Line selection decision method of off-station type distributed line selection device

Abstract

This invention relates to the field of distribution network fault location technology and discloses a fault location decision method for an off-site distributed fault location device. It utilizes an off-site power grid sensor acquisition system to collect current information from each outgoing line and busbar, establishing typical fault samples and forming a typical fault sample space. Then, a zero-sequence network diagram is constructed based on the current information collected by the power grid sensor acquisition system. Circuit equations are written based on the zero-sequence network diagram, and the adiabatic equivalent of the circuit equations is applied using the servicing principle. Finally, the sequence parameters that indicate the phase transition development trend of the system are solved. This off-site distributed fault location device fault location decision method can correctly solve the uncertainty between fault symptoms and faulty lines, achieving fault location for low-current grounding systems. Furthermore, the correlation analysis-based fault diagnosis requires fewer samples, involves less computation, and is easy to master.

Description

Technical Field

[0001] This invention relates to the field of fault location technology in distribution networks, specifically to a fault location decision method for an off-site distributed fault location device. Background Technology

[0002] With the development of the national economy and society, people have increasingly higher demands for reliable power supply and supervision, especially for the safe operation and timely fault diagnosis of the distribution network, which is closely connected to the public. Single-phase grounding is the most common fault in low-current grounding systems of distribution networks, accounting for over 80%, and many short-circuit faults are also caused by single-phase grounding faults. When a single-phase grounding occurs, there are significant safety hazards such as voltage rise in non-faulty phases, ferroresonant overvoltage, arcing grounding overvoltage, and electric shock to people and animals, requiring immediate disconnection of the faulty power supply. In recent years, the State Grid Corporation of China has improved the technical principles for handling single-phase grounding faults in low-current grounding systems. The principle of "operable for 2 hours," established in the 1970s based on Soviet experience, has been modified to the principle of quickly isolating the fault at the nearest location after surviving a transient grounding fault. This has changed from "2-hour operation + grounding line selection" to "segment tripping." To achieve the goal of quickly distinguishing between faulty and non-faulty lines, power supply units have widely installed low-current grounding line selection devices.

[0003] Currently, there are various methods for selecting fault lines with low current, each with its own advantages and disadvantages. For different fault signal characteristics, each method has certain applicable conditions. When the applicable conditions are met, the selection result of the method will definitely be correct; otherwise, the selection result may be incorrect. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides a fault selection decision method for an off-site distributed fault selection device. This method can correctly resolve the uncertainty between fault symptoms and faulty lines, enabling fault selection in low-current grounding systems. Furthermore, the correlation analysis-based fault diagnosis has advantages such as requiring fewer samples, less computational load, and ease of mastery. It solves the problem that various low-current fault selection methods exist, each with its own advantages and disadvantages. For different fault signal characteristics, each method has certain applicable conditions; when these conditions are met, the fault selection result is guaranteed to be correct; otherwise, the result may be incorrect.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: a route selection decision method for an off-site distributed route selection device, comprising the following steps:

[0008] S1. Use the power grid sensor acquisition system deployed outside the station to collect the current information of each outgoing line and bus, establish typical fault samples, and form a typical fault sample space.

[0009] S2. Using certain data processing methods, transform the data of the fault samples to be detected and typical fault samples.

[0010] S3. Select an appropriate resolution coefficient;

[0011] S4. Calculate the correlation coefficient between the fault sample to be tested and the typical fault sample.

[0012] S5. Calculate the correlation degree between the fault sample to be tested and the typical fault sample using the correlation coefficient, sort the correlation degrees, and obtain the correlation order.

[0013] S6. Based on the correlation sequence, the line with the highest correlation degree is determined to be the faulty line.

[0014] Preferably, the steps for establishing the fault sample space in S1 are as follows:

[0015] 1) List the zero-sequence network diagram for the current information collected by the power grid sensor acquisition system, write the circuit equation based on the zero-sequence network diagram, apply the servo principle to perform adiabatic equivalence on the circuit equation, and finally solve for the sequence parameter of the phase transition development trend of the display system.

[0016] 2) Based on the obtained sequence parameters, select a set of sampled data from the faulty circuit of the system as prototype samples. The number of data points depends on the sampling frequency and time.

[0017] 3) Preprocess the prototype samples by using the K-means clustering algorithm to obtain the initial values ​​of each prototype vector, then normalize and zero-mean the vectors to obtain the prototype vector V. k ,Right now:

[0018]

[0019] 4) Use the inverse method to find the adjoint vector of the prototype vector. Adjoint vector It can be described as V k transpose vector The linear superposition, i.e.

[0020]

[0021] The adjoint vector of the prototype vector It will be calculated step by step as the prototype vector is given.

[0022] 5) Experimental mode q input from each fault location (0) and formula Find the initial values ​​of the order parameters. Under the condition that λ=B=C, use the following formula to perform evolution or directly find the order parameter with the largest absolute value.

[0023]

[0024] That is, ξ k The maximum value of (0) is to directly find the maximum initial sequence parameter value, and the corresponding prototype mode is the fault sample of the line.

[0025] Preferably, in step S2, mean value processing is used when transforming the data of the fault sample to be detected and the typical fault sample;

[0026] Given an original sequence: χ0={χ0(1), χ0(2), ..., χ0(n)}

[0027] Let its average value be:

[0028] Then, after applying the mean value to the sequence χ0, we get:

[0029]

[0030] The sequence Y0 is the result of the sequence χ0 after mean-averaging.

[0031] Preferably, the correlation coefficient in S4 is calculated in the following manner;

[0032] Let χ0 = {χ0(1), χ0(2), ..., χ0(n)} be the reference phasor, χ i ={χ i (1), χ i (2), ..., χ i Let (m)} be the comparison phasor, where i = 1, 2, ..., m, then χ0(k) and χ i The correlation coefficient ξ of (k) 0i (k) can be calculated using the following formula.

[0033]

[0034] In the formula Δ 0i (k)=|χ0(k)-χ i (k)|, where χ0(k) and χ are two sequences. i The absolute difference of (k), Δ max and Δ min For all comparison phasors X i The maximum and minimum values ​​of the absolute difference between the sequence and X0 are ρ, which is the resolution coefficient used to improve the significance of the difference between the correlation coefficients. The value of ρ is [0,1], and it is generally taken as 0.5.

[0035] Preferably, in step S5, the correlation coefficients of each element of each comparison vector are concentrated in one value, which is the correlation degree. This value is used to reflect the magnitude of the relative changes between factors during the development of the system. If the relative changes of two factors are basically the same during the development process, then the correlation degree between the two is considered to be large, and vice versa.

[0036] The correlation between two sequences is calculated as the average of the correlation coefficients between corresponding elements of the two sequences, i.e.:

[0037]

[0038] In the formula γ 0i To compare subphasors X i The degree of correlation with the reference phasor X0, where n is the number of sequence elements.

[0039] (III) Beneficial Effects

[0040] Compared with the prior art, the present invention provides a route selection decision method for an off-site distributed route selection device, which has the following beneficial effects:

[0041] 1. By establishing typical fault samples, a typical fault sample space is formed. The more fault samples contained in the typical fault sample space, the more accurate the line selection. In order to improve the reliability of the criterion, lines with high gray correlation can be used as references for line selection conclusions. Correlation analysis can correctly solve the uncertainty between fault symptoms and fault lines, and realize fault line selection for low current grounding systems. At the same time, fault diagnosis by correlation analysis has the advantages of requiring fewer samples, less analysis and calculation, and ease of mastery. In addition, the essence of gray correlation analysis is gray correlation order, and the relationship between the correlation degree between the fault to be detected and each typical fault sample is relatively ambiguous. In other words, the correlation degree cannot characterize the credibility of the fault to be detected belonging to each typical fault sample. Therefore, the information represented by the gray correlation order is relatively limited. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the route selection decision method of an off-site distributed route selection device proposed in this invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Example 1:

[0045] See attached document Figure 1 A route selection decision method for an off-site distributed route selection device includes the following steps:

[0046] S1. Use the power grid sensor acquisition system deployed outside the station to collect the current information of each outgoing line and bus, establish typical fault samples, and form a typical fault sample space.

[0047] The steps for establishing the fault sample space are as follows:

[0048] 1) List the zero-sequence network diagram for the current information collected by the power grid sensor acquisition system, write the circuit equation based on the zero-sequence network diagram, apply the servo principle to perform adiabatic equivalence on the circuit equation, and finally solve for the sequence parameter of the phase transition development trend of the display system.

[0049] 2) Based on the obtained sequence parameters, select a set of sampled data from the faulty circuit of the system as prototype samples. The number of data points depends on the sampling frequency and time.

[0050] 3) Preprocess the prototype samples by using the K-means clustering algorithm to obtain the initial values ​​of each prototype vector, then normalize and zero-mean the vectors to obtain the prototype vector V. k ,Right now:

[0051]

[0052] 4) Use the inverse method to find the adjoint vector of the prototype vector. Adjoint vector It can be described as V k transpose vector The linear superposition, i.e.

[0053]

[0054] The adjoint vector of the prototype vector It will be calculated step by step as the prototype vector is given.

[0055] 5) Experimental mode q input from each fault location (0) and formula Find the initial values ​​of the order parameters. Under the condition that λ=B=C, use the following formula to perform evolution or directly find the order parameter with the largest absolute value.

[0056]

[0057] That is, ξ k The maximum value of (0) is to directly find the maximum initial sequence parameter value, and the corresponding prototype mode is the fault sample of the line.

[0058] S2. Using certain data processing methods, transform the data of the fault samples and typical fault samples to be detected, and use mean value processing.

[0059] Given an original sequence: χ0={χ0(1), χ0(2), ..., χ0(n)}

[0060] Let its average value be:

[0061] Then, after applying the mean value to the sequence χ0, we get:

[0062]

[0063] The sequence Y0 is the result of the sequence χ0 after mean-averaging.

[0064] S3. Select an appropriate resolution coefficient;

[0065] S4. Calculate the correlation coefficient between the fault sample to be tested and the typical fault sample. The correlation coefficient is calculated in the following way.

[0066] Let χ0 = {χ0(1), χ0(2), ..., χ0(n)} be the reference phasor, χ i ={χ i (1), χ i (2), ..., χ i Let (m)} be the comparison phasor, where i = 1, 2, ..., m, then χ0(k) and χ i The correlation coefficient ξ of (k) 0i (k) can be calculated using the following formula.

[0067]

[0068] In the formula Δ 0i (k)=|χ0(k)-χ i (k)|, where χ0(k) and χ are two sequences. i The absolute difference of (k), Δ max and Δ min For all comparison phasors X i The maximum and minimum values ​​of the absolute difference between the sequence and X0 are ρ, which is the resolution coefficient used to improve the significance of the difference between the correlation coefficients. The value of ρ is [0,1], and it is generally taken as 0.5.

[0069] S5. Calculate the correlation degree between the fault sample to be tested and the typical fault sample using the correlation coefficient, sort the correlation degree to obtain the correlation order, and concentrate the correlation coefficient of each element of each comparison vector into one value. This value is the correlation degree, which is used to reflect the relative change between factors in the process of system development. If the relative change of two factors is basically the same in the process of development, then the correlation between the two is considered to be large, and vice versa.

[0070] The correlation between two sequences is calculated as the average of the correlation coefficients between corresponding elements of the two sequences, i.e.:

[0071]

[0072] In the formula γ 0i To compare subphasors X i The degree of correlation with the reference phasor X0, where n is the number of sequence elements;

[0073] Based on the above correlation calculation, for the reference phasor X0 and the comparison phasor X... i (i = 1, 2, ..., m), where the comparison phasors contain m sequences with correlation degrees of γ. i (i = 1, 2, ..., m), sorting from largest to smallest is called the associative order. Let the associative order be γ1 > γ2 > ... > γ m This indicates that X1 is closest to X0, followed by X2, and X... m In correlation analysis, the most important factor is the order of correlation degree, not the actual magnitude of the correlation degree. The magnitude of the correlation degree is merely an external manifestation of the interaction and mutual influence between elements, while the correlation order is the essence. The goal of correlation analysis is to find the correlation order, and based on the correlation order, the line with the highest correlation degree is determined as the faulty line.

[0074] Obviously, the more fault samples the typical fault sample space contains, the more accurate the fault selection will be. In order to improve the reliability of the criterion, lines with a high degree of gray correlation can be used as a reference for the fault selection conclusion. Correlation analysis can correctly solve the uncertainty between fault symptoms and fault lines, and realize fault selection for low-current grounding systems. At the same time, fault diagnosis by correlation analysis has the advantages of requiring fewer samples, less analysis and calculation, and ease of use. In addition, the essence of gray correlation analysis is gray correlation order, but the relationship between the correlation degree between the fault to be detected and each typical fault sample is relatively ambiguous. In other words, the correlation degree cannot characterize the credibility of the fault to be detected belonging to each typical fault sample. Therefore, the information represented by the gray correlation order is limited. To address this problem, by integrating the fault feature quantities of multiple fault selection methods, the credibility problem of fault selection can be solved better. The more typical fault samples established, the more reliable and accurate the fault selection result will be. Since the data involved in the diagnosis often have different physical dimensions and the values ​​differ greatly, it is necessary to perform dimensionless and normalized data processing.

[0075] It should be noted that the term "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0076] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A route selection decision method for an off-site distributed route selection device, characterized in that, Includes the following steps: S1. Using an off-site power grid sensor acquisition system, collect current information for each outgoing line and busbar to establish typical fault samples, forming a typical fault sample space. The steps for establishing the fault sample space are as follows: 1) List the zero-sequence network diagram for the current information collected by the power grid sensor acquisition system, write the circuit equation based on the zero-sequence network diagram, apply the servo principle to perform adiabatic equivalent on the circuit equation, and finally solve for the sequence parameter of the phase transition development trend of the display system. 2) Based on the obtained sequence parameters, select a set of sampled data from the faulty circuit of the system as prototype samples. The number of data points depends on the sampling frequency and time. 3) Preprocess the prototype samples by using the K-means clustering algorithm to obtain the initial values ​​of each prototype vector, then normalize and zero-mean the vectors to obtain the prototype vectors. ,Right now: 4) Use the inverse method to find the adjoint vector of the prototype vector. , accompanying vector Can be described as transpose vector The linear superposition, i.e. The adjoint vector of the prototype vector It will be calculated step by step as the prototype vector is given. 5) Experimental modes input from each fault location and formula Find the initial values ​​of the order parameters, in Under these conditions, the following formula is used for evolution or the order parameter with the largest absolute value is directly searched; Right now The maximum value is obtained by directly finding the maximum initial sequence parameter value, and the corresponding prototype mode is the fault sample of the line. S2. Using a certain data processing method, the fault samples to be detected and typical fault samples are transformed. The data processing method adopts mean value processing. Given the original sequence: Let its average value be: Then for the sequence The result is obtained after mean normalization. sequence That is, a sequence The result after averaging; S3. Select an appropriate resolution coefficient; S4. Calculate the correlation coefficient between the fault sample to be tested and the typical fault sample. S5. Calculate the correlation degree between the fault sample to be tested and the typical fault sample using the correlation coefficient, sort the correlation degrees, and obtain the correlation order. S6. Based on the correlation sequence, the line with the highest correlation degree is determined to be the faulty line.

2. The route selection decision method of an off-site distributed route selection device according to claim 1, characterized in that: The correlation coefficient in S4 is calculated in the following way; set up For reference phasor, For comparison phasors, where ,but and correlation coefficient It can be calculated using the following formula: In the formula , for two sequences and The absolute difference and For all comparison phasors The sequence and The maximum and minimum values ​​in the absolute difference, The resolution coefficient is used to improve the significance of differences between correlation coefficients. The value range of is [0,1].

3. The route selection decision method of an off-site distributed route selection device according to claim 1, characterized in that: In step S5, the correlation coefficients of each element of each comparison vector are concentrated in one value, which is the correlation degree. This value is used to reflect the magnitude of the relative changes between factors during the development of the system. If the relative changes of two factors are basically the same during the development process, then the correlation degree between the two is considered to be large, and vice versa. The correlation between two sequences is calculated as the average of the correlation coefficients between corresponding elements of the two sequences, i.e.: In the formula For comparison subphasors Compared with the reference phasor The degree of correlation, where n is the number of elements in the sequence.

Images