Flexible grounding system high-resistance fault line selection method and system considering DG injection harmonic waves

Abstract

The invention discloses a flexible grounding system high-resistance fault line selection method and system considering DG injection harmonic waves, and the method comprises the following steps: judging the exit condition of a small resistor according to the change of the zero-sequence voltage of a bus in a flexible grounding system; if the condition that the small resistor quits operation occurs, sampling a zero-sequence current sequence, and performing pairwise cross wavelet transform to obtain a high-correlation time-frequency region; judging whether the high-correlation time-frequency region is a DG harmonic frequency band or not, determining an adaptive time-frequency window, solving a phase difference obtained after cross wavelet transformation among all feeder lines, and calculating a phase difference vector; and setting a phase difference threshold value, comparing the phase difference vector with the phase difference threshold value, and judging a fault line selection result based on a comparison result.

Description

Technical Field

[0001] The present invention belongs to the technical field of relay protection fault line selection, and particularly relates to a method and system for high-resistance fault line selection in a flexible grounding system considering DG-injected harmonics. Background Technique

[0002] With the development of the power system, the operating environment of the distribution network has become increasingly complex. High-resistance single-phase grounding faults have gradually become one of the more concealed and challenging fault types in the distribution network. Especially in a flexible grounding distribution network, due to the lack of research on the characteristic mechanism of distributed power source access, the applicability of traditional fault location methods is not strong, and the detection and location of high-resistance single-phase grounding faults have become more complex.

[0003] Especially in a flexible grounding system, after a high-resistance grounding fault occurs in the flexible grounding system, there is a problem of refusal to operate in the parallel stage of the neutral point small resistor, and the harmonics injected by inverter-type distributed power sources are likely to affect the existing transient line selection methods, bringing new challenges to the safe operation of the distribution system and the identification of the fault area. Therefore, considering the problem of refusal to operate under high-resistance faults in the flexible grounding system and the harmonic interference of DG injection is a hot issue that must be considered today. Summary of the Invention

[0004] The present invention aims to solve the deficiencies of the prior art and provides the following solutions:

[0005] A method for high-resistance fault line selection in a flexible grounding system considering DG-injected harmonics, comprising the following steps:

[0006] Judge the withdrawal situation of the small resistor according to the change of the zero-sequence voltage of the bus in the flexible grounding system;

[0007] If the situation of the small resistor exiting the operation occurs, sample the zero-sequence current sequence and perform pairwise cross-wavelet transform to obtain a high-correlation time-frequency region;

[0008] Judge whether the high-correlation time-frequency region is the DG harmonic frequency band, determine the adaptive time-frequency window, obtain the phase difference after cross-wavelet transform between all feeders, and calculate the phase difference vector;

[0009] Set a phase difference threshold, compare the phase difference vector with the phase difference threshold, and judge the fault line selection result based on the comparison result.

[0010] Preferably, the method for judging the withdrawal situation of the small resistor includes:

[0011] Obtain the zero-sequence voltage u of the neutral point in the first stage of the flexible grounding system 01 and the zero-sequence voltage u of the neutral point in the second stage 02 ;

[0012] When u appears02 > 4.7%u 01 When it is, it is determined that the small resistance in the second stage exits operation.

[0013] Preferably, the method for sampling the zero-sequence current sequence includes:

[0014] Based on the edge effect of wavelet transform, obtain the zero-mode current sampling sequence i of all lines within the time range of 30 ms before refusal to operate and 40 ms after refusal to operate k (n), where n is the sampling serial number and k is the line number.

[0015] Preferably, the method for calculating the phase difference vector includes:

[0016] Determine the average phase difference between feeders according to the adaptive time window to obtain the phase difference matrix

[0017]

[0018] Among them, Indicates the phase difference between line k and line k';

[0019] Let Be the phase difference matrix The maximum value element in, then the phase difference coefficient DIF between line k and all other feeders k Is:

[0020]

[0021] Among them, N represents a natural number;

[0022] Obtain the phase difference vector based on the phase difference coefficient:

[0023] DIF = [DIF1 DIF2 … DIF N .

[0024] Preferably, the method for obtaining the fault line selection result includes:

[0025] Set the phase difference threshold D set , obtain the maximum value DIF of the phase difference vector max ;

[0026] Based on the phase difference threshold D set And the maximum value DIF of the phase difference vector max , obtain the fault line selection result:

[0027] When DIF max < D set When it is, it is determined that a bus fault has occurred;

[0028] When DIFmax ≥ D set When it is, it is determined as the corresponding DIF max A fault occurs in the feeder line.

[0029] The present invention also provides a high-resistance fault line selection system for a flexible grounding system considering DG-injected harmonics. The system applies the method described in any one of the above, and includes: a small resistance judgment module, a wavelet transform module, a phase difference calculation module, and a fault judgment module;

[0030] The small resistance judgment module judges the withdrawal situation of the small resistance according to the change of the zero-sequence voltage of the bus in the flexible grounding system;

[0031] If the situation of the small resistance exiting the operation occurs, the wavelet transform module samples the zero-sequence current sequence and performs pairwise cross-wavelet transform to obtain a high-correlation time-frequency region;

[0032] The phase difference calculation module is used to judge whether the high-correlation time-frequency region is the DG harmonic frequency band, determine an adaptive time-frequency window, obtain the phase difference after cross-wavelet transform between all feeders, and calculate the phase difference vector;

[0033] The fault judgment module is used to set a phase difference threshold, compare the phase difference vector with the phase difference threshold, and judge the fault line selection result based on the comparison result.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] Based on the disturbance generated by the withdrawal of the small resistance in parallel with the neutral point, the present invention realizes the identification of the fault feeder even when the small resistance in parallel with the neutral point exits the operation while excluding the interference of DG-injected harmonics. By establishing a flexible grounding distribution network model, the reason for the withdrawal of the small resistance in parallel with the neutral point under high-resistance faults is revealed, and according to the phase relationship between the fault feeder, the healthy feeder and the neutral point voltage after the small resistance exits, the phase difference characteristics between the fault feeder and the healthy feeder under high-resistance faults are discovered, and the influence of DG-injected harmonics on the traditional transient line selection method of the distribution network is clarified. By using the cross-wavelet transform (XWT) method, the advantages of complex wavelet transform in frequency band analysis and real wavelet in mutation detection are fully combined, the high-correlation time-frequency window corresponding to the transient main resonance process is obtained, the accurate perception of high-resistance faults is realized, and the accurate identification of the fault feeder is completed. Description of the Drawings

[0036] To more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the embodiments are briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other accompanying drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 Schematic diagram of the method flow of the embodiment of the present invention;

[0038] Figure 2 Block diagram of the method flow of the embodiment of the present invention;

[0039] Figure 3 Equivalent zero-sequence model when the neutral point small resistance of the embodiment of the present invention is withdrawn;

[0040] Figure 4 Projection of the healthy feeder and the faulty feeder on the neutral point voltage in the embodiment of the present invention;

[0041] Figure 5 Phase change diagram of the embodiment of the present invention, where a is the undisturbed sequence and b is the disturbed sequence;

[0042] Figure 6 Schematic diagram of the simulation structure and fault point setting of the 10kV flexible grounding distribution network power distribution system in the embodiment of the present invention;

[0043] Figure 7 Schematic diagram of the zero-sequence current at the head of the feeder in the embodiment of the present invention;

[0044] Figure 8 Schematic diagram of the cross-wavelet transform result of l1 and l2 in the embodiment of the present invention;

[0045] Figure 9 Schematic diagram of the cross-wavelet transform result of l1 and l3 in the embodiment of the present invention;

[0046] Figure 10 Schematic diagram of the cross-wavelet power spectrum of l2 and l3 in the embodiment of the present invention;

[0047] Figure 11 Schematic diagram of the Emanuel arc model in the embodiment of the present invention;

[0048] Figure 12 Schematic diagram of the current at the head of the faulty feeder in the embodiment of the present invention;

[0049] Figure 13 Schematic diagram of the current at the head of the non-faulty feeder in the embodiment of the present invention. Detailed implementation manners

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Before describing the embodiments of the present invention, first, a feature analysis of the phase difference of the faulty zero-sequence current is carried out:

[0053] The simplified zero-sequence model of single-phase grounding when the small resistor is withdrawn is as Figure 3 shown. When a high-resistance grounding fault occurs in the system, the neutral small resistor closes, triggering the protection action and causing the withdrawal. The grounding mode of the system is converted to resonant grounding, and its transient process originates from the parallel resonance interaction between the equivalent inductance L p of the arc suppression coil and the capacitance C 0Σ to the ground. From Figure 3 the transient equation for the withdrawal of the small resistor can be listed:

[0054]

[0055] In the formula, for simplicity of calculation, R = 3R f , i 0Lp represents the current flowing through the arc suppression coil, represents the virtual power supply at the fault point, U ph represents the amplitude of the fault-phase voltage, ω represents the power frequency, represents the fault initial phase angle (the phase angle of the fault-phase voltage at the fault moment).

[0056] The damping characteristics of the system can be divided into two states: underdamping and overdamping according to the value of the transition resistance. When the transition resistance exceeds 1000 Ω, both the zero-sequence voltage and the zero-sequence current at the neutral point show significant attenuation characteristics, and the system is in the underdamping state at this time. Based on this characteristic, the underdamping transient process of the flexible grounding system after the withdrawal of the neutral small resistor (i.e., in the high-resistance state) will be mainly studied.

[0057] When the transition resistance R satisfies:

[0058]

[0059] the system is in the underdamping state. At this time, the transient component of the zero-sequence current flowing through the arc suppression coil is:

[0060] i 0L_T = e -δt[A1cos(ω f t)+A2sin(ω f t)] (3)

[0061] where: B represents the amplitude of the power frequency component, ω f represents the main resonance frequency, and δ represents the attenuation factor.

[0062] Furthermore, the transient component of the zero-sequence voltage of the bus can be obtained:

[0063]

[0064] The transient current of the capacitance to the ground of each sound feeder is:

[0065]

[0066] The transient component of the zero-sequence current at the fault point can be expressed as:

[0067]

[0068] The zero-sequence current at the outlet of the faulty line is:

[0069]

[0070] According to the theoretical analysis of Equation (5), the amplitude of the transient zero-sequence current of the sound feeder is positively correlated with its capacitance to the ground and the transient voltage of the neutral point, and the transient components of each sound feeder have the same-phase characteristic. For the transient phase relationship between the faulty feeder and the sound feeder, Equations (5) and (7) do not explicitly represent their mathematical correlation. Theoretical derivation shows that the transient zero-sequence currents of the two types of feeders can be decomposed into projection components in the direction of the neutral point voltage. By comparing the difference in the projection values of the transient currents of the faulty feeder and the sound feeder in the direction of the transient voltage of the neutral point, the phase offset representing the fault characteristics can be effectively extracted.

[0071] From the transient current of the capacitance to the ground of each sound feeder, it can be seen that the transient current of the sound feeder is proportional to the derivative of the transient voltage of the neutral point, and since the transient zero-sequence voltage is a decaying sine function, the transient current and the transient voltage are not orthogonal, and the projection of the former on the latter is not zero, as shown in Equation (8):

[0072]

[0073] It can be proved that the projection of the transient current of the sound line on the transient voltage is a very small negative value. The two show an approximately orthogonal relationship. According to Equations (4) and (6), the transient zero-sequence current at the fault point is proportional to the transient zero-sequence voltage of the neutral point, and the proportionality factor is the transition resistance at the fault point, and the two are in the same phase, as shown in Equation (9):

[0074]

[0075] Equation (9) shows that the zero-sequence transient current at the outlet of the faulty feeder consists of the line's intrinsic capacitance component and the transient component i 0f_T at the fault point. Due to the characteristic relationship, the zero-sequence current at the outlet of the faulty feeder in Equation (10) can be equivalently simplified to the zero-sequence transient current at the fault point.

[0076]

[0077] The projection coefficient of the transient zero-sequence current of the faulty line on the transient zero-sequence voltage is:

[0078]

[0079] And there is a relationship:

[0080]

[0081] It can be seen that: the projection coefficient of the transient zero-sequence current of the faulty feeder on the neutral point voltage is opposite in polarity to that of the healthy feeder, and its amplitude is also much larger than the latter. The phase difference between the faulty feeder and the healthy feeder is as Figure 4 shown.

[0082] According to Equations (3) to (7), the main resonance frequency precisely corresponds to the intrinsic frequency of the transient component excited when the small neutral resistor is withdrawn. This frequency-domain characteristic completely characterizes the dynamic response characteristics of the fault current. It can be observed that the phase difference between non-faulty feeders is extremely small and approximately in-phase, while the phase difference between the faulty feeder and the healthy feeder is obvious and exceeds 90°. Therefore, the phase difference characteristics between the faulty feeder and the non-faulty feeder after the withdrawal of the small neutral resistor can be used for line selection.

[0083] Influence of DG-injected harmonics on the transient characteristics of zero-sequence current:

[0084] Inverter-type DGs based on pulse rectifiers generally generate harmonics of order 6m ± 1 (m = 1, 2,...), among which the 5th and 7th harmonics are the most significant. Let the DG be connected to line n, and the mth unbalanced harmonic zero-sequence component at the injection point, and the mth harmonic at the head of non-DG-injected feeder k is:

[0085] i 0km = [Z n (ω m ) + Z DG i 0m / Z k (ω m ) (13)

[0086] Among them, ω m represents the mth harmonic frequency, and Z k(ω m ) represents the input impedance of the non-DG connected line k, Z n (ω m ) represents the input impedance of line n from the DG connection point to the bus, Z DG represents the equivalent output impedance of the DG.

[0087] After the DG injects unbalanced harmonic current into the system, the zero-sequence current detected at the head of each feeder is superimposed with i 0km , i 0km is the mth DG harmonic at the head of feeder k. The zero-sequence current of the faulty feeder detected at the head of the feeder is the zero-sequence current after the DG harmonic is superimposed, as shown in Equation (14).

[0088] i′ 0k = i 0k + i 0km (14)

[0089] Since the phase of each i 0km is affected by the position of the DG connection and remains fixed, not affected by the transient to steady-state transition during the withdrawal of the small resistance at the neutral point, is non-time-varying, and as can be seen from Equation (13), the magnitude relationship of each i 0km depends on the signal frequency and line parameters, and for i 0k related to the transient process of single-phase grounding fault, its phase depends on the location where the fault occurs.

[0090] Therefore, when the faulty feeder and the DG connected line are not on the same line, there is an obvious phase difference between the main resonance component detected at the head of the faulty feeder and the non-faulty feeder, and the harmonic component detected at the head of the DG connected line has the largest energy among all lines and the polarity is opposite to that of other lines. If the main resonance frequency of the feeder does not coincide with the DG harmonic frequency, the polarity relationship of the DG harmonic frequency band in different lines may interfere with the phase difference of the lines in the main resonance frequency band, affecting the method of fault line selection using the phase difference feature.

[0091] Embodiment 1

[0092] In this embodiment, as Figure 1 , Figure 2 shown, a method for high-resistance fault line selection in a flexible grounding system considering DG injected harmonics includes the following steps:

[0093] S1. According to the change of the zero-sequence voltage of the bus in the flexible grounding system, judge the withdrawal situation of the small resistance.

[0094] After a high-resistance fault occurs, the flexible grounding system enters the second stage, and the zero-sequence voltage of the bus drops to 1.7% - 4.7% before the small resistor is put into operation. However, after the small resistor is put into and then taken out of operation in the second stage, the bus voltage will rise sharply. Therefore, the method for judging the disconnection of the small resistor includes: obtaining the zero-sequence voltage u of the neutral point in the first stage of the flexible grounding system 01 and the zero-sequence voltage u of the neutral point in the second stage 02 ; when u 02 > 4.7%u 01 occurs, it is determined that the small resistor in the second stage is taken out of operation.

[0095] S2. If the small resistor is taken out of operation, sample the zero-sequence current sequence and perform pairwise cross-wavelet transform to obtain the high-correlation time-frequency region.

[0096] The method for sampling the zero-sequence current sequence includes: based on the edge effect of wavelet transform, obtaining the zero-mode current sampling sequence i k (n) of all lines within the time range of 30 ms before refusal to operate and 40 ms after refusal to operate, where n is the sampling serial number and k is the line number.

[0097] When the small resistor at the neutral point of the flexible grounding system is taken out of operation, there is a significant phase difference in the main resonance components of the zero-sequence current between the faulty feeder and the healthy feeder, while the phase consistency among the healthy feeders is relatively high. The fault line selection can be carried out by using the phase difference characteristics between the feeders. However, the DG harmonic injection will interfere with the phase characteristics of the main resonance components, resulting in a decrease in the reliability of the traditional line selection method.

[0098] Therefore, in this embodiment, the cross-wavelet transform (XWT) is used to characterize the time-frequency correlation and phase relationship of the feeder zero-sequence current. This method effectively avoids the problems of mode mixing and over-decomposition existing in the traditional method by quantifying the correlation degree of the two signals in the time domain. Specifically, by jointly analyzing the time-frequency domain correlation of non-stationary signals, the high-correlation time-frequency window (HRW) corresponding to the disconnection of the small resistor is extracted.

[0099] The high-correlation time-frequency window is the high-correlation overlapping region after the cross-wavelet transform of the selected line and the reference line. This region usually contains one or more non-connected time-frequency sets. Usually, the overlapping region containing the highest frequency peak is defined as the adaptive time-frequency window. However, the adaptive time-frequency window is usually the overlapping region of the main resonance frequency band and the harmonic frequency band. In this embodiment, the DG harmonic interference will be removed by using the stationarity test.

[0100] When the main resonance frequency and the harmonic frequency are in different frequency bands, as shown in Figure 5 (a), the phase of the harmonic frequency band signal does not change significantly before and after the small resistor of the system is taken out of operation. This phenomenon indicates that the harmonic component is not interfered by the transient process. However, when the main resonance frequency and the harmonic frequency are in the same frequency band, as shown in Figure 5(As shown in (b), the transient process caused by the withdrawal of the neutral - point small resistor will significantly interfere with the harmonic frequency band, resulting in abnormal phase - shift characteristics before and after the switching. Define the average phase difference between the candidate line k' and line k within the HRW as shown in Equation (15).

[0101]

[0102] Where, N HRW represents the total number of points within the HRW, and (s, τ) represents the time period.

[0103] If the target time - frequency window falls within the frequency band where the DG - injected harmonics are located (the frequency bands of the 5th and 7th harmonics), it is necessary to intercept the phase sequences 5 ms long before and after the fault in the harmonic frequency band and and perform stationarity verification according to Equation (16).

[0104]

[0105] Where, σ1 and σ2 respectively represent and the standard deviations of.

[0106] When ρ < 3, it is determined that the phase sequence has not been disturbed after the withdrawal of the neutral - point small resistor, and this high - correlation time - frequency region is discarded. Without considering this high - correlation time - frequency region, re - determine the time - frequency set with the highest peak as the adaptive time - frequency window. When ρ > 3, it is determined that the phase sequence has been disturbed after the withdrawal of the neutral - point small resistor, and this high - correlation time - frequency region is the adaptive time window.

[0107] S3. Determine whether the high - correlation time - frequency region is the DG harmonic frequency band, and determine the adaptive time - frequency window, calculate the phase difference obtained after the cross - wavelet transform between all feeders, and calculate the phase - difference vector.

[0108] The method for calculating the phase - difference vector includes: determining the average phase difference between feeders according to the adaptive time window to obtain the phase - difference matrix

[0109]

[0110] Where, represents the phase difference between line k and line k'; let be the element with the largest value in the phase - difference matrix then the phase - difference coefficient DIF between line k and all the other feeders k is:

[0111]

[0112] Where, N represents a natural number; when feeder k is the fault feeder, close to the value by N - 1 times, DIF k close to the value value. When the feeder k is a sound feeder, close to the value DIF k is close to 0. Based on the phase difference coefficient, a phase difference vector is obtained:

[0113] DIF = [DIF1 DIF2 … DIF N (19).

[0114] When a certain DIF k is significantly greater than the phase difference coefficients between the remaining feeders, then the feeder k is a faulty feeder.

[0115] S4. Set a phase difference threshold, and compare the phase difference vector with the phase difference threshold. Based on the comparison result, judge the faulty line selection result.

[0116] The method for obtaining the faulty line selection result includes: setting a phase difference threshold D set , and obtaining the maximum value DIF of the phase difference vector max ; based on the phase difference threshold D set and the maximum value DIF of the phase difference vector max , obtain the faulty line selection result:

[0117] When DIF max < D set , it is determined that a bus fault has occurred;

[0118] When DIF max ≥ D set , it is determined that the feeder corresponding to DIF max has a fault.

[0119] Embodiment 2

[0120] In this embodiment, a high - resistance fault line selection system for a flexible grounding system considering DG - injected harmonics includes: a small - resistance judgment module, a wavelet transform module, a phase difference calculation module, and a fault judgment module.

[0121] The small - resistance judgment module judges the withdrawal situation of the small resistance according to the change of the zero - sequence voltage of the bus in the flexible grounding system.

[0122] If the situation of the small resistance exiting the operation occurs, the wavelet transform module samples the zero - sequence current sequence and performs pairwise cross - wavelet transform to obtain a high - correlation time - frequency region.

[0123] The phase difference calculation module is used to determine whether the high-correlation time-frequency region is the DG harmonic frequency band, determine the adaptive time-frequency window, obtain the phase difference after cross-wavelet transform between all feeders, and calculate the phase difference vector.

[0124] The fault judgment module is used to set the phase difference threshold, compare the size of the phase difference vector with the phase difference threshold, and judge the fault line selection result based on the comparison result.

[0125] Embodiment III

[0126] In this embodiment, to verify the effectiveness of the method in this paper for fault line selection and section location when the neutral point shunt small resistor is taken out of operation during a single-phase grounding fault in a flexible grounding distribution network, a 10 kV distribution system with inverter-type DG is built in the MATLB / Simulink simulation environment, as Figure 6 shown, and the specific parameters are shown in Table 1.

[0127] Table 1

[0128]

[0129] I. Algorithm reliability analysis

[0130] To verify the effectiveness of the line selection method in this paper when the neutral point small resistor is taken out, the interference of different fault feeders, transition resistors, and fault initial phase angles will be considered to verify the method in this paper. Taking the fault of feeder l22 as an example, the transition resistor is 500 Ω, and the neutral point small resistor is taken out of operation at 0.1 s. The time-domain waveform diagram of the sampling signal is as Figure 7 shown. It can be seen from Figure 7 that the zero-sequence current at the head of each feeder is significantly affected by harmonic interference. The cross-wavelet transform is performed pairwise on the zero-sequence currents at the head of the feeders. Due to space limitations, only the cross-wavelet power spectra of l1 and l2 and l2 and l3 are shown in this paper, as Figure 8 and Figure 9 shown. In the figure, the vertical coordinate represents the scale coefficient j, and the corresponding frequency is f s / 2 j , and f s represents the sampling frequency.

[0131] In Figure 8 and Figure 9 , the black envelope is the time-frequency window with the highest correlation between the currents of the two feeders and suitable for line selection. The black arrow represents the phase difference between the two signals. When the phase differences are 0°, 90°, 180°, and 270° respectively, the arrow directions are horizontally to the right, vertically downward, horizontally to the left, and vertically upward. In this embodiment, for the convenience of subsequent phase comparison, the phase difference is converted from 0° to 360° to -180° to 180°, and the absolute value is taken.

[0132] In the edge calculation problem of wavelet transform, as the scale increases (frequency decreases), the time length required for signal analysis will also increase accordingly. Conversely, when the scale decreases, the required time length becomes shorter. Therefore, signals near the time window edge in the low-frequency region (i.e., high scale) cannot be accurately captured and analyzed by wavelet transform. This effect usually appears as a "U-shaped" structure in the time-scale coefficient spectrum. This "U-shaped" envelope is called the Cone of Influence (COI), which defines the range where wavelet transform cannot perform accurate calculations in the time window edge region.

[0133] From Figure 8 and Figure 9 it can be seen that the high-correlation time-frequency regions are respectively located in the frequency bands with scale factors of 3 - 5 and 16 and above. First, perform a stationarity check on the frequency band with a scale factor of 3 - 5. The scale factors 3 - 5 correspond to the frequency bands where the 5th and 7th harmonics are located (DG harmonic frequencies). After performing a stationarity check on the frequency band with a scale factor of 3 - 5, ρ = 0.231 is obtained. Therefore, the region with a scale factor of 3 - 5 is excluded. The frequency band with a scale factor of 16 and above is determined as the adaptive time-frequency window, and the time window is [100.1, 123] ms. The phase differences between l1 and l2 and between l1 and l3 within the adaptive time-frequency window are 110.31° and 2.37° respectively. According to the calculation method of the phase difference coefficient: DIF = [1.62 109.28 2.43 1.72 2.57], DIF max = 109.28° > DIF set = 15°. Feeder l2 is the faulty feeder, and the line selection is correct.

[0134] To further verify the test method of the present invention, different faulty feeders, fault distances D, and transition resistances R are respectively considered f to verify the method of the present invention. It can be seen from Table 2 that the method of the present invention is still effective when the grounding resistance is 3000Ω.

[0135] Table 2

[0136]

[0137] Example 4

[0138] In this example, to verify the tolerance ability of the line selection method to noise interference, this example takes a single-phase grounding fault with a fault at 2 km from the busbar of feeder l 32 , adding white noise with a signal-to-noise ratio SNR of 60 dB, and a transition resistance of 1000Ω as an example. The cross-wavelet power spectrum is as Figure 10 shown.

[0139] From Figure 10It can be seen that white noise appears in the time scale range of 0.125 - 0.25, while the DG harmonic frequency band and the main resonance frequency band of the feeder appear in a higher time scale range and are not affected by white noise. Next, different faulty feeders, fault distance D, transition resistance R f , and signal-to-noise ratio SNR will be considered to verify the method of this paper.

[0140] Table 3

[0141]

[0142] As can be seen from Table 3, after the sampling signal is superimposed with noise, the DIF value of the faulty feeder is still much larger than that of the healthy feeder, and the method of the present invention can still correctly select the line and has a certain anti-noise ability.

[0143] To verify the applicability of the invention in non-linear arcs, the Emanuel arc model is used in this embodiment to simulate arc faults, and the arc model is as Figure 11 shown. The model circuit consists of two DC power supplies (V p , V n ) and two diodes (VD p , VD n ). V p and V n are used to simulate the arc voltage, and their values change according to the voltage level of the system and the requirements of asymmetric modeling, and these voltage values fluctuate randomly and independently with time. In addition, by adjusting the different values of resistors R p and R n , the model can simulate different asymmetries of the arc current. Three different arc parameter configurations are shown in Table 4.

[0144] Table 4

[0145]

[0146] From Figure 12 and Figure 13 in the zero-sequence currents at the head of the faulty feeder in the three cases, it can be seen that there are obvious differences in the arc extinction times of the three current waveforms, and after the small neutral resistance is withdrawn, obvious non-linear distortions occur in the feeder current. In this embodiment, the effectiveness of the line selection method will be verified based on the three arc cases, and the judgment results are shown in Table 5.

[0147] Table 5

[0148]

[0149] As can be seen from Table 5, during arc faults, the method of the present invention can still accurately determine the faulty feeder.

[0150] The embodiments described above are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for selecting high-resistance fault line in a flexible grounding system considering DG-injected harmonics, characterized in that: The following steps are involved: According to the change of busbar zero-sequence voltage in the flexible grounding system, the exit status of small resistors can be judged; If a small resistance exits operation, the zero-sequence current sequence is sampled and a pairwise cross wavelet transform is performed to obtain a high-correlation time-frequency region; Determine whether the high-correlation time-frequency region is a DG harmonic frequency band, determine an adaptive time-frequency window, obtain the phase difference obtained after cross-wavelet transform between all feeders, and calculate the phase difference vector; A phase difference threshold is set, and the magnitude of the phase difference vector is compared with the phase difference threshold, and a fault line selection result is determined based on the comparison result.

2. According to claim 1, a method for selecting high-resistance fault line in a flexible grounding system considering DG-injected harmonics is characterized in that: Methods for determining the exit status of small resistors include: Get the first stage neutral point zero sequence voltage u in the flexible grounding system 01 、The second stage neutral point zero sequence voltage u 02 ; When u appears 02 >4.7%u 01 When , it is determined that the small resistance of the second stage has exited operation.

3. According to claim 1, a method for selecting high-resistance fault line in a flexible grounding system considering DG injected harmonics is characterized in that: Methods for sampling zero-sequence current sequence include: Based on the edge effect of wavelet transform, the zero-mode current sampling sequence i of all lines within the time range of 30ms before and 40ms after the refusal to operate is obtained. k (n), n is the sampling number, k is the line number.

4. According to claim 1, a method for selecting high-resistance fault line in a flexible grounding system considering DG injected harmonics is characterized in that: Methods for calculating the phase difference vector include: Determine the average phase difference between feeders according to the adaptive time window and obtain the phase difference matrix in, represents the phase difference between line k and line k'; set up is the phase difference matrix The largest element in the value, then the phase difference coefficient DIF between line k and all other feeders k for: Wherein, N represents a natural number; The phase difference vector is obtained based on the phase difference coefficient: DIF=[DIF1DIF2…DIF N ]。 5. According to claim 4, a method for selecting high-resistance fault line in a flexible grounding system considering DG injected harmonics is characterized in that: Methods for obtaining fault line selection results include: Set the phase difference threshold D set , obtain the maximum value DIF of the phase difference vector max ; Based on the phase difference threshold D set and the maximum value of the phase difference vector DIF max , and obtain the fault line selection result: When DIF max <D set When , it is determined that the busbar is faulty; When DIF max ≥D set When , it is determined to correspond to DIF max The feeder line fails.

6. A high-resistance fault line selection system for a flexible grounding system considering DG-injected harmonics, the system applying the method described in any one of claims 1 to 5, characterized in that: include: Small resistance judgment module, wavelet transformation module, phase difference calculation module and fault judgment module; The small resistance judgment module judges the exit status of the small resistance according to the change of the bus zero-sequence voltage in the flexible grounding system; If a small resistance exits operation, the wavelet transform module samples the zero-sequence current sequence and performs a pairwise cross wavelet transform to obtain a high-correlation time-frequency region; The phase difference calculation module is used to determine whether the high-correlation time-frequency region is a DG harmonic frequency band, determine the adaptive time-frequency window, obtain the phase difference obtained after the cross-wavelet transform between all feeders, and calculate the phase difference vector; The fault judgment module is used to set a phase difference threshold, compare the phase difference vector with the phase difference threshold, and judge the fault line selection result based on the comparison result.

Images