A power distribution network abnormal state monitoring method and device based on a trend cumulative effect

Abstract

The application relates to a power distribution network abnormal state monitoring method and device based on a trend cumulative effect. The method collects zero sequence current I0 and zero sequence voltage U0 of a power distribution network in real time, sequentially carries out data recording, zero drift and calibration processing on the collected data, then carries out growth trend accumulation judgment, redundant rising trend accumulation judgment, decreasing trend accumulation judgment and repeated zero-crossing judgment according to the integral results after the zero drift and calibration processing, and carries out corresponding fault processing according to the judgment results. The trend of zero sequence power before and after an abnormal state is accumulated through a period sliding window, the grounding condition of the power distribution network is judged according to the accumulated trend, the problem that a small signal component under high resistance grounding is difficult to identify can be avoided, and the accuracy of the abnormal state judgment of the power distribution network is effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network monitoring technology, and particularly relates to a method and device for monitoring abnormal states of power distribution networks based on the cumulative effect of trends. Background Technology

[0002] Currently, the power distribution network is enormous, with numerous lines. Rural distribution networks are dominated by overhead lines, while urban distribution networks are primarily composed of cable lines. However, both overhead and cable lines, due to their inherent distribution network characteristics, are characterized by wide distribution, diversity, and complexity. Furthermore, because distribution networks mainly consist of neutral point arc suppression coils and low-resistance grounding systems, coupled with their widespread presence and complexity, single-phase grounding faults have become the most frequent type of fault in distribution networks.

[0003] Meanwhile, the applicant discovered that single-phase ground faults are accompanied by changes in zero-sequence current and zero-sequence voltage. However, the specific quantities are usually affected by the grounding impedance and cannot guarantee that the zero-sequence current and zero-sequence voltage will rapidly rise to extremely high values. Especially in cases of grounding on grass, concrete, asphalt, or trees, different types of ground faults do not cause significant changes in zero-sequence current, particularly in high-impedance grounding, where changes in zero-sequence voltage and zero-sequence current are limited. However, existing methods relying on excessive zero-sequence current and zero-sequence voltage for single-phase ground fault detection are far from sufficient for monitoring single-phase ground faults in current distribution networks. Therefore, employing specific algorithmic strategies to monitor abnormal states in distribution networks under high-impedance grounding and to provide early warnings in the event of a fault has become a necessary approach.

[0004] When a ground fault occurs, the location of a single-phase grounding point can be considered equivalent to a zero-sequence power source. In this case, the arc suppression coil or small resistor at the neutral point becomes the load of the zero-sequence power source, and the zero-sequence power should appear negative when observed from the power source side of the grounding point. However, in practical applications, due to sampling errors, grounding transient changes, and the influence of harmonic and transient components in the grounding signal, the sign of the zero-sequence power cannot be simply used as a constraint. Furthermore, during arcing grounding, the influence of transient signals is even more severe, all of which lead to misjudgments when relying solely on the sign of the zero-sequence power. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, this invention provides a method and device for monitoring abnormal states in power distribution networks based on the cumulative effect of trends.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] Firstly, the present invention provides a distribution network abnormal state monitoring method based on trend accumulation effect, which collects the zero-sequence current I0 and zero-sequence voltage U0 of the distribution network in real time, and performs data recording and zero drift and calibration processing on the collected data in sequence. Then, based on the integral results after zero drift and calibration processing, it performs cumulative judgment of growth trend, cumulative judgment of redundancy rising trend, cumulative judgment of decreasing trend, and judgment of repeated zero crossing, and performs corresponding fault handling based on the judgment results, notifying the power grid operation and maintenance personnel that the boundary section needs to be searched, or notifying the power grid operation and maintenance personnel that a fault has occurred and the fault point needs to be investigated.

[0008] The distribution network abnormal state monitoring method of the present invention performs periodic sliding window accumulation of the trend of zero-sequence power before and after the abnormal state, and judges the grounding status of the distribution network based on the accumulated trend. This avoids the problem that transient components, especially small signal components under high-resistance grounding, are difficult to identify, and effectively improves the accuracy of distribution network abnormal state judgment.

[0009] Furthermore, the method for monitoring abnormal states in a distribution network based on the trend accumulation effect specifically includes:

[0010] S1. Real-time acquisition of zero-sequence current I0 and zero-sequence voltage U0 in the power distribution network;

[0011] S2. Data recording: When the zero-sequence current I0 and the zero-sequence voltage U0 change abruptly or excessively, record multiple data waveforms at the moment of the abrupt change or excessive change;

[0012] S3. Zero drift and calibration processing: The recorded data waveform is sampled, and the sampled data is integrated to obtain multiple integration results. Then, the integration results are subjected to zero drift and calibration processing.

[0013] S4. Cumulative growth trend judgment: Based on the integral value of the integral result after zero drift and calibration processing, calculate the integral difference of the integral results in two adjacent sliding window periods, and then judge the cumulative growth trend based on the integral difference;

[0014] Redundancy upward trend cumulative judgment: Based on the integral difference and integral value obtained in the cumulative judgment process of the growth trend, the redundancy upward trend is judged and accumulated;

[0015] Decreasing trend cumulative judgment: Based on the integral difference and integral value obtained in the process of cumulative judgment of growth trend, the cumulative decreasing trend is judged;

[0016] Repeated zero crossing judgment: Based on the integral value of the integral results of two adjacent sliding window cycles within each sampling full data window, it is determined whether there is repeated zero crossing;

[0017] S5. Fault Handling: Based on the judgment result of S4, perform corresponding fault handling. When the positive growth trend is valid, determine that an off-boundary grounding state has occurred and notify the power grid operation and maintenance personnel that the on-boundary section needs to be searched. When the redundancy upward trend is valid, determine that an off-boundary grounding state has occurred and notify the power grid operation and maintenance personnel that the on-boundary section needs to be searched. When the decreasing trend is valid, determine that an on-boundary grounding state has occurred, output an alarm signal, notify the power grid operation and maintenance personnel that a fault has occurred and the fault point needs to be investigated. When repeatedly crossing zero, determine that an off-boundary grounding state has occurred and notify the power grid operation and maintenance personnel that the on-boundary section needs to be searched.

[0018] Furthermore, the formula for the integration operation in the zero drift and calibration process is as follows:

[0019] S=∫I0(t)*U0(t)dt

[0020] In the formula, S represents the integration result, t represents the sampling time, and dt represents the integration step size.

[0021] Furthermore, the data waveforms recorded in the data recording are the first m data waveforms and the last n data waveforms at the moment of abrupt change or excessive change.

[0022] Furthermore, in the zero-drift and calibration processing, the sampling frequency of the recorded data waveform is fs. Therefore, the recorded waveform data has (m+n)*0.02*fs data points. Starting from 0.02*fs data points, with an integration window length of fs / 50 data points, integration is performed on all data points within each integration window, resulting in (m+n-1)*0.02*fs integration results. The zero-drift and calibration processing of these integration results specifically involves: averaging the first (m-2)0.02*fs integration results; when the average value < 0, subtracting the average value from all (m+n-1)*0.02*fs integration results; when the average value > the positive integration judgment value, subtracting the difference between the average value and the positive integration judgment value from all (m+n-1)*0.02*fs integration results; otherwise, keeping the integration results unchanged.

[0023] Furthermore, in the cumulative growth trend judgment, the judgment based on the integral difference and the cumulative growth trend are specifically as follows: when the integral difference > 0, the positive growth counter is incremented by 1; otherwise, the positive growth counter is reset to zero. When the value of the positive growth counter during the sliding window period is > fs / 25, the positive growth trend is determined to be valid.

[0024] Furthermore, in the cumulative judgment of the redundancy upward trend, the judgment and accumulation of the redundancy upward trend based on the integral difference and integral value obtained in the cumulative judgment process of the growth trend is specifically as follows: when the integral difference > 0 and the integral value > 0, and the redundancy upward counter < fs / 50*0.1, the redundancy upward counter is incremented by 1; otherwise, the redundancy upward counter is reset to zero. After the redundancy upward counter ≥ fs / 50*0.1, when the integral difference > 0, the redundancy upward counter is incremented by 1; when the integral difference < 0, if the integral value of the integral result is greater than the integral value of the number of fs / 50*0.1, the redundancy upward counter is still incremented by 1. When none of the above conditions are met, the redundancy upward counter is reset to zero. When the redundancy upward counter > fs / 25 within the sliding window period, the redundancy upward trend is determined to be valid.

[0025] Furthermore, in the cumulative judgment of the decreasing trend, the judgment of the cumulative decreasing trend based on the integral difference and integral value obtained in the cumulative judgment of the growth trend is specifically as follows: when the integral difference is <0 and the integral value is <0, the negative growth counter is incremented by 1; otherwise, the negative growth counter is reset to zero. When the value of the negative growth counter during the sliding window period is >fs / 25, the negative growth trend is determined to be valid.

[0026] Furthermore, in the repeated zero-crossing judgment, when the product of the integral values ​​of the integral results in two adjacent sliding window periods is ≤0, the zero-crossing counter is incremented by 1; when the value of the zero-crossing counter in the full data window is greater than 2, repeated zero crossing is judged.

[0027] Secondly, the present invention also provides a distribution network abnormal state monitoring device based on trend accumulation effect. When the device is running, it executes the above-mentioned distribution network abnormal state monitoring method, including a zero-sequence current and voltage acquisition unit, a data recording module, a zero drift and calibration processing module, a growth trend accumulation judgment module, a redundancy rising trend accumulation judgment module, a decreasing trend accumulation judgment module, a repeated zero crossing judgment module, and a fault processing module.

[0028] For the various aspects of the second aspect mentioned above and the technical effects that each aspect may achieve, please refer to the above description of the technical effects that can be achieved for the first aspect or the various possible solutions in the first aspect, which will not be repeated here. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the distribution network abnormality monitoring method based on trend accumulation effect as described in this invention.

[0030] Figure 2 This is a schematic diagram of the distribution network abnormal state monitoring device based on trend accumulation effect described in this invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0032] The present invention discloses a distribution network abnormal state monitoring method based on trend accumulation effect, which collects the zero-sequence current I0 and zero-sequence voltage U0 of the distribution network in real time, and performs data recording and zero drift and calibration processing on the collected data in sequence. Then, based on the integral results after zero drift and calibration processing, it performs cumulative judgment of growth trend, cumulative judgment of redundancy rising trend, cumulative judgment of decreasing trend, and judgment of repeated zero crossing.

[0033] like Figure 1 As shown, the method for monitoring abnormal states of distribution networks based on trend accumulation effects specifically includes the following steps:

[0034] Step S1. Real-time acquisition of zero-sequence current I0 and zero-sequence voltage U0 of the distribution network.

[0035] Step S2. Data recording: When the zero-sequence current I0 and zero-sequence voltage U0 of the distribution network acquired in real time in step S1 undergo sudden or excessive changes, record multiple data waveforms at the moment of the sudden or excessive change.

[0036] Specifically, when the zero-sequence current I0 and the zero-sequence voltage U0 undergo a sudden change or excessive change, the recorded data waveforms are the first m data waveforms and the last n data waveforms at the moment of the sudden change or excessive change. Here, m and n are user-defined. For example, the common waveform recording requirement in the distribution network is 4 data waveforms before and 8 data waveforms after the fault moment, or 8 data waveforms before and 12 data waveforms after the fault moment. That is, when a sudden change or analog quantity exceeds a certain value at a certain moment is collected, the data waveforms of the 4 or 8 cycles before this moment and the data waveforms of the 8 or 12 cycles after this moment are recorded from the moment the value exceeds the certain value.

[0037] Step S3. Zero Drift and Calibration Processing: The data waveform recorded in Step S2 is sampled, and the sampled data is integrated to obtain multiple integration results for subsequent trend judgment. Then, the integration results are subjected to zero drift and calibration processing to ensure that the integration result is between 0 and the positive integration judgment value, which is beneficial for subsequent trend judgment (because under normal circumstances, the integration result is positive and between 0 and the positive integration judgment value, but due to acquisition interference, zero drift, etc., the integration result may be less than zero or greater than the positive integration judgment value); The formula for the integration operation is as follows:

[0038] S=∫I0(t)*U0(t)dt

[0039] In the formula, S represents the integration result, t represents the sampling time, and dt represents the integration step size.

[0040] The zero drift and calibration process refers to calculating the average of multiple integration results. If the average is less than 0, the average is subtracted from all integration results to raise the integration result to a position greater than 0. If the average is greater than the positive integration judgment value, the difference between the average and the positive integration judgment value is subtracted from all integration results to lower the integration result to below the positive integration judgment value. Otherwise, the integration results remain unchanged.

[0041] In one possible implementation scheme, the sampling frequency of the recorded data waveform in the zero drift and calibration process is fs (fs ≥ 6.8kHz, because when intermittent arcing occurs, the peak time of the intermittent arcing current is very short. If the sampling rate is not high enough, the zero-sequence current I0 cannot be accurately collected, thus causing the subsequent trend judgments to fail). Then the recorded waveform data has a total of (m+n)*0.02*fs data points. Starting from the 0.02*fs data point, with the length of fs / 50 data points as the integration window, the integration operation is performed on all data points in each integration window to obtain (m+n-1)*0.02*fs integration results. That is, the integration point of the current point is obtained from the current point and the fs / 50-1 points before it. From the first to fs / 50 data points, the first integration point can be obtained by integration, resulting in a total of (m+n-1)*0.02*fs integration results. Then, zero drift and calibration processing is performed on the integration results. Specifically, the average value of the first (m-2)0.02*fs integration results is calculated. When the average value is less than 0, the average value is subtracted from all (m+n-1)*0.02*fs integration results. When the average value is greater than the positive integration judgment value, the difference between the average value and the positive integration judgment value is subtracted from all (m+n-1)*0.02*fs integration results. In other cases, the integration results remain unchanged.

[0042] Step S4. Cumulative growth trend judgment: Based on the integral value of the integral result after zero drift and calibration processing in step S3, calculate the integral difference of the integral results in two adjacent sliding window periods, and then judge the cumulative growth trend based on the integral difference. Specifically, when the integral difference is > 0, the positive growth counter is incremented by 1; otherwise, the positive growth counter is reset to zero. When the value of the positive growth counter in the sliding window period is > fs / 25, the positive growth trend is determined to be valid.

[0043] Redundancy Rising Trend Cumulative Judgment: Based on the integral difference and integral value obtained during the cumulative judgment of the growth trend, the redundancy rising trend is judged and accumulated. Specifically, it can be as follows: when the integral difference > 0 and the integral value > 0, and the redundancy rising counter < fs / 50*0.1, the redundancy rising counter is incremented by 1; otherwise, the redundancy rising counter is reset to zero. After the redundancy rising counter ≥ fs / 50*0.1, when the integral difference > 0, the redundancy rising counter is incremented by 1; when the integral difference < 0, if the integral value of the integral result is greater than the integral value of fs / 50*0.1, the redundancy rising counter is still incremented by 1. When none of the above conditions are met, the redundancy rising counter is reset to zero. When the redundancy rising counter > fs / 25 within the sliding window period, the redundancy rising trend is determined to be valid.

[0044] Decreasing trend cumulative judgment: Based on the integral difference and integral value obtained in the process of increasing trend cumulative judgment, the cumulative decreasing trend is judged. Specifically, when the integral difference is <0 and the integral value is <0, the negative growth counter is incremented by 1; otherwise, the negative growth counter is reset to zero. When the value of the negative growth counter in the sliding window period is >fs / 25, the negative growth trend (decreasing trend) is determined to be valid.

[0045] Repeated zero crossing judgment: Based on the integral value of the integral result of two adjacent sliding window periods within each sampling full data window, it is determined whether there is repeated zero crossing. Specifically, when the product of the integral values ​​of the integral results in two adjacent sliding window periods is ≤0, the zero crossing counter is incremented by 1; when the value of the zero crossing counter within the full data window is greater than 2, it is determined that there is repeated zero crossing.

[0046] Step S5. Fault Handling: Based on the judgment result of step S4, perform corresponding fault handling; specifically: when the positive growth trend is valid, determine that an off-boundary grounding state has occurred, and notify the power grid operation and maintenance personnel that an on-boundary section needs to be located; when the redundancy upward trend is valid, determine that an off-boundary grounding state has occurred, and notify the power grid operation and maintenance personnel that an on-boundary section needs to be located; when the decreasing trend is valid, determine that an on-boundary grounding state has occurred, output an alarm signal, and notify the power grid operation and maintenance personnel that a fault has occurred and fault point investigation is required; when repeatedly crossing zero, determine that an off-boundary grounding state has occurred, and notify the power grid operation and maintenance personnel that an on-boundary section needs to be located.

[0047] The present invention discloses a distribution network abnormal state monitoring method based on trend accumulation effect. This method uses a periodic sliding window to accumulate the trend of zero-sequence power before and after an abnormal state, and judges the grounding status of the distribution network based on the accumulated trend. This avoids the problem of difficulty in identifying transient components, especially small-signal components under high-resistance grounding, effectively improving the accuracy of distribution network abnormal state judgment. Specifically, it is manifested as follows:

[0048] (1) This invention calculates the cumulative effect of energy trends by integration, divides trends into different types, and processes them by type, thus avoiding the problem of refusal to move or false movement caused by simply relying on the sign of energy.

[0049] (2) By accumulating trends, this invention avoids the influence of a transient instantaneous variable on the entire judgment process. The trend after integration includes the entire process within the integration period, thereby reducing the possibility of false judgments, and in particular, avoiding the influence of instantaneous factors on the entire judgment.

[0050] like Figure 2 As shown, the present invention also provides a distribution network abnormal state monitoring device based on trend accumulation effect, including a zero-sequence current and voltage acquisition unit 100, a data recording module 200, a zero drift and calibration processing module 300, a growth trend accumulation judgment module 400, a redundancy rising trend accumulation judgment module 500, a decreasing trend accumulation judgment module 600, a repeated zero crossing judgment module 700, and a fault processing module 800. The device executes the above-mentioned distribution network abnormal state monitoring method of the present invention during operation.

[0051] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for monitoring abnormal states in a distribution network based on the cumulative effect of trends, characterized in that, The zero-sequence current I0 and zero-sequence voltage U0 of the distribution network are collected in real time. The collected data are then processed for data recording, zero drift and calibration. Based on the integral results after zero drift and calibration, the system performs cumulative judgment on growth trend, cumulative judgment on redundancy increase trend, cumulative judgment on decrease trend, and judgment on repeated zero crossing. Based on the judgment results, the system performs corresponding fault handling, notifies the grid operation and maintenance personnel that the section within the boundary needs to be searched, or notifies the grid operation and maintenance personnel that a fault has occurred and the fault point needs to be investigated. Specifically, the following steps are included: S1. Real-time acquisition of zero-sequence current I0 and zero-sequence voltage U0 in the power distribution network; S2. Data recording: When the zero-sequence current I0 and the zero-sequence voltage U0 undergo sudden or excessive changes, record multiple data waveforms at the moment of the sudden or excessive change; S3. Zero drift and calibration processing: The recorded data waveform is sampled, and the sampled data is integrated to obtain multiple integration results. Then, the integration results are subjected to zero drift and calibration processing. S4. Cumulative growth trend judgment: Based on the integral value of the integral result after zero drift and calibration processing, calculate the integral difference of the integral results in two adjacent sliding window periods, and then judge the cumulative growth trend based on the integral difference; Redundancy upward trend cumulative judgment: Based on the integral difference and integral value obtained in the cumulative judgment process of the growth trend, the redundancy upward trend is judged and accumulated; Decreasing trend cumulative judgment: Based on the integral difference and integral value obtained in the process of cumulative judgment of growth trend, the cumulative decreasing trend is judged; Repeated zero crossing judgment: Based on the integral value of the integral results of two adjacent sliding window cycles within each sampling full data window, it is determined whether there is repeated zero crossing; S5. Fault Handling: Based on the judgment result of S4, perform corresponding fault handling. When the positive growth trend is valid, determine that an off-boundary grounding state has occurred and notify the power grid operation and maintenance personnel that the on-boundary section needs to be searched. When the redundancy upward trend is valid, determine that an off-boundary grounding state has occurred and notify the power grid operation and maintenance personnel that the on-boundary section needs to be searched. When the decreasing trend is valid, determine that an on-boundary grounding state has occurred, output an alarm signal, notify the power grid operation and maintenance personnel that a fault has occurred and the fault point needs to be investigated. When repeatedly crossing zero, determine that an off-boundary grounding state has occurred and notify the power grid operation and maintenance personnel that the on-boundary section needs to be searched.

2. The method according to claim 1, characterized in that, The formula for integration calculation in the zero drift and calibration process is as follows: S =∫ I 0( t )* U 0( t ) dt In the formula S Indicates the result of integration. t Indicates the sampling time. dt This represents the integration step size.

3. The method according to claim 1 or 2, characterized in that, The data waveforms recorded in the data recording are the first m data waveforms and the last n data waveforms at the moment of sudden change or excessive change.

4. The method according to claim 3, characterized in that, In the zero-drift and calibration processing, the sampling frequency of the recorded data waveform is fs. Therefore, the recorded waveform data has (m+n)*0.02*fs data points. Starting from 0.02*fs data points, with an integration window length of fs / 50 data points, integration is performed on all data points within each integration window, resulting in (m+n-1)*0.02*fs integration results. The zero-drift and calibration processing of these integration results is as follows: the first (m-2)0.02*fs integration results are averaged. When the average value < 0, the average value is subtracted from all (m+n-1)*0.02*fs integration results. When the average value > the positive integration judgment value, the difference between the average value and the positive integration judgment value is subtracted from all (m+n-1)*0.02*fs integration results. Otherwise, the integration results remain unchanged.

5. The method according to claim 4, characterized in that, In the cumulative judgment of growth trend, the judgment based on integral difference and cumulative growth trend is specifically as follows: when integral difference > 0, the positive growth counter is incremented by 1; otherwise, the positive growth counter is reset to zero. When the value of the positive growth counter is greater than fs / 25 within the sliding window period, the positive growth trend is determined to be valid.

6. The method according to claim 4, characterized in that, In the cumulative judgment of redundancy upward trend, the judgment and accumulation of redundancy upward trend based on the integral difference and integral value obtained in the cumulative judgment process is as follows: when the integral difference > 0 and the integral value > 0, and the redundancy upward counter is less than fs / 50*0.1, the redundancy upward counter is incremented by 1; otherwise, the redundancy upward counter is reset to zero. After the redundancy upward counter ≥ fs / 50*0.1, when the integral difference > 0, the redundancy upward counter is incremented by 1; when the integral difference < 0, if the integral value of the integral result is greater than the integral value of the number of fs / 50*0.1, the redundancy upward counter is still incremented by 1. When none of the above conditions are met, the redundancy upward counter is reset to zero. When the redundancy upward counter > fs / 25 within the sliding window period, the redundancy upward trend is determined to be valid.

7. The method according to claim 4, characterized in that, In the cumulative judgment of the decreasing trend, the judgment of the cumulative decreasing trend based on the integral difference and integral value obtained in the cumulative judgment of the growth trend is as follows: when the integral difference is <0 and the integral value is <0, the negative growth counter is incremented by 1; otherwise, the negative growth counter is reset to zero. When the value of the negative growth counter during the sliding window period is >fs / 25, the negative growth trend is determined to be valid.

8. The method according to claim 4, characterized in that, In the repeated zero-crossing judgment, when the product of the integral values ​​of the integral results in two adjacent sliding window periods is ≤0, the zero-crossing counter is incremented by 1; when the value of the zero-crossing counter in the full data window is greater than 2, repeated zero crossing is judged.

9. A power distribution network abnormality monitoring device based on trend cumulative effect, characterized in that, When the device is in operation, it performs the distribution network abnormal state monitoring method described in any of claims 1-8, including a zero-sequence current and voltage acquisition unit, a data recording module, a zero drift and calibration processing module, a growth trend accumulation judgment module, a redundancy rising trend accumulation judgment module, a decreasing trend accumulation judgment module, a repeated zero crossing judgment module, and a fault handling module.

Images