Method for improving sensitivity of zero sequence voltage variation detection in low resistance grounded system

Abstract

The present application relates to a kind of methods for improving the sensitivity detection of small resistance grounding system zero sequence voltage variation, including based on the zero sequence voltage and zero sequence current collected, it includes the following steps: setting the sampling point number of each cycle zero sequence voltage, obtains corresponding original zero sequence voltage recording data stream;Original zero sequence voltage recording data stream is converted into weighted cumulative differential zero sequence voltage sequence;Waveform mutation detection is carried out to the obtained weighted cumulative differential zero sequence voltage sequence, and mutation time time label is obtained;According to time label, obtain the recording data of all installation points of zero sequence current in this cycle at the same time;The correlation coefficient of the recording sequence of each installation point zero sequence current in this cycle and the data sequence of weighted cumulative differential zero sequence voltage in this cycle is calculated, and fault discrimination is carried out.Thereby, the recognition degree and anti-interference ability of zero sequence current under high resistance grounding fault can be improved using weighted integration.

Description

Technical Field

[0001] This invention relates to a voltage change detection method, and more particularly to a method for improving the sensitivity of zero-sequence voltage change detection in a low-resistance grounding system. Background Technology

[0002] Urban power distribution networks increasingly utilize cable lines, which experience large short-circuit currents during single-phase ground faults. To limit overvoltage levels and quickly disconnect faulty lines, low-resistance grounding is becoming more widely used in cable lines. Low-resistance grounding typically employs zero-sequence overcurrent protection, whose setting needs to avoid the ground capacitance current flowing through the lines within the protection zone when a metallic ground fault occurs on lines outside the protection zone. Because the ground capacitance current of cable lines is relatively large, the setting is generally high, for example, 60A in some areas.

[0003] However, due to factors such as the natural environment, high-resistance grounding faults of non-ideal conductors often occur in the distribution network. At this time, the fault current is less than the zero-sequence current protection setting value, causing the protection to fail to operate. The fault is not easy to be cleared and may also cause serious faults such as fire at the fault point, injury or death to people and animals.

[0004] For 10kV distribution lines, a simple estimate can be made. If the zero-sequence overcurrent protection setting is 60A, the corresponding grounding resistance is no more than 90 ohms. For high-resistance grounding systems exceeding 90 ohms, it will not operate. Similarly, for low-resistance grounding systems, because...

[0005] When the neutral point grounding resistance is low, the zero-sequence voltage at the neutral point is very small in high-resistance grounding. Directly using zero-sequence voltage activation to improve sensitivity has limited effect due to a dead zone. For example, using the commonly used zero-sequence voltage of 15V as the activation threshold in low-current grounding systems, it can only detect single-phase grounding faults with a grounding resistance not exceeding 100 ohms in low-resistance grounding systems. As analyzed above, for low-resistance grounding systems, simply using zero-sequence current overcurrent protection cannot detect high-resistance faults. Similarly, using a simple zero-sequence voltage over-limit algorithm also results in a high-resistance dead zone, preventing successful detection and location of high-resistance faults.

[0006] In view of the above-mentioned shortcomings, the designer has actively researched and innovated in order to create a method to improve the sensitivity of zero-sequence voltage change detection in low-resistance grounding systems, making it more valuable for industrial applications. Summary of the Invention

[0007] To address the aforementioned technical problems, the purpose of this invention is to provide a method for improving the sensitivity of zero-sequence voltage change detection in low-resistance grounding systems.

[0008] The present invention provides a method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, comprising the following steps based on the acquired zero-sequence voltage and zero-sequence current:

[0009] Step 1: Set the number of sampling points for each cycle of zero-sequence voltage and obtain the corresponding raw zero-sequence voltage waveform data stream;

[0010] Step 2: Convert the original zero-sequence voltage waveform data stream into a weighted, accumulated differential zero-sequence voltage sequence;

[0011] Step 3: Perform waveform abrupt change detection on the obtained weighted cumulative differential zero-sequence voltage sequence to obtain the time label of the abrupt change moment;

[0012] Step 4: Obtain the waveform data of the zero-sequence current of all installation points at the same time for that cycle based on the time stamp of the abrupt change.

[0013] Step 5: Calculate the correlation coefficient between the zero-sequence current recording sequence of each installation point and the weighted cumulative differential zero-sequence voltage data sequence of that cycle, and then use this correlation coefficient to identify the fault.

[0014] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, the zero-sequence voltage is obtained through monitoring data at the installation points, and it originates from the bus or from a certain point on the line; the zero-sequence current originates from each installation point.

[0015] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, in step one, the number of sampling points for each cycle of zero-sequence voltage is set to N, ranging from 128 to 512. The original zero-sequence voltage recording data stream after recording is set to Uo(m*N+n), where m represents the m-th recording cycle and n represents the n-th recording data in the m-th recording cycle.

[0016] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in low-resistance grounding systems, the default value of N is 256.

[0017] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in low-resistance grounding systems, in step two, a new weighted cumulative zero-sequence voltage sequence Us is obtained by transformation based on the original zero-sequence voltage waveform data stream.

[0018] in,

[0019] In the above formula, x(l) represents [U] O ((m-L+l)*N+n)-U O The fundamental amplitude is obtained by Fourier transforming the sequence [(mL)*N+n)].

[0020] Among them, the maximum cumulative number of recorded waves, L, is set, with a value range of 8 to 256.

[0021] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, the default value of L is 32.

[0022] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, in step three, for each weighted accumulated zero-sequence voltage sequence Us, a sampling point is calculated and compared with a pre-set threshold THres_Us, wherein the threshold THres_Us is 3 to 20.

[0023] If the threshold is exceeded, increment by one;

[0024] If, in a series of P statistical analyses, the threshold is exceeded by more than Q calculations, then the zero-sequence voltage is considered to have undergone a sudden change, and the current time stamp Ts of the sudden change is recorded.

[0025] Where Q ≤ P.

[0026] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, the threshold THres_Us is 10 by default, Q is 3 by default, and P is 5 by default.

[0027] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system, in step four, based on the time stamp Ts of the abrupt change, a one-cycle zero-sequence current sequence Io(k,n) is obtained for each installation point with the time stamp Ts of the abrupt change as the starting point, and a one-cycle data Us(n) of the weighted additional zero-sequence voltage with the starting point at that time is also obtained, where k represents the kth installation point.

[0028] Furthermore, in the above-mentioned method for improving the sensitivity detection of zero-sequence voltage changes in low-resistance grounding systems, in step five, the correlation coefficient Corr(k) between the cycle-wise zero-sequence current sequence Io(k,n) and the cycle-wise data sequence Us(n) is calculated at each installation point.

[0029]

[0030] If the correlation coefficient Corr(k) ≥ the threshold THres_C1, then the installation point is considered not to be upstream of the fault point on the faulty line.

[0031] If the correlation coefficient Corr(k) ≤ the threshold THres_C2, then the installation point is considered to be upstream of the fault point on the faulty line.

[0032] The default value of the threshold THres_C1 is 0.9, and the default value of the threshold THres_C2 is -0.9.

[0033] By means of the above-described solution, the present invention has at least the following advantages:

[0034] Weighted integration can be used to improve the identification accuracy and anti-interference capability of zero-sequence current under high-resistance grounding faults. Simultaneously, by directly calculating the correlation between the integration results and zero-sequence voltage, a simple and effective method can be constructed to improve the accuracy of grounding fault location.

[0035] It can overcome the problem of dead zone of zero-sequence voltage in high-resistance faults under small resistance grounding by integrating and accumulating the differential zero-sequence voltage, and realize the detection and location of high-resistance grounding faults under small resistance grounding by combining the phase relationship between the zero-sequence voltage and the zero-sequence current.

[0036] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0037] Figure 1 This is the original zero-sequence voltage waveform.

[0038] Figure 2 It is the waveform after weighted difference accumulation.

[0039] Figure 3 yes Figure 2 A magnified view of a portion of the image. Detailed Implementation

[0040] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0041] like Figures 1 to 3 A method for improving the sensitivity of zero-sequence voltage change detection in low-resistance grounding systems, including methods based on acquired zero-sequence voltage and zero-sequence current, is unique in that it includes the following steps:

[0042] Step 1: Set the number of sampling points for each cycle of zero-sequence voltage and obtain the corresponding raw zero-sequence voltage waveform data stream. During implementation, set the number of sampling points for each cycle of zero-sequence voltage to N, ranging from 128 to 512. A larger value for N results in a greater computational load but higher fidelity for transient signals. Therefore, the value of N can be set as needed; a default value of 256 is acceptable for general requirements. Next, let the raw zero-sequence voltage waveform data stream after waveform recording be Uo(m*N+n), where m represents the m-th waveform recording cycle and n represents the n-th waveform data in the m-th waveform recording cycle.

[0043] Step two involves transforming the original zero-sequence voltage waveform data stream into a weighted, cumulative differential zero-sequence voltage sequence. Specifically, based on the original zero-sequence voltage waveform data stream, a new weighted, cumulative zero-sequence voltage sequence Us is obtained through transformation.

[0044] During the implementation period, UO((mL)*N+n)],(m≥L).

[0045] In the above formula, x(l) represents [U] O ((m-L+l)*N+n)-U O The fundamental amplitude is obtained by Fourier transforming the sequence [(mL)*N+n)]. The maximum accumulated wave number L is set, ranging from 8 to 256. A smaller value results in faster high-resistivity fault detection, but reduces the response capability. A larger value results in longer detection time, but higher sensitivity. Multiple comparative experiments showed that a default value of 32 for L is sufficient.

[0046] Step 3: Perform waveform abrupt change detection on the acquired weighted cumulative differential zero-sequence voltage sequence to obtain the time stamp of the abrupt change. That is, for each sample point of the weighted cumulative zero-sequence voltage sequence Us, compare it with the pre-set threshold THres_Us, which ranges from 3 to 20.

[0047] During implementation, if the threshold value is exceeded, it is incremented by one. If the threshold is exceeded more than Q times in P consecutive statistical analyses, the zero-sequence voltage is considered to have undergone a sudden change, and the current time stamp Ts of the sudden change is recorded. Where Q ≤ P. For ease of implementation, the threshold THres_Us defaults to 10, Q defaults to 3, and P defaults to 5.

[0048] Step four: Obtain the zero-sequence current recording data for that cycle at all installation points at the same time based on the time stamp of the abrupt change. Specifically, based on the time stamp Ts of the abrupt change, obtain the zero-sequence current sequence Io(k,n) for one cycle at each installation point, starting from the time stamp Ts. Simultaneously, obtain the weighted zero-sequence voltage data Us(n) starting from that time. Here, k represents the k-th installation point.

[0049] Step 5: Calculate the correlation coefficient between the zero-sequence current waveform sequence for each installation point and the weighted cumulative differential zero-sequence voltage waveform sequence for that cycle, and use this correlation coefficient to determine the fault. Specifically, calculate the correlation coefficient Corr(k) between the zero-sequence current waveform sequence Io(k,n) and the waveform data sequence Us(n) for each installation point.

[0050] Use the following formula,

[0051]

[0052] If the correlation coefficient Corr(k) ≥ the threshold THres_C1, the installation point is considered not upstream of the fault point on the faulty line. If the correlation coefficient Corr(k) ≤ the threshold THres_C2, the installation point is considered upstream of the fault point on the faulty line. For ease of implementation, the default value for the threshold THres_C1 is 0.9, and the default value for the threshold THres_C2 is -0.9.

[0053] According to a preferred embodiment of the present invention, the zero-sequence voltage is obtained through monitoring data at the installation points, originating from the busbar or a point on the line; the zero-sequence current originates from each installation point. Therefore, the final fault location result will be between two installation points, downstream of one installation point, or outside the area (indicating the fault is not downstream of any installation point).

[0054] The working principle of this invention is as follows:

[0055] A high-resistance grounding fault occurred in a certain area, and its original zero-sequence voltage waveform is as follows: Figure 1 As shown. Then, a weighted difference summation is performed to obtain the result as shown. Figure 2 The zero-sequence voltage waveform is shown. During this period, the maximum weighted accumulation count L = 32 is used. Subsequently, the zero-sequence voltage after weighted differential accumulation of L = 32 is locally magnified to obtain the following waveform: Figure 3 The zero-sequence voltage waveform shown is a single-cycle waveform.

[0056] Next, based on the differentially weighted cumulative zero-sequence voltage, the correlation coefficient between it and the zero-sequence current upstream of the fault point is -0.97, and the correlation coefficient between it and the zero-sequence current downstream of the fault point is 0.95. Then, it is determined whether the monitoring point is located upstream of the fault point on the faulty line.

[0057] As can be seen from this example, the zero-sequence voltage processed by the weighted differential accumulation method can significantly improve the quality of the original zero-sequence voltage signal when the grounding is high in a low-resistance grounding system. The correlation coefficient between the zero-sequence voltage processed by the weighted differential accumulation method and the zero-sequence current at the monitoring point can clearly determine whether the monitoring point is located upstream of the fault point of the single-phase grounding fault line.

[0058] As can be seen from the above description and the accompanying drawings, the present invention has the following advantages:

[0059] Weighted integration can be used to improve the identification accuracy and anti-interference capability of zero-sequence current under high-resistance grounding faults. Simultaneously, by directly calculating the correlation between the integration results and zero-sequence voltage, a simple and effective method can be constructed to improve the accuracy of grounding fault location.

[0060] It can overcome the problem of dead zone of zero-sequence voltage in high-resistance faults under small resistance grounding by integrating and accumulating the differential zero-sequence voltage, and realize the detection and location of high-resistance grounding faults under small resistance grounding by combining the phase relationship between the zero-sequence voltage and the zero-sequence current.

[0061] Furthermore, the orientations or positional relationships described in this invention are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the purpose of facilitating the description of this invention and simplifying the description, and are not intended to indicate or imply that the device or structure referred to must have a specific orientation, or to operate in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0062] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. Methods to improve the sensitivity of zero-sequence voltage change detection in low-resistance grounding systems, including methods based on the acquired zero-sequence voltage and zero-sequence current. Its features Includes the following steps: Step 1: Set the number of sampling points for each cycle of zero-sequence voltage and obtain the corresponding raw zero-sequence voltage waveform data stream; Step 2: Convert the original zero-sequence voltage waveform data stream into a weighted, accumulated differential zero-sequence voltage sequence; The number of sampling points for the zero-sequence voltage per cycle is set to N, ranging from 128 to 512. The original zero-sequence voltage recording data stream after recording is set to Uo(m*N+n), where m represents the m-th recording cycle and n represents the n-th recording data in the m-th recording cycle. The default value of N is 256. Based on the original zero-sequence voltage waveform data stream, a new weighted cumulative zero-sequence voltage sequence Us is obtained after transformation, where... In the above formula, x(l) represents [U] O ((m-L+l)*N+n)-U O The fundamental amplitude value obtained by Fourier transform of the sequence [(mL)*N+n)] is given by setting the maximum cumulative number of recorded waves L, which ranges from 8 to 256, and the default value of L is 32. Step 3: Perform waveform abrupt change detection on the acquired weighted cumulative differential zero-sequence voltage sequence to obtain the time label of the abrupt change moment; Step 4: Obtain the waveform data of the zero-sequence current of all installation points at the same time for that cycle based on the time stamp of the abrupt change. Step 5: Calculate the correlation coefficient between the zero-sequence current recording sequence of each installation point and the weighted cumulative differential zero-sequence voltage data sequence of that cycle, and then use this correlation coefficient to identify the fault.

2. The method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system according to claim 1, characterized in that: The zero-sequence voltage is obtained through monitoring data at the installation points, and it originates from the bus or from a point on the line; the zero-sequence current originates from each installation point.

3. The method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system according to claim 1, characterized in that: In step three, for each sample point of the new weighted cumulative zero-sequence voltage sequence Us, it is compared with a pre-set threshold THres_Us, which is 3 to 20. If the threshold is exceeded, increment by one; If, in a series of P statistical analyses, the threshold is exceeded by more than Q calculations, then the zero-sequence voltage is considered to have undergone a sudden change, and the current time stamp Ts of the sudden change is recorded. Where Q ≤ P.

4. The method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system according to claim 3, characterized in that: The threshold THres_Us defaults to 10, Q defaults to 3, and P defaults to 5.

5. The method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system according to claim 1, characterized in that: In step four, based on the time stamp Ts of the abrupt change, a cycle of zero-sequence current sequence Io(k,n) is obtained for each installation point with the time stamp Ts of the abrupt change as the starting point. At the same time, a cycle of weighted additional zero-sequence voltage Us(n) is also obtained with the time stamp Ts as the starting point, where k represents the kth installation point.

6. The method for improving the sensitivity detection of zero-sequence voltage changes in a low-resistance grounding system according to claim 1, characterized in that: In step five, the correlation coefficient Corr(k) between the zero-sequence current sequence Io(k,n) and the individual cycle data sequence Us(n) is calculated at each installation point. If the correlation coefficient Corr(k) ≥ the threshold THres_C1, then the installation point is considered not to be upstream of the fault point on the faulty line. If the correlation coefficient Corr(k) ≤ the threshold THres_C2, then the installation point is considered to be upstream of the fault point on the faulty line. The default value of the threshold THres_C1 is 0.9, and the default value of the threshold THres_C2 is -0.9.

Images