Distribution network single-phase earth fault detection method and system based on adaptive fusion algorithm

Abstract

The invention provides a distribution network single-phase earth fault detection method and system based on an adaptive fusion algorithm, and relates to the technical field of distribution network earth fault detection methods. The method comprises the following steps: dynamically tracking the waveform of a target distribution network, comparing a zero-sequence voltage abrupt change value with a dynamic zero-sequence voltage abrupt change starting constant value, and determining a zero-sequence voltage abrupt change starting constant value; self-adaptive zero-sequence voltage abrupt change detection is carried out, and whether zero-sequence voltage abnormity occurs or not is judged; when zero sequence voltage abnormity occurs, original sampling data in the waveform is traced, a waveform sudden change position is identified from the original sampling data, a dynamic data window is delimited based on the waveform sudden change position, and alternating current channel data in the dynamic data window is synchronously acquired from the instantaneous value waveform; and through an adaptive fusion algorithm, carrying out ground fault type identification on the AC channel data in the dynamic data window to obtain a single-phase ground fault detection result of the target power distribution network. According to the method, the accuracy and robustness of distribution network single-phase earth fault detection are improved through the dynamic zero-sequence voltage abrupt change starting constant value and the dynamic data window.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network grounding fault detection methods, and in particular to a distribution network single-phase grounding fault detection method and system based on an adaptive fusion algorithm. Background Art

[0002] For small current grounding faults, due to the different neutral point grounding methods and the complexity of combined fault types, a single processing algorithm cannot meet the requirements of all grounding fault judgments. According to statistics, the vast majority of substations supplying power to the distribution network use a neutral point ungrounded method, and one-third use an arc suppression coil grounding method. For distribution lines with an ungrounded neutral point, since its steady-state angle can fully characterize the location characteristics of the grounding area, the steady-state power angle method is the best small current grounding algorithm. For the arc suppression coil grounding method, a transient grounding algorithm is often used to judge grounding faults. Nowadays, many places have proposed that distribution network equipment needs to have an adaptive neutral point grounding method, and have further put forward requirements for the accuracy of grounding faults. The original single grounding algorithm method can no longer meet market demand. Summary of the Invention

[0003] In order to solve the above problems, the present invention proposes a distribution network single-phase grounding fault detection method and system based on an adaptive fusion algorithm, which improves the accuracy and robustness of distribution network single-phase grounding fault detection by starting a fixed value and a dynamic data window through dynamic zero-sequence voltage mutation.

[0004] According to some embodiments, the present invention adopts the following technical solutions: A distribution network single-phase grounding fault detection method based on an adaptive fusion algorithm includes: Dynamically track the waveform of the target distribution network, the waveform including original sampling data and AC channel data, the AC channel data including zero-sequence voltage, phase current, and zero-sequence current; Based on the zero-sequence voltage, the zero-sequence voltage mutation value is calculated in real time. By comparing the zero-sequence voltage mutation value with the dynamic zero-sequence voltage mutation starting value, adaptive zero-sequence voltage mutation detection is performed to determine whether zero-sequence voltage abnormality occurs; When a zero-sequence voltage anomaly occurs, the original sampling data in the waveform is traced back to identify the waveform mutation location. Based on the waveform mutation location, a dynamic data window is defined, and the AC channel data in the dynamic data window is synchronously obtained from the instantaneous value waveform. Through the adaptive fusion algorithm, the AC channel data in the dynamic data window is used to identify the grounding fault type and obtain the single-phase grounding fault detection result of the target distribution network.

[0005] According to some embodiments, the present invention adopts the following technical solutions: A distribution network single-phase grounding fault detection system based on an adaptive fusion algorithm includes: The waveform tracking module is configured to dynamically track the waveform of the target distribution network, wherein the waveform includes original sampling data and AC channel data, and the AC channel data includes zero-sequence voltage, phase current, and zero-sequence current; The mutation detection module is configured to: calculate the zero-sequence voltage mutation value in real time based on the zero-sequence voltage, perform adaptive zero-sequence voltage mutation detection by comparing the zero-sequence voltage mutation value with the dynamic zero-sequence voltage mutation starting constant, and determine whether a zero-sequence voltage abnormality occurs; The data acquisition module is configured to: when a zero-sequence voltage anomaly occurs, trace back the original sampling data in the waveform, identify the waveform mutation location, define a dynamic data window based on the waveform mutation location, and synchronously acquire the AC channel data in the dynamic data window from the instantaneous value waveform; The fault identification module is configured to: identify the grounding fault type of the AC channel data in the dynamic data window through an adaptive fusion algorithm to obtain a single-phase grounding fault detection result of the target distribution network.

[0006] According to some embodiments, the present invention adopts the following technical solutions: A computer program product includes a computer program. When the computer program is executed by a processor, the computer program implements the distribution network single-phase grounding fault detection method using an adaptive fusion algorithm.

[0007] According to some embodiments, the present invention adopts the following technical solutions: A non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm is implemented.

[0008] According to some embodiments, the present invention adopts the following technical solutions: An electronic device includes: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the distribution network single-phase grounding fault detection method that implements an adaptive fusion algorithm.

[0009] Compared with the prior art, the present invention has the following beneficial effects: (1) For transient algorithms, obtaining the correct fault transient waveform is the key to correctly identifying ground faults. However, due to the possible imbalance of the line itself, the zero-sequence voltage value before the ground fault occurs cannot be determined. The existing algorithm sets a fixed zero-sequence voltage start mutation constant, which cannot meet the ground fault identification under all imbalances. Therefore, the present invention adopts a dynamic zero-sequence voltage mutation start constant to adaptively start the ground fault detection link to meet the ground fault identification under all imbalances.

[0010] (2) For transient algorithms, high-frequency components are extremely important features in transient data windows. If high-frequency components dominate, fixed-length data windows (half-wave or quarter-cycle) cannot accurately reflect the characteristics of high-frequency components at the moment of fault. Therefore, the present invention adopts flexible data window identification and uses a dynamic data window length scheme to reduce the transient high-frequency component interference caused by fixed-length data windows.

[0011] (3) The present invention adopts the fusion of multiple sub-algorithms, including the commonly used zero-sequence power direction method, vector superposition power direction method, three-phase unbalanced mutation method and other sub-algorithms. The number of algorithms can be expanded in the future. By taking a weighted average of the "reliability", the fusion of various algorithms can give a reliable result. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0013] Figure 1 This is a flow chart of the method of Example 1. DETAILED DESCRIPTION

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0016] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "comprising" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0017] Example 1 An embodiment of the present invention provides a method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm, including: Step 1: Dynamically track the waveform of the target distribution network, where the waveform includes original sampling data and AC channel data, where the AC channel data includes zero-sequence voltage, phase current, and zero-sequence current; Step 2: Based on the zero-sequence voltage, the zero-sequence voltage mutation value is calculated in real time. By comparing the zero-sequence voltage mutation value with the dynamic zero-sequence voltage mutation starting value, adaptive zero-sequence voltage mutation detection is performed to determine whether a zero-sequence voltage anomaly occurs; Step 3: If a zero-sequence voltage anomaly occurs, trace back the original sampling data in the waveform to identify the waveform mutation location, define a dynamic data window based on the waveform mutation location, and synchronously obtain the AC channel data in the dynamic data window from the instantaneous value waveform; Step 4: Use the adaptive fusion algorithm to identify the ground fault type of the AC channel data in the dynamic data window and obtain the single-phase ground fault detection result of the target distribution network.

[0018] As an embodiment, the present invention provides a method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm, which includes four steps: starting an adaptive zero-sequence voltage mutation detection step, starting a grounding fault detection step, identifying each grounding algorithm fault, and weighted fusion of each identification result. Figure 1 The specific implementation process is as follows: Step S1: Start the adaptive zero-sequence voltage mutation detection link Using the dynamic data tracking method, according to the historical zero-sequence voltage data of the line, set the dynamic zero-sequence voltage mutation starting value, and judge whether the zero-sequence voltage mutation value is greater than the dynamic zero-sequence voltage mutation starting value. If so, enter step S2; if not, continue to track the zero-sequence voltage mutation value, specifically: S1.1 calculates the zero-sequence voltage fundamental value for each cycle As the data source, the zero-sequence voltage fundamental value is subjected to first-order low-pass filtering to increase the data smoothness. Here, taking the frequency of 50HZ as an example, 1 cycle = 20ms, based on the zero-sequence voltage fundamental value and the zero-sequence voltage data of the previous cycle, the zero-sequence voltage data of the current cycle is calculated, which can be expressed as follows:

[0019] in, is the calculated real-time zero-sequence voltage fundamental value, is the zero-sequence voltage data of the nth cycle after first-order low-pass filtering. Assuming that the nth cycle is the current cycle, is the zero-sequence voltage data of the n-1th cycle, that is, the historical zero-sequence voltage data, is the filter coefficient; the filter coefficient can be adjusted and will affect the startup sensitivity.

[0020] S1.2 sets the dynamic zero-sequence voltage sudden change starting value of the current cycle based on the zero-sequence voltage data of the previous cycle:

[0021] in, It is the dynamic zero-sequence voltage sudden start setting value. It is the sudden start coefficient; the default value of the sudden start coefficient is set according to the waveform of actual line fault, grounding true type experimental fault, etc.

[0022] S1.3 determines whether the zero-sequence voltage mutation value is greater than the dynamic zero-sequence voltage mutation starting value: when When , it means that the zero-sequence voltage is abnormal and the ground fault detection link is entered.

[0023] Step S2: Start the ground fault detection process Trace back N cycles of data to find the location of waveform mutations, and synchronously obtain zero-sequence voltage data and other AC channel data through the dynamic data window.

[0024] Specifically, when it is detected that the zero-sequence voltage mutation value is greater than the dynamic zero-sequence voltage mutation starting value, first, trace back the previous N cycle data (here, 4 cycle data can be selected) and determine the waveform mutation position therefrom. Then, scan forward and backward from the waveform mutation position for waveforms that satisfy the zero-sequence voltage mutation value greater than the dynamic zero-sequence voltage mutation starting value, confirm the fault packet width, and thus form a dynamic data window. In the dynamic data window, zero-sequence voltage data and other AC channel data are synchronously acquired. Here, the other AC channel data include: phase current and zero-sequence current. The following steps are included: (1) Calculate the original sampling data in the traceability data area (the four cycles before the algorithm is started) point by point, and calculate the differential data of the whole cycle. The calculation formula is as follows:

[0025] in, is the differential sampling point data of the entire cycle (hereinafter referred to as differential data), is the original sampling data, is the number of sampling points per cycle.

[0026] here and The difference is: It is the zero-sequence voltage fundamental wave or the zero-sequence voltage data derived from the fundamental wave, and its update period is 20 milliseconds; The data is collected by the analog-to-digital converter AD, and the update period depends on the sampling frequency. The common sampling frequency is 128 points per cycle (20ms), that is, the update period is 156.25 microseconds.

[0027] The fundamental wave of the former is calculated by FFT from 20 milliseconds of actual sampling data, reflecting the overall change trend of the zero-sequence voltage amplitude (used for startup), while the latter reflects the transient (extremely short time) data characteristics of the zero-sequence voltage (used for fine feature analysis).

[0028] (2) Scan the differential data point by point to check whether it is consistent with the dynamic start setting value. The relationship between the waveform mutation position is determined as follows: First, in order to prevent misjudgment due to external reasons (such as electromagnetic interference), two consecutive sampling points are required to undergo mutations, which can be expressed as follows:

[0029]

[0030] in, It is the conversion coefficient from zero-sequence voltage fundamental value to AD sampling value.

[0031] Then, the differential data is further differentiated, which is actually a second-order full-wave differential, reflecting the speed of change of the data, that is, the rate of change of the rate of change. Since it is alternating current, its mutation is polarized. Therefore, the following formula is used to ensure that the mutation direction of two consecutive points is consistent, that is, the mutation polarity is consistent, that is, either both mutations increase or both mutations decrease:

[0032] (3) Reduce the influence of noise zero drift. Record the mutation point position n, search forward and backward for positions p and q respectively, and reduce the influence of noise zero drift by the following search conditions. The search conditions are:

[0033]

[0034] in, is the secondary transformer precision voltage.

[0035] In order to prevent misjudgment caused by too small a data window, the following requirements are required:

[0036] Finally, the dynamic data window [p,q] is obtained.

[0037] (4) Select the AC channel data in the dynamic data window [p,q] and input each transient grounding algorithm.

[0038] Step S3: Fault identification of each grounding algorithm Zero-sequence voltage data and other AC channel data obtained from the dynamic data window are input into various grounding algorithms, and each grounding algorithm outputs a result that includes algorithm reliability and characterizes the type of grounding fault. The various grounding algorithms include: zero-sequence power direction method, vector superposition power direction method, and three-phase unbalanced mutation method.

[0039] Here we take the zero-sequence power direction method as an example to illustrate: S3.1. Input data into the dynamic data window, including: zero-sequence voltage, phase current, and zero-sequence current; S3.2. Extend the dynamic data window length to a full cycle length and fill the extended area data with zeros; S3.3. Calculate the phase difference (0-180°) between the zero-sequence voltage and zero-sequence current within the dynamic data window. S3.4. Normalize the phase difference to (-1, 1) using the following formula:

[0040] Among them, angle is the phase difference between zero-sequence voltage and zero-sequence current, represents the normalized output of the algorithm, i.e. the result with the reliability of the algorithm and characterizing the type of ground fault, Indicates 100% reliability grounding. Indicates 100% reliability of external grounding, When it is near 0, it means that the data is unreliable. When the absolute value is around 0.5, it means that the data is relatively reliable.

[0041] Other transient algorithms also input data from the same dynamic data window and use similar processing methods to output results from each algorithm ranging from -1 to 1, and finally perform weighted averaging on the algorithm results.

[0042] Step S4: Weighted fusion of each recognition result The output of each grounding algorithm Perform weighted averaging and output the final result with the reliability of the fusion algorithm and the characterization of the ground fault type , which can be expressed as:

[0043] Where m is the number of grounding algorithms, A positive number indicates an internal ground fault. A negative number indicates an out-of-bounds ground fault.

[0044] Here, if each algorithm identifies the waveform as a reliable feature, it produces a highly reliable result (near +1 or -1); if it identifies it as an unreliable feature, it produces a less reliable result (near 0). Through the final weighted average, low-reliability results are automatically ignored. This processing method allows each algorithm to fully utilize its capabilities. For example, for arc grounding faults, the three-phase current imbalance method has an extremely high recognition rate, with an absolute reliability of almost 1. Other algorithms cannot produce highly reliable data (results are ambiguous, ranging from -0.5 to +0.5). The final weighted average collapses the data to the highly reliable result of the three-phase current imbalance method, thus producing the correct result.

[0045] Example 2 In one embodiment of the present invention, a distribution network single-phase grounding fault detection system using an adaptive fusion algorithm is provided, comprising: The waveform tracking module is configured to dynamically track the waveform of the target distribution network, wherein the waveform includes original sampling data and AC channel data, and the AC channel data includes zero-sequence voltage, phase current, and zero-sequence current; The mutation detection module is configured to: calculate the zero-sequence voltage mutation value in real time based on the zero-sequence voltage, perform adaptive zero-sequence voltage mutation detection by comparing the zero-sequence voltage mutation value with the dynamic zero-sequence voltage mutation starting constant, and determine whether a zero-sequence voltage abnormality occurs; The data acquisition module is configured to: when a zero-sequence voltage anomaly occurs, trace back the original sampling data in the waveform, identify the waveform mutation location, define a dynamic data window based on the waveform mutation location, and synchronously acquire the AC channel data in the dynamic data window from the instantaneous value waveform; The fault identification module is configured to: identify the grounding fault type of the AC channel data in the dynamic data window through an adaptive fusion algorithm to obtain a single-phase grounding fault detection result of the target distribution network.

[0046] Example 3 In one embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm is implemented.

[0047] Example 4 In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided, which is used to store computer instructions. When the computer instructions are executed by a processor, the distribution network single-phase grounding fault detection method based on the adaptive fusion algorithm is implemented.

[0048] Example 5 In one embodiment of the present invention, an electronic device is provided, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the distribution network single-phase grounding fault detection method that implements the adaptive fusion algorithm.

[0049] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0050] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0051] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A distribution network single-phase grounding fault detection method based on an adaptive fusion algorithm, characterized in that: include: Dynamically track the waveform of the target distribution network, the waveform including original sampling data and AC channel data, the AC channel data including zero-sequence voltage, phase current, and zero-sequence current; Based on the zero-sequence voltage, the zero-sequence voltage mutation value is calculated in real time. By comparing the zero-sequence voltage mutation value with the dynamic zero-sequence voltage mutation starting value, adaptive zero-sequence voltage mutation detection is performed to determine whether zero-sequence voltage abnormality occurs; When a zero-sequence voltage anomaly occurs, the original sampling data in the waveform is traced back to identify the waveform mutation location. Based on the waveform mutation location, a dynamic data window is defined, and the AC channel data in the dynamic data window is synchronously obtained from the instantaneous value waveform. Through the adaptive fusion algorithm, the AC channel data in the dynamic data window is used to identify the grounding fault type and obtain the single-phase grounding fault detection result of the target distribution network.

2. The method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm according to claim 1, wherein: The dynamic zero-sequence voltage sudden change starting setting is calculated in real time based on historical zero-sequence voltage data, specifically: Based on the zero-sequence voltage fundamental value, the zero-sequence voltage data is calculated and expressed as follows: in, is the calculated real-time zero-sequence voltage fundamental value, is the zero-sequence voltage data of the nth cycle after first-order low-pass filtering, is the zero-sequence voltage data of the n-1th cycle, is the filter coefficient; Based on the historical zero-sequence voltage data, the dynamic zero-sequence voltage sudden start setting is calculated and expressed as: in, It is the dynamic zero-sequence voltage sudden start setting value. is the mutation initiation coefficient.

3. The method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm according to claim 1, wherein: The zero-sequence voltage mutation value is the zero-sequence voltage fundamental value Zero-sequence voltage data of the n-1th cycle The difference.

4. The method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm according to claim 1, wherein: The dynamic data window is defined based on the waveform mutation position, specifically: Scan forward and backward from the waveform mutation position to find the waveform that satisfies the zero-sequence voltage mutation value greater than the dynamic zero-sequence voltage mutation starting value, confirm the fault packet width, and thus form a dynamic data window.

5. The method for detecting single-phase grounding fault in a distribution network using an adaptive fusion algorithm according to claim 1, wherein: The adaptive fusion algorithm is used to identify the ground fault type of the AC channel data in the dynamic data window, specifically: The AC channel data is input into several grounding algorithms, and each grounding algorithm outputs a result including the reliability of the algorithm and characterizing the type of grounding fault; The results output by each grounding algorithm are weighted averaged to obtain the final fault detection result with the reliability of the fusion algorithm and characterizing the grounding fault type.

6. The method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm according to claim 5, wherein: The grounding algorithms include but are not limited to the zero-sequence power direction method, the vector superposition power direction method, and the three-phase unbalanced mutation method.

7. A distribution network single-phase grounding fault detection system based on an adaptive fusion algorithm, characterized in that: include: The waveform tracking module is configured to dynamically track the waveform of the target distribution network, wherein the waveform includes original sampling data and AC channel data, and the AC channel data includes zero-sequence voltage, phase current, and zero-sequence current; The mutation detection module is configured to: calculate the zero-sequence voltage mutation value in real time based on the zero-sequence voltage, perform adaptive zero-sequence voltage mutation detection by comparing the zero-sequence voltage mutation value with the dynamic zero-sequence voltage mutation starting constant, and determine whether a zero-sequence voltage abnormality occurs; The data acquisition module is configured to: when a zero-sequence voltage anomaly occurs, trace back the original sampling data in the waveform, identify the waveform mutation location, define a dynamic data window based on the waveform mutation location, and synchronously acquire the AC channel data in the dynamic data window from the instantaneous value waveform; The fault identification module is configured to: identify the grounding fault type of the AC channel data in the dynamic data window through an adaptive fusion algorithm to obtain a single-phase grounding fault detection result of the target distribution network.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for detecting single-phase grounding faults in a distribution network using an adaptive fusion algorithm as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by the processor, the distribution network single-phase grounding fault detection method of the adaptive fusion algorithm according to any one of claims 1 to 6 is implemented.

10. An electronic device, characterized in that: include: A processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to execute a distribution network single-phase grounding fault detection method that implements an adaptive fusion algorithm as described in any one of claims 1 to 6.

Images