Automatic line fault identification method for multi-terminal complex direct-current power supply system

Abstract

The invention belongs to the technical field of direct-current power supply, and particularly relates to an automatic line fault identification method for a multi-terminal complex direct-current power supply system. According to the method, by combining a signal frequency band decomposition and feature value extraction method and a dynamic planning shortest path algorithm for a complex circuit topological structure, accurate processing of fault signals and accurate judgment of fault positions are achieved. Compared with traditional single electrical quantity monitoring or simple threshold value judgment, the comprehensive application method has the advantages that fault features can be recognized more accurately, and the false alarm rate and the missing report rate are reduced. According to the method, the wave head of the first traveling wave is accurately calibrated and the fault distance vector is calculated, so that the high-resistance fault can be reliably identified, the method adapts to the characteristics of a multi-terminal complex direct-current power supply system, and the accuracy and reliability of fault identification are improved. According to the method, the fault point can be quickly responded and positioned through the fault starting self-discrimination function module and subsequent distance vector construction and calculation.

Description

Technical Field

[0001] The present invention belongs to the technical field of direct current power supply, and in particular relates to a method for automatically identifying line faults in a multi-terminal complex direct current power supply system. Background Art

[0002] In modern power supply systems, DC power supply is widely used in long-distance power transmission and distributed energy systems due to its high efficiency and stable power supply characteristics. However, with the expansion of system scale and complexity of structure, multi-terminal DC power supply systems face the challenge of line fault detection and protection. Traditional fault detection methods often rely on data from a single terminal, which may lead to misjudgment or missed judgment in multi-terminal systems, especially in high-resistance faults and complex grid structures.

[0003] Traditional DC power supply system fault detection methods mostly rely on single electrical quantity monitoring (such as voltage and current changes) or simple threshold judgment. These methods may be effective in simple DC systems, but in multi-terminal complex DC power supply systems, due to the mutual coupling between lines and the complex and changeable fault characteristics, it is often difficult to accurately and quickly identify the fault location. Especially in the case of high-resistance faults or strong noise interference, the false alarm and missed alarm rates are high, which seriously affects the stability and reliability of the system. In addition, although the existing fault location technology, such as the traveling wave ranging method, has improved the positioning accuracy to a certain extent, it relies on accurate wave velocity measurement and wave head identification. For multi-terminal complex DC systems, the wave propagation path is complex, the wave head identification is difficult, and it is easily affected by factors such as system structure changes and parameter mismatch, resulting in increased positioning errors.

[0004] Therefore, developing a method that can adapt to the characteristics of multi-terminal complex DC power supply systems and can automatically, quickly and accurately identify line faults has important practical significance and application value for improving the overall stability and reliability of the system and reducing power outages caused by faults. Summary of the invention

[0005] The purpose of the present invention is to propose a method for automatically identifying line faults in a multi-terminal complex DC power supply system in order to solve the problem of line fault detection and location in a multi-terminal complex DC power supply system. The method of the present invention comprehensively applies signal processing, dynamic programming algorithm and electrical quantity monitoring technology, aiming to achieve accurate, fast and reliable identification of line faults in a DC power grid, so as to improve the overall stability and safety of the power supply system.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0007] A method for automatically identifying line faults in a multi-terminal complex DC power supply system comprises the following steps:

[0008] The fault start self-discrimination function module uses the change of multi-terminal voltage Δu and the controllable setting threshold Δu set The logical relationship rule calculation is implemented; the inherent distance vector M ij The construction function module uses the dynamic programming shortest path algorithm to calculate the complex circuit topology of the multi-terminal DC power supply system. The first wave head capture function module uses the combination of signal frequency band decomposition and eigenvalue extraction method to achieve accurate calibration of the first wave head, thereby determining the time signal of the first wave arriving at each terminal. The fault distance vector D ij The self-acquisition function module calculates the difference between the distance from each receiving terminal to the fault point and the distance from the reference terminal to the fault point based on the time signal of the first wave reaching each terminal; the fault section identification vector L ij The functional module is constructed by transforming the intrinsic distance vector M ij and the fault distance vector D ij Obtained after making a difference.

[0009] Further, as a preferred technical solution of the present invention, in the fault start self-discrimination function module, the threshold Δu set The setting needs to be determined based on the maximum value of the voltage change at both ends of the ground transition resistance of the normal line; by comparing the real-time voltage data at both ends of the power supply line with the threshold Δu set The relationship between |Δu| and Δu is used to determine whether the line is abnormal, that is, |Δu|>Δu set An exception occurs when the startup criteria are met.

[0010] Further, as a preferred technical solution of the present invention, the inherent distance vector M ij The construction function module is realized by analyzing the circuit topology structure of the multi-terminal complex DC power supply system using a dynamic programming shortest path algorithm, which can calculate the shortest path length between any two lines, including the following steps:

[0011] S301: Retrieve the DC power supply system topology. If there is a direct connection line (i, j) between power supply terminals i and j, then dist[i][j] = M ij ; If there is no direct connection line (i, j) between power supply terminals i and j, first set dist[i][j] to a very large value;

[0012] S302: For each power supply terminal i, all the line paths reaching itself are initialized to dist[i][i]=0;

[0013] S303: After initializing the inherent distance vector, recursive updating is performed. For each possible intermediate power supply terminal k between two power supply terminals, the shortest line paths of all power supply terminal pairs (i, j) are gradually updated:

[0014] dist[i][j]=min(dist[i][j],dist[i][k]+dist[k][j]) (1)

[0015] Among them, i and j represent the power supply terminal numbers, M ij Represents the shortest line distance between power supply terminals i and j. After initializing its shortest line path matrix dist[i][j], it is gradually updated to obtain the shortest path.

[0016] Further, as a preferred technical solution of the present invention, the capture of the first wave head is achieved by combining signal band decomposition with eigenvalue extraction method to achieve accurate calibration of the wave head, thereby determining the time signal when the first wave arrives at each terminal. The specific steps are as follows:

[0017] S401: First, use phase mode transformation to transform the positive and negative voltage traveling wave signals u at both ends of the line. p 、u n Decoupling is performed to obtain the line mode component u1 and the zero mode component u0. The decoupling operation equation is as follows:

[0018]

[0019] S402: Decompose the line-mode voltage u1 by signal frequency band to obtain multiple single-frequency functions, and update the decomposition results using the quadratic penalty factor α and the Lagrange multiplier index λ. The decomposition model is as shown in equations (3) and (4), where u1 g is the decomposed function component, w k g is the center frequency corresponding to each function component, δ(t) represents the pulse function, is the gradient of each function component; f is the original multi-component signal, and “*” is the convolution calculation;

[0020]

[0021] S403: Take the first decomposed function voltage signal U=[u (1) ,u (2) ,u (3) ,…,u (n) ] is transformed and decomposed by eigenvalue extraction, as shown in formula (5); after obtaining a matrix with equal elements on each sub-diagonal corresponding to the signal, only two eigenvalues ​​can be obtained; among them, the first eigenvalue obtained will be relatively large, while the second eigenvalue will be relatively small, namely the main signal M1 and the detail component D1, wherein the modulus maximum of the detail component realizes the detection of the sudden change point appearing in the signal;

[0022]

[0023] S404: Repeat the decomposition step of S403 for the detail component D1, and continue to select the modulus maximum value of the detail component of the second layer for analysis, so as to achieve the calibration of the wave head and obtain the time signal t k , where k∈(1,2,…,n).

[0024] Further, as a preferred technical solution of the present invention, the fault distance vector D ij The self-acquisition function module calculates the difference between the distance from each receiving terminal to the fault point and the distance from the reference terminal to the fault point based on the acquired time signal. The specific steps are as follows:

[0025] S501: After accurately acquiring the time signal, after multi-terminal communication, the time signals obtained by each receiving end are compared in length, and the time signals are arranged in order from short to long;

[0026] S502: Obtain the time signal t with the smallest value k After that, it is determined that the reference terminal on one side of the line where the fault point is located is S k ;

[0027] S503: The reference terminal on one side of the faulty line is S k , the target fault distance vector is D kj , the calculation formula of the vector value is:

[0028] D kj =||t n -t k |·v1 (6)

[0029] Where v1 is the line mode component of the traveling wave velocity of the fault voltage, (t k ×v1) represents the distance from the fault point to the reference terminal S k The distance value, (t n ×v1) represents the distance from the fault point to the reference terminal S n By taking the absolute value of the difference, we can obtain the distance from each receiving end to the fault point and the receiving end S k The difference in distance to the fault point can determine the fault distance vector D kj Every item included.

[0030] Further, as a preferred technical solution of the present invention, the fault segment identification vector L ij The building block is to transform the inherent shortest distance vector M ij and the fault distance vector D ij After the difference is made, a new vector is obtained, and another reference terminal is determined based on the information of the non-zero items in the vector; for the fault segment identification vector L ijThe non-zero terms in L ia , L ib , if S a With S i There is a direct connection between them, and S b With S i If there is no direct connection between them, the fault segment is judged to be S i -S a If t a <t b , then the fault segment is judged to be S i -S a part.

[0031] The method for automatically identifying line faults in a multi-terminal complex DC power supply system described in the present invention adopts the above technical solution and has the following technical effects compared with the prior art:

[0032] (1) Comprehensive application of signal processing technology and dynamic programming algorithm: The present invention combines signal frequency band decomposition and eigenvalue extraction methods, as well as a dynamic programming shortest path algorithm for complex circuit topology structures, to achieve accurate processing of fault signals and accurate judgment of fault locations. Compared with traditional single electrical quantity monitoring or simple threshold judgment, this comprehensive application method can more accurately identify fault characteristics and reduce false alarm and missed alarm rates.

[0033] (2) Applicable to high-resistance faults and complex systems: The existing technology often has difficulty in accurately and quickly identifying the fault location in the case of high-resistance faults or strong noise interference. The method of the present invention can reliably identify high-resistance faults by accurately calibrating the first-line wave head and calculating the fault distance vector, and adapts to the characteristics of multi-terminal complex DC power supply systems, thereby improving the accuracy and reliability of fault identification.

[0034] (3) Rapid response and location of fault points: The present invention can quickly respond to and locate fault points through the fault-starting self-discrimination function module and the subsequent distance vector construction and calculation. Compared with traditional methods that require manual timed monitoring or rely on complex wave velocity measurement and wave head identification, the method of the present invention can significantly shorten the fault location time and improve the efficiency of fault handling.

[0035] (4) Reduce power outage time and improve system stability: Fast and accurate fault location helps to take timely measures to repair the fault and reduce the power outage time caused by the fault. This is of great significance for ensuring the stable operation of the power supply system and meeting the power demand of users.

[0036] (5) Applicable to multi-terminal DC power supply systems of various scales: The method of the present invention is not limited by the scale of the system and can be widely applied to multi-terminal DC power supply systems of various scales. This makes the method more widely applicable and can meet the needs of systems of different scales. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a topological diagram of the single-pole structure of the four-terminal DC power supply system of the present invention;

[0038] Figure 2 Graph of the transformation decomposition process for eigenvalue extraction;

[0039] Figure 3 The figure is a flow chart of a fault identification method according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0040] The present invention is further explained below in detail with reference to the accompanying drawings so that those skilled in the art can more deeply understand the present invention and be able to implement it. However, the following reference examples are only used to explain the present invention and are not intended to limit the present invention.

[0041] Example 1

[0042] The specific embodiment of the present invention is a method for automatically identifying line faults in a multi-terminal complex DC power supply system. Figure 1 The present invention is applicable to various types of DC lines, including overhead lines and cable lines, to single-pole grounding faults and bipolar short-circuit faults, and to DC transmission networks of various voltage levels and topological structures. It can be used as a backup protection to cooperate with a single-ended ultra-fast main protection method. The specific embodiment of the present invention includes five parts: a fault start self-discrimination function module, an inherent distance vector Mij construction function module, a first-line wave head capture function module, a fault distance vector Dij self-acquisition function module, and a fault section identification vector Lij construction function module.

[0043] The specific embodiment of the present invention is a method for automatically identifying line faults in a multi-terminal complex DC power supply system. Figure 1 As shown, the signal measurement point R 12 and R 21 is the protection measurement point in the fault line area, and L is the current limiting reactor of the DC line.

[0044] In the fault start self-discrimination function module of the specific embodiment of the present invention, the threshold Δu set The setting of needs to be determined based on the maximum value of the voltage change at both ends of the ground transition resistance of the normal line. set The relationship between |Δu| and Δu is used to determine whether the line is abnormal, that is, |Δu|>Δu set An exception occurs when the startup criteria are met.

[0045] The intrinsic distance vector M of the specific embodiment of the present invention ijThe construction function module is realized by analyzing the circuit topology of the multi-terminal complex DC power supply system using the dynamic programming shortest path algorithm, which can calculate the shortest path length between any two lines. It includes the following steps:

[0046] S301: Retrieve the DC power supply system topology. If there is a direct connection line (i, j) between power supply terminals i and j, then dist[i][j] = M ij ; If there is no direct connection line (i, j) between power supply terminals i and j, dist[i][j] is first set to a very large value, indicating that it cannot be reached directly.

[0047] S302: For each power supply terminal i, all the line paths reaching itself are initialized to dist[i][i]=0.

[0048] S303: After initializing the inherent distance vector, recursive updating is performed. For each possible intermediate power supply terminal k between two power supply terminals, the shortest line paths of all power supply terminal pairs (i, j) are gradually updated:

[0049] dist[i][j]=min(dist[i][j],dist[i][k]+dist[k][j]) (1)

[0050] The meaning of formula (1) is to check whether a shorter path can be obtained from power supply terminal i to j through intermediate terminal k. Where i and j represent the power supply terminal numbers respectively, and M ij Represents the shortest line distance between power supply terminals i and j. After initializing its shortest line path matrix dist[i][j], it is gradually updated to obtain the shortest path.

[0051] The capture of the first wave head in the specific embodiment of the present invention is to achieve accurate calibration of the wave head by combining signal frequency band decomposition with eigenvalue extraction method, so as to determine the time signal when the first wave arrives at each terminal. The specific steps are as follows:

[0052] S401: First, use phase mode transformation to transform the positive and negative voltage traveling wave signals u at both ends of the line. p 、u n Decoupling is performed to obtain the line mode component u1 and the zero mode component u0. The decoupling operation equation is shown in formula (2).

[0053]

[0054] S402: Decompose the line-mode voltage u1 by signal frequency band to obtain multiple single-frequency functions. And update the decomposition result by using the quadratic penalty factor α and the Lagrange multiplier index λ. The decomposition model is as shown in equations (3) and (4), where u1 gis the decomposed function component, w k g is the center frequency corresponding to each function component, δ(t) represents the pulse function, is the gradient of each function component; f is the original multi-component signal, and “*” is the convolution calculation.

[0055]

[0056] S403: Take the first decomposed function voltage signal U=[u (1) ,u (2) ,u (3) ,…,u (n) ] is transformed and decomposed by eigenvalue extraction, as shown in formula (5). After obtaining a matrix with equal elements on each sub-diagonal corresponding to the signal, only two eigenvalues ​​can be obtained. Among them, the first eigenvalue obtained will be relatively large, while the second eigenvalue will be smaller, namely the main signal M1 and the detail component D1. The modulus maximum of the detail component can realize the detection of sudden change points in the signal. The decomposition process is as follows Figure 2 shown.

[0057]

[0058] S404: Repeat the decomposition step of S403 for the detail component D1, and continue to select the modulus maximum value of the detail component of the second layer for analysis, so as to achieve the calibration of the wave head and obtain the time signal t k , where k∈(1,2,…,n).

[0059] The fault distance vector D of the specific embodiment of the present invention ij The self-acquisition function module calculates the difference between the distance from each receiving terminal to the fault point and the distance from the reference terminal to the fault point based on the acquired time signal. The specific steps are as follows:

[0060] S501: After accurately acquiring the time signal, after multi-terminal communication, the time signals obtained by each receiving end are compared in length, and the time signals are arranged in order from short to long.

[0061] S502: Obtain the time signal t with the smallest value k After that, it is determined that the reference terminal on one side of the line where the fault point is located is S k .

[0062] S503: The reference terminal on one side of the faulty line is S k , the target fault distance vector is D kj , the calculation formula of the vector value is:

[0063] D kj=|t n -t k |·v1 (6)

[0064] Where v1 is the line mode component of the traveling wave velocity of the fault voltage, (t k ×v1) represents the distance from the fault point to the reference terminal S k The distance value, (t n ×v1) represents the distance from the fault point to the reference terminal S n By taking the absolute value of the difference, we can obtain the distance from each receiving end to the fault point and the receiving end S k The difference in distance to the fault point can determine the fault distance vector D kj Every item included.

[0065] The fault segment identification vector L of the specific embodiment of the present invention ij The building block is to transform the inherent shortest distance vector M ij and the fault distance vector D ij After the difference is made, a new vector is obtained, which is used as a tool to determine the reference terminal on the other side and the specific location of the transmission line where the fault point is located. The other reference terminal is determined based on the information of the non-zero items in the vector. For the fault section identification vector L ij The non-zero terms in L ia , L ib , if S a With S i There is a direct connection between them, and S b With S i If there is no direct connection between them, the fault segment is judged to be S i -S a If t a <t b , it can be determined that the fault segment is S i -S a part.

[0066] The present invention proposes a method for automatically identifying line faults in a multi-terminal complex DC power supply system, aiming to overcome the difficulty of traditional single-ended ultra-fast main protection in identifying line high-resistance faults. The method of the present invention establishes a new fault segment identification vector in the identification of the fault segment, and judges the specific location of the fault point according to the characteristics of each item in the vector. The difference in fault feature quantity caused by the fault direction is expanded, and it has strong anti-interference ability. It can be used as a backup protection method in conjunction with single-ended ultra-fast main protection, meeting the practical application prospects.

[0067] The specific implementation scheme described above further describes in detail the purpose, technical scheme and beneficial effects of the present invention. It should be understood that the above is only a specific implementation scheme of the present invention and is not intended to limit the scope of the present invention. Any equivalent changes and modifications made by any technician in the field without departing from the concept and principle of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for automatically identifying line faults in a multi-terminal complex DC power supply system, characterized in that: The following steps are involved: The fault start self-discrimination function module uses the change of multi-terminal voltage Δu and the controllable setting threshold Δu set The logical relationship rule calculation is implemented; the inherent distance vector M ij The construction function module uses the dynamic programming shortest path algorithm to calculate the complex circuit topology of the multi-terminal DC power supply system. The first wave head capture function module uses the combination of signal frequency band decomposition and eigenvalue extraction method to achieve accurate calibration of the first wave head, thereby determining the time signal of the first wave arriving at each terminal. The fault distance vector D ij The self-acquisition function module calculates the difference between the distance from each receiving terminal to the fault point and the distance from the reference terminal to the fault point based on the time signal of the first wave reaching each terminal; Fault segment identification vector L ij The functional module is constructed by transforming the intrinsic distance vector M ij and the fault distance vector D ij Obtained after making a difference.

2. According to claim 1, a method for automatically identifying line faults in a multi-terminal complex DC power supply system is characterized in that: In the fault start self-judgment function module, the threshold Δu set The setting needs to be determined based on the maximum value of the voltage change at both ends of the ground transition resistance of the normal line; by comparing the real-time voltage data at both ends of the power supply line with the threshold Δu set The relationship between |Δu| and Δu is used to determine whether the line is abnormal, that is, |Δu|>Δu set An exception occurs when the startup criteria are met.

3. The method for automatically identifying line faults in a multi-terminal complex DC power supply system according to claim 1 is characterized in that: The intrinsic distance vector M ij The construction function module is realized by analyzing the circuit topology structure of the multi-terminal complex DC power supply system using a dynamic programming shortest path algorithm, which can calculate the shortest path length between any two lines, including the following steps: S301: Retrieve the DC power supply system topology. If there is a direct connection line (i, j) between power supply terminals i and j, then dist[i][j] = M ij ; If there is no direct connection line (i, j) between power supply terminals i and j, first set dist[i][j] to a very large value; S302: For each power supply terminal i, all the line paths reaching itself are initialized to dist[i][i]=0; S303: After initializing the inherent distance vector, recursive updating is performed. For each possible intermediate power supply terminal k between two power supply terminals, the shortest line paths of all power supply terminal pairs (i, j) are gradually updated: dist[i][j]=min(dist[i][j],dist[i][k]+dist[k][j]) (1) Among them, i and j represent the power supply terminal numbers, M ij Represents the shortest line distance between power supply terminals i and j. After initializing its shortest line path matrix dist[i][j], it is gradually updated to obtain the shortest path.

4. The method for automatically identifying line faults in a multi-terminal complex DC power supply system according to claim 1 is characterized in that: The capture of the first wave head is achieved by combining signal frequency band decomposition with eigenvalue extraction method to achieve accurate calibration of the wave head, thereby determining the time signal when the first wave arrives at each terminal. The specific steps are as follows: S401: First, use phase mode transformation to transform the positive and negative voltage traveling wave signals u at both ends of the line. p 、u n Decoupling is performed to obtain the line mode component u1 and the zero mode component u0. The decoupling operation equation is as follows: S402: Decompose the line-mode voltage u1 by signal frequency band to obtain multiple single-frequency functions, and update the decomposition results using the quadratic penalty factor α and the Lagrange multiplier index λ. The decomposition model is as shown in equations (3) and (4), where u1 g is the decomposed function component, w k g is the center frequency corresponding to each function component, δ(t) represents the pulse function, is the gradient of each function component; f is the original multi-component signal, "*" is the convolution calculation; S403: Take the first decomposed function voltage signal U=[u (1) ,u (2) ,u (3) ,…,u (n) ] is transformed and decomposed by eigenvalue extraction, as shown in formula (5); after obtaining a matrix with equal elements on each sub-diagonal corresponding to the signal, only two eigenvalues ​​can be obtained; among them, the first eigenvalue obtained will be relatively large, while the second eigenvalue will be relatively small, namely the main signal M1 and the detail component D1, wherein the modulus maximum of the detail component realizes the detection of the sudden change point appearing in the signal; S404: Repeat the decomposition step of S403 for the detail component D1, and continue to select the modulus maximum value of the detail component of the second layer for analysis, so as to achieve the calibration of the wave head and obtain the time signal t k , where k∈(1,2,…,n).

5. The method for automatically identifying line faults in a multi-terminal complex DC power supply system according to claim 1 is characterized in that: The fault distance vector D ij The self-acquisition function module calculates the difference between the distance from each receiving terminal to the fault point and the distance from the reference terminal to the fault point based on the acquired time signal. The specific steps are as follows: S501: After accurately acquiring the time signal, after multi-terminal communication, the time signals obtained by each receiving end are compared in length, and the time signals are arranged in order from short to long; S502: Obtain the time signal t with the smallest value k After that, it is determined that the reference terminal on one side of the line where the fault point is located is S k ; S503: The reference terminal on one side of the faulty line is S k , the target fault distance vector is D kj , the calculation formula of the vector value is: D kj =|t n -t k |·v1 (6) Where v1 is the line mode component of the traveling wave velocity of the fault voltage, (t k ×v1) represents the distance from the fault point to the reference terminal S k The distance value, (t n ×v1) represents the distance from the fault point to the reference terminal S n By taking the absolute value of the difference, we can obtain the distance from each receiving end to the fault point and the receiving end S k The difference in distance to the fault point can determine the fault distance vector D kj Every item included.

6. The method for automatically identifying line faults in a multi-terminal complex DC power supply system according to claim 1 is characterized in that: The fault segment identification vector L ij The building block is to transform the inherent shortest distance vector M ij and the fault distance vector D ij After the difference is made, a new vector is obtained, and another reference terminal is determined based on the information of the non-zero items in the vector; For the fault segment identification vector L ij The non-zero terms in L ia , L ib , if S a With S i There is a direct connection between them, and S b With S i If there is no direct connection between them, the fault segment is judged to be S i -S a If t a <t b , then the fault segment is judged to be S i -S a part.

Images