit for fault point DC characteristic identification.

[0045] Figure 5 Schematic diagram of voltage and current data extraction for equation solving. DETAILED DESCRIPTION

[0046] In order to make the purpose and technical solution of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment 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.

[0047] The present invention relates to a method for detecting arc-type grounding faults in low-voltage power distribution lines based on parameter identification. According to electrical theory, when a low-voltage arc burns stably, the arc voltage is approximately the sum of its near-anode and near-cathode voltage drops, that is:

[0048] (1)

[0049] In formula (1), U h Indicates the arc stable burning voltage, U c represents the voltage drop near the cathode, U a Indicates the voltage drop near the anode. For a specific metal electrode and gas medium, the voltage drop near the cathode U c and near anode voltage drop U a All are constant. Voltage drop near cathode U c and near anode voltage drop U a Generally between 8 and 10V. U c + U a "Generally more than 15V.

[0050] Figure 1 Figure 1 is a schematic diagram of the arc voltage waveform at the fault point when an arc-type ground fault occurs. Figure 1 From the waveform characteristics shown in the figure, when an arc-type ground fault occurs, it first goes through the arc starting stage T1, and the DC arc burning voltage V It is equivalent to connecting a square wave voltage source in parallel between the phase line and the equipment housing (or PE line). The arcing period is equivalent to a DC voltage source. When the arc-type grounding fault is removed, it goes through the arc extinction stage T2. The DC arcing voltage V Return to normal.

[0051] Figure 2 is the equivalent circuit diagram of ground fault. Figure 2 As shown, the fault point in the equivalent circuit is equivalent to an equivalent DC power supply. U h . Equivalent AC power supply UAC Connection equivalent inductance L S One end of the equivalent inductance L S The other end of the first inductor is connected L 1, the first inductor L The other end of 1 is connected to the first resistor R 1, the first resistor R The other end of 1 is connected to an equivalent DC power supply U h The positive terminal and the second inductor L 2, equivalent DC power supply U h The negative pole of the second inductor is grounded. L The other end of 2 is connected to the second resistor R 2, one end of the second resistor R The other end of 2 is connected to one end of the load, and the other end of the load is grounded.

[0052] If the monitoring point is set upstream of the fault point, that is, the equivalent inductance L S With the first inductor L 1, according to Kirchhoff's voltage law (KVL),

[0053] (2)

[0054] In formula (2), u s ( t ) represents the voltage sampled at the monitoring point, i h ( t ) represents the current sampled at the monitoring point, R 1 represents the resistance of the first resistor (i.e. the line resistance between the fault point and the monitoring point), L 1 represents the inductance value of the first inductor (i.e. the inductance between the fault point and the monitoring point), U h Indicates the voltage of the equivalent DC power supply at the fault point.

[0055] Since the distance between the fault point and the monitoring point is unknown, the R 1. L 1. U h are all unknown quantities. From formula (2), we can see that since the equivalent load parameters in this formula are only R 1. L 1, so the system belongs to a linear system. By measuring the sampling point voltage and sampling point current of multiple sampling points at the monitoring point, an overdetermined equation can be constructed. R 1.L 1. U h Solve the identification, when the voltage of the equivalent DC power supply at the identified fault point is U h When it is greater than the set threshold, it can be considered that a DC voltage source unique to arc burning exists, and an arc-type grounding fault is considered to have occurred.

[0056] The equation for identifying the existence of DC arcing voltage using overdetermined equations is as follows:

[0057] (3)

[0058] From formula (3), we can see that by continuously adjusting the voltage at the sampling point u s ( t ) and sampling point current i h ( t ) to perform synchronous sampling and obtain multiple synchronously sampled discrete values i h (1) i h (2) ..., u s (1) u s (2), ..., and , ,…….in, , , The value of i h (1) i h (2) The values ​​of ... are calculated as follows:

[0059] (4)

[0060] In formula (4), ΔT Indicates the sampling time interval.

[0061] Figure 3 FIG. 1 is a flow chart of a method for detecting arc-type ground faults in low-voltage power distribution lines based on parameter identification according to the present invention. Figure 3 The arc type ground fault detection method of the present invention has the following steps:

[0062] Step S1, setting monitoring points on the low-voltage distribution line, synchronously sampling the voltage and current of the monitoring points, and extracting data for constructing overdetermined equations.

[0063] Step S2, based on the data, using the above formula (3) in units of half-wave, calculate the unknown quantity under the positive half-wave and the negative half-wave R 1. L 1. U h The value of .

[0064] Step S3, judging the identification result of the ground fault in units of half-wave, and the judgment criterion is:

[0065] (5)

[0066] In formula (5), α Indicates the judgment threshold of arc burning voltage identification, The voltage of the equivalent DC power supply at the fault point representing the positive half-wave, The voltage of the equivalent DC power supply at the fault point representing the negative half-wave.

[0067] In the above formula (1), the voltage drop near the cathode is U c and near anode voltage drop U a Generally between 8 and 10V. U c + U a "Generally more than 15V. Affected by the presence of harmonics, the actual identification result using the overdetermined equation has a certain error with the ideal value. When there is no arc-type grounding fault, the identification result is near 0V. When an arc-type grounding fault occurs, the identification result is generally not less than 12V. Therefore, the judgment threshold can be set in practice α When a DC voltage source of 12V or above is detected, it is considered that an arc-type grounding fault has occurred between the phase line and the equipment casing or the PE line.

[0068] In the above formula (3), although there are only three unknowns, the equation can be solved by using sampling value information greater than three, that is, by constructing an overdetermined equation with a number of equations greater than the number of unknowns, and using the least squares solution of the overdetermined equation as the final result of the equation. This can greatly improve the accuracy of the solution of the equation, and the more input variables used to solve the equation, the higher the solution accuracy.

[0069] According to the existing literature 1: Wang Wei, Xu Bingyin, Sun Zhongyu, Liang Dong. Voltage differential detection method for arc fault in low voltage distribution network[J]. Proceedings of the CSEE. 2023, 43(22): 8674-8688., combined with Figure 1From the arc voltage waveform at the fault point in the figure, it can be seen that the entire arc fault process is divided into three stages: arcing, arcing and arcing extinction. The method proposed in the present invention is theoretically intended to identify the existence of DC arcing voltage in the arcing stage. Therefore, in the implementation of this method, when extracting sampling points as inputs for constructing the above formula (3), the data during the arcing period should be used, and the data during the arcing and arcing periods should be avoided.

[0070] Figure 4 Based on Figure 2 Schematic diagram of the equivalent circuit for fault point DC characteristic identification. Figure 4 In the equivalent circuit of the fault point shown in Figure 1, the arc starting and arc extinction stages of the arc fault correspond to a narrow area near the current zero crossing point, while the arc burning stage corresponds to a wider area near the current peak. Generally speaking, the higher the power supply voltage, the more stable the arc burning. Although the actual load types in the low-voltage system are diverse, they are based on Figure 4 For the method of the present invention, the line load only includes line resistance and line inductance, so the present invention is not affected by the load type. The method can adapt to arc-type grounding fault detection under various load conditions.

[0071] Therefore, the sampling point voltage u s ( t ) is used as the center of the window for solving the equation, that is, u s ( t ) is taken as the center and the window data of a certain width on both sides are extracted. Figure 5 Schematic diagram of voltage and current data extraction for equation solving.

[0072] Based on the above principle, in the above step S1, the specific steps of extracting data for constructing the overdetermined equation are as follows.

[0073] Step S11, set the number of sampling points per cycle to N, and measure the voltage of the sampling point at the monitoring point in units of cycles. u s ( t ) and sampling point current i h ( t ) for synchronous sampling, and the sampled discrete signal is recorded as the sampling point voltage discrete signal u s ( n ) and the discrete current signal of the sampling point i h ( n ).

[0074] Step S12: Due to the uncertainty of the discrete signal of the sampling point voltage u s (n ) and the discrete current signal of the sampling point i h ( n ), therefore, the voltage discrete signal at the sampling point is calculated using the fast Fourier transform (FFT) u s ( n )'s initial phase angle of the fundamental component θ 1. The specific calculation method is:

[0075] First, the voltage discrete signal of the sampling point is obtained by fast Fourier transform u s ( n )'s real and imaginary parts of the fundamental component a , b , ; then through , calculate the initial phase angle θ 1.

[0076] Step S13, for the positive and negative half-waves, the peak voltage phases correspond to π / 2 and 3π / 2 respectively, so the initial phase angle is calculated. θ The angle deviations of 1, π / 2 and 3π / 2 can be used to know the interval ΔN between the initial sampling point of the current cycle and the peak voltage points of the positive and negative half-waves. The intervals ΔN1 and ΔN2 between the initial sampling point of the current cycle and the positive half-wave and negative half-wave can be calculated respectively. The calculation formula is:

[0077] Positive half wave: (6)

[0078] Negative half wave: (7)

[0079] In formula (6) and formula (7), “|…|” means taking the absolute value, and “mod |…| 2π " indicates taking the angle modulo. When calculating the sampling point interval ΔN, due to the discrete signal of the sampling point voltage u s ( n ) can be located in any quadrant between 0 and 2π, so 2π is introduced in equations (6) and (7) to ensure that the shift in each case is in the same direction.

[0080] Step S14: If the voltage discrete signal at the sampling point u s ( n ) is denoted as N 0, then the sampling points of the peak voltage points of the positive half-wave and the negative half-wave are:

[0081] Positive half wave: (8)

[0082] Negative half wave: (9)

[0083] In formula (8) and formula (9), represents the sampling point of the peak voltage point of the positive half-wave, The sampling point representing the peak voltage point of the negative half-wave.

[0084] Step S15, assuming that the width of the data selection window is set to M+1 sampling points, the voltage discrete signals of the sampling points are respectively u s ( n ) is the sampling point of the peak voltage point of the positive half wave , the sampling point of the peak voltage point of the negative half wave Select the center of the data window, select M / 2 sampling points on the left and right sides respectively, and complete the sampling point voltage discrete signal with a length of M+1 u s ( n ) in the data extraction. Preferably, the total time length corresponding to the data selection window can be controlled within 3ms~6ms.

[0085] Step S16, obtaining the current discrete signal of the sampling point i h ( n ), and the current discrete signal of the sampling point i h ( n ) to filter out the high-frequency components in the current and obtain a low-frequency current discrete signal i' h ( n ).

[0086] The present invention essentially utilizes low-frequency fault information for identification. Therefore, when the current discrete signal of the sampling point is i h ( n ) Before selection, it is necessary to pre-filter out the high-frequency components in the current sampled at the monitoring point to improve the recognition accuracy. The filtering process can use digital filtering methods such as smoothing filtering or FIR low-pass filtering.

[0087] The filtered sampling point current discrete signal i h ( n ) is recorded as the low-frequency current discrete signal i' h ( n ). It should be noted that if the filtering algorithm causes i' h ( n )and ih ( n ) has shifted, the phase of the filtered signal needs to be restored. For example, after the FIR filter filters the signal, a phase shift of K / 2 will occur, where K is the filter order. i' h ( n ) moves K / 2 points in the opposite direction to perform phase restoration.

[0088] Step S17, respectively using discrete signals of the sampling point voltage u s ( n ) is the sampling point of the peak voltage point of the positive half wave , the sampling point of the peak voltage point of the negative half wave Corresponding synchronous current sampling points , As a low-frequency current discrete signal i' h ( n ) of the positive half-wave data acquisition window and the negative half-wave data acquisition window, that is, the synchronous current sampling point , With M / 2 sampling points selected on the left and right sides, a low-frequency current discrete signal with a length of M+1 is completed. i' h ( n ) in the data extraction.

[0089] Step S18, according to the above formula (4), that is,

[0090] ,

[0091] use i' h (1) i' h (2) , ..., the value of , ,…….

[0092] So far, the above formula (3) requires i' h (1) i' h (2) ..., u s (1) u s (2), ..., and , , …and other data extraction completed.

[0093] In summary, the method of the present invention solves the protection problem of arc-type grounding fault in the TN-CS system in the scenario where the TN-S part lacks residual current protection, that is, the grounding fault occurs when the phase line is connected to the equipment housing or the phase line is connected to the PE line. The traditional overcurrent protection technology may not be able to perceive the arc-type grounding fault, and the present invention uses the common feature that the arc voltage at the fault point is approximately a DC voltage source during the arcing period to carry out fault identification, thereby solving the protection blind spot of the prior art.

[0094] In practical applications, the present invention can be implemented through intelligent circuit breakers, leakage protectors and smart meters before entering the home, to detect and warn of arc-type grounding faults occurring on the user side (i.e., the TN-S part) of the TN-CS system, thus making up for the protection blind spots of the prior art.

[0095] In a low-voltage AC power distribution system, for voltage signals, power quality issues will not introduce a DC voltage component into the AC voltage. The DC voltage component introduced by the arc at the fault point is a unique feature caused by the arc phenomenon. Therefore, the present invention is theoretically unique in identifying arc-type grounding faults.

[0096] Although the present invention has been disclosed as above by way of embodiments, it is not intended to limit the present invention. Any person having ordinary knowledge in the technical field may make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the scope of the attached patent application.

Claims

1. A method for detecting arc-type grounding faults in low-voltage distribution lines based on parameter identification, characterized in that: The low-voltage power distribution line is a TN-CS system, the monitoring point is set in the TN-S part of the TN-CS system, and the fault detection method includes the following steps: Step S1, setting a monitoring point on the low-voltage distribution line, synchronously sampling the voltage and current of the monitoring point, and extracting data for constructing an overdetermined equation; Step S2, based on the data, using half-wave as a unit, respectively using overdetermined equations to calculate the line resistance under positive half-wave and negative half-wave R 1. Inductor L 1. Voltage U h The value of , the overdetermined equation is, In the above formula, i h ( t ) represents the sampling point current, u s ( t ) represents the sampling point voltage (t=1,2,3…), R 1 represents the line resistance between the fault point and the monitoring point, L 1 represents the inductance between the fault point and the monitoring point, U h Represents the voltage of the equivalent DC power supply at the fault point, , , …is calculated as: In the formula, ΔT Indicates the sampling time interval; Step S3, judging the identification result of the ground fault in units of half-wave, and the judgment criterion is: In the formula, α Indicates the judgment threshold of arc burning voltage identification, The voltage of the equivalent DC power supply at the fault point representing the positive half-wave, The voltage of the equivalent DC power supply at the fault point representing the negative half-wave; In the above step S1, extracting data for constructing an overdetermined equation includes the following steps: Step S11, set the number of sampling points per cycle to N, and measure the voltage of the sampling point at the monitoring point in units of cycles. u s ( t ) and sampling point current i h ( t ) for synchronous sampling, and the sampled discrete signal is recorded as the sampling point voltage discrete signal u s ( n ) and the discrete current signal of the sampling point i h ( n ) ; Step S12, using fast Fourier transform to obtain the discrete voltage signal of the sampling point u s ( n )'s real and imaginary parts of the fundamental component a , b , , and then through , calculate the initial phase angle θ 1; Step S13, respectively calculating the interval ΔN1 between the initial sampling point of the current cycle and the positive half-wave and the interval ΔN2 between the initial sampling point of the current cycle and the negative half-wave, the calculation formula is: Positive half wave: Negative half wave: In the formula, "|…|" means taking the absolute value, "mod |…| 2π " means taking the modulus of the angle; Step S14: If the voltage discrete signal at the sampling point u s ( n ) is denoted as N 0, then the sampling points of the peak voltage points of the positive half-wave and the negative half-wave are: Positive half wave: Negative half wave: In the formula, represents the sampling point of the peak voltage point of the positive half-wave, The sampling point representing the peak voltage point of the negative half-wave; Step S15, assuming that the width of the data selection window is set to M+1 sampling points, the voltage discrete signals of the sampling points are respectively u s ( n ) is the sampling point of the peak voltage point of the positive half wave , the sampling point of the peak voltage point of the negative half wave Select the center of the data window, select M / 2 sampling points on the left and right sides respectively, and complete the sampling point voltage discrete signal with a length of M+1 u s ( n ) in data extraction; Step S16, obtaining the discrete current signal of the sampling point i h ( n ), and the current discrete signal of the sampling point i h ( n ) to filter out the high-frequency components in the current and obtain a low-frequency current discrete signal i' h ( n ); Step S17, respectively using discrete signals of the sampling point voltage u s ( n ) is the sampling point of the peak voltage point of the positive half wave , the sampling point of the peak voltage point of the negative half wave Corresponding synchronous current sampling points , As a low-frequency current discrete signal i' h ( n ) is the center of the positive half-wave data acquisition window and the negative half-wave data acquisition window, and M / 2 sampling points are selected on the left and right sides to complete the low-frequency current discrete signal with a length of M+1. i' h ( n ) in data extraction; Step S18, according to the following formula, using i' h (1) i' h (2) , ..., the value of , ,……, 。 2. The method for detecting arc-type ground faults in low-voltage power distribution lines based on parameter identification according to claim 1, characterized in that: In step S16, the filtering process adopts smoothing filtering or FIR low-pass filtering.

3. The method for detecting arc-type ground faults in low-voltage power distribution lines based on parameter identification as claimed in claim 2, characterized in that: If the filtering algorithm results in i' h ( n )and i h ( n ) has moved, the phase of the filtered signal needs to be restored.

4. The method for detecting arc-type ground faults in low-voltage power distribution lines based on parameter identification according to claim 1, characterized in that: The total time length corresponding to the data selection window is 3ms~6ms.

5. The method for detecting arc-type ground faults in low-voltage power distribution lines based on parameter identification according to claim 1, characterized in that: The judgment threshold of arc burning voltage identification α is 12V.

Images