Power distribution network fault section positioning method

Abstract

The invention relates to the technical field of power distribution network fault section positioning, in particular to a power distribution network fault section positioning method, which comprises the steps of determining a topological structure of a power distribution network according to nodes of the power distribution network; obtaining a network description matrix according to the topological structure and the line parameters between the nodes, and establishing a fault information matrix according to information uploaded by the nodes; obtaining a fault discrimination matrix according to the network description matrix and the fault information matrix; and determining the section corresponding to the line only containing the element 1 and the element 0 in the fault judgment matrix as a suspected fault section set. According to the technical scheme provided by the invention, the characteristics of a matrix algorithm and an optimization algorithm are combined, the problems that the fault section cannot be accurately positioned when distortion occurs, and dimension explosion of a large amount of data occurs are effectively solved, and the accuracy and reliability of the power distribution network fault section positioning method are improved.

Description

Technical Field

[0001] The present application relates to the technical field of fault section locating in a distribution network, and in particular to a method for locating a fault section in a distribution network. Background Art

[0002] As an important part of the power system, the distribution network is responsible for the distribution and supply of electric energy. However, due to factors such as aging of distribution equipment, external environmental influences and human operation, frequent failures have become one of the main challenges restricting the safe and stable operation of the distribution network, seriously affecting users' daily life and production activities.

[0003] In recent years, with the widespread application of distributed generation (DG), especially renewable energy such as solar energy and wind energy, they have gradually been integrated into the distribution network due to their advantages of cleanliness, pollution-free and low carbon. However, while this trend brings many benefits, it also makes the topology of the distribution network more complex. When a fault occurs, the bidirectional fault current is difficult to judge, which further increases the difficulty of fault location, making it difficult for traditional fault location methods to cope with complex and changeable working conditions.

[0004] With the rapid development of distribution network automation technology, a large number of feeder terminal units (FTUs) have been widely deployed in distribution networks for real-time monitoring of operating status. At present, the fault section location methods based on FTU are mainly divided into matrix methods and artificial intelligence algorithms. The matrix method uses graph theory to simplify the topological structure of the distribution network into a network description matrix, and combines the fault information uploaded by the FTU to construct a fault information matrix. By performing specific operations on these two types of matrices, a fault location discrimination matrix is ​​generated to achieve fault location. The advantages of this method are simple principle and fast calculation speed, but when the fault information is distorted, the fault section cannot be accurately determined. The artificial intelligence algorithm constructs a mathematical model and uses an optimization algorithm to solve the fault section, which has strong versatility and high fault tolerance. However, in practical applications, artificial intelligence algorithms are prone to fall into local optimal solutions, especially in the iterative process, and the number of iterations is large, which affects the efficiency and accuracy of the algorithm.

[0005] Therefore, it has become a top priority to develop an efficient and accurate distribution network fault location solution to ensure that the power system can maintain high reliability, stability, and safety when facing complex topologies and dynamic environments. Summary of the invention

[0006] In view of this, it is necessary to provide a method for locating a fault section of a distribution network to solve the shortcomings of the prior art, such as the inability to accurately determine the fault section when fault information is distorted, and the inefficiency of fault location.

[0007] A method for locating a fault section of a distribution network, comprising:

[0008] Determine the topological structure of the distribution network according to the nodes of the distribution network;

[0009] Obtaining a network description matrix according to the topology structure and line parameters between nodes, and establishing a fault information matrix according to information uploaded by the nodes;

[0010] Obtaining a fault discrimination matrix according to the network description matrix and the fault information matrix;

[0011] The sections corresponding to the rows of the fault discrimination matrix containing only element 1 and element 0 are determined as a set of suspected fault sections.

[0012] Preferably, the sections corresponding to the rows of the fault discrimination matrix containing only element 1 and element 0 are determined as the suspected fault section set, and then the method further includes:

[0013] Obtain all candidate fault sections in the suspected fault section set by exhaustive method;

[0014] Determine whether the node information of the candidate fault section is distorted;

[0015] If the node information of the candidate fault section is not distorted, the corresponding candidate fault section is determined to be the final fault section;

[0016] If the node information of the candidate fault section is distorted, the candidate fault section is used as the fault hypothesis of the optimization model, and the final fault section is determined by finding the optimal solution.

[0017] Preferably, if the node information of the candidate fault section is distorted, the candidate fault section is used as a fault hypothesis of the optimization model, and the final fault section is determined by finding the optimal solution, including:

[0018] If the node of the suspected fault section is distorted, a switch function is used as an optimization model;

[0019] Calculating the optimal solution of the switching function by a binary whale algorithm;

[0020] If the element 1 appears in the i-th column of the optimal solution, it is determined that the i-th section of the suspected fault section has a fault.

[0021] Preferably, the switching function satisfies:

[0022]

[0023] Among them, I ju (s) and I jd (s) are the switch functions of the upstream and downstream segments of node j, respectively; Ij is the switch function of node j; s j,us 、s j,ds M is the segment status from node j to the upstream power source and the downstream power source; s 、N s is the number of power sources upstream and downstream of node j; M x 、N x is the total number of upstream and downstream sections of node j; sj,u and sj,d are the section states from node j to upstream and downstream; k u , k d Indicates the access status of DG. It is 1 when DG is connected, otherwise it is 0.

[0024] Preferably, determining whether the node information of the candidate fault section is distorted includes:

[0025] Calculating the evaluation function of the candidate fault section;

[0026] If the result of the evaluation function is 0, it is determined that the node information of the candidate fault section is not distorted;

[0027] If the result of the evaluation function is not 0, it is determined that the node information of the candidate fault section is distorted.

[0028] Preferably, the method further includes: verifying that the node information of the candidate fault section is distorted, including:

[0029] Determine the area type of the candidate fault section, the area type includes a load area and other areas;

[0030] If the region type of the candidate fault section is a load region, a load region switch function is used for verification; the load region switch function satisfies:

[0031]

[0032] If the area type of the suspected fault section is other areas, other area switch functions are used for verification; the other area switch functions satisfy:

[0033]

[0034] Among them, I j (s) represents the switching function of node j; S j,Su Represents the node j to the upstream initial node, S j,Sd represents the state of the segment between node j and the downstream electrical terminal node; S j,d represents the status of all downstream segments of node j, S j,u Represents the status of all upstream segments of node j.

[0035] Preferably, verifying that the node information of the candidate fault section is distorted includes:

[0036] Obtaining an ideal state value of the candidate fault section through a switch function corresponding to the region type of the candidate fault section;

[0037] Comparing the ideal state value with the actual state value of the faulty segment node;

[0038] If the actual state value is consistent with the ideal state value, it is determined that the node information of the candidate fault section is not distorted; if the actual state value is inconsistent with the ideal state value, it is determined that the node information of the candidate fault section is distorted.

[0039] Preferably, the optimal solution of the switching function is determined by a fitness function; wherein the solution corresponding to the highest value of the fitness function is determined as the optimal solution of the switching function;

[0040] The fitness function satisfies:

[0041] Among them, I j is the actual fault current information of the jth switch node; I ej (s) is the expected fault current value of the jth switch node; sj is the state of the jth section; N is the total number of switch nodes in the distribution network; M is the total number of sections in the distribution network; θ is the weight coefficient.

[0042] Preferably, the evaluation function satisfies:

[0043]

[0044] f is the fitness value of the suspected fault section; D1 is the number of sections in the area; I j is the fault current information of node j in the area; Ij(s) represents the switching function of node j in the area.

[0045] Preferably, determining the topology of the distribution network according to the nodes of the distribution network includes:

[0046] Determine the node connection relationship of the distribution network;

[0047] Devices without feeder terminal units are ignored, the feeder terminal units are considered as nodes, and the topology of the distribution network is determined.

[0048] Compared with the existing technology, the beneficial effect of the present application is that the technical solution provided by the present application combines the characteristics of matrix algorithm and optimization algorithm, effectively solves the problems of being unable to accurately locate the fault section when distortion occurs and the "dimensional explosion" faced with large amounts of data, and improves the accuracy and reliability of the distribution network fault section location method. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of a method for locating a fault section in a distribution network provided in an embodiment of the present application.

[0050] Figure 2 Schematic diagram A of a flow chart of a method for locating a fault section in a distribution network according to another embodiment of the present application.

[0051] Figure 3 Schematic diagram B of the method for locating a fault section in a distribution network provided in an embodiment of the present application.

[0052] Figure 4 It is a schematic diagram of a 9-node distribution network containing distributed power sources provided in an embodiment of the present application.

[0053] Figure 5 This is an example diagram of a network description matrix of a 9-node distribution network containing distributed power sources provided in an embodiment of the present application.

[0054] Figure 6 This is an example diagram of a fault discrimination matrix of a 9-node distribution network containing distributed power sources provided in an embodiment of the present application.

[0055] Figure 7 It is a schematic diagram of a 33-node distribution network including distributed power sources provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0057] See also Figure 1 , Figure 1 1 is a flow chart of a method for locating a fault section of a distribution network provided in an embodiment of the present application. The method for locating a fault section of a distribution network comprises the following steps:

[0058] S101: Determine the topology of the distribution network according to the nodes of the distribution network.

[0059] First, clarify the connection method between each node within the distribution network (such as substations, switch stations, distributed power sources, etc.), understand the starting and ending nodes of each feeder, and determine the topological structure of the distribution network.

[0060] Furthermore, since the equipment without feeder terminal unit (FTU) can not provide effective real-time monitoring data, it can be ignored and the FTU can be regarded as a node, and the line between two nodes is called a section. In this way, the topology of the distribution network is simplified, the calculation of invalid or uncritical nodes is reduced, and the overall calculation speed is improved.

[0061] S102: Obtain a network description matrix according to the topological structure and line parameters between nodes, and establish a fault information matrix according to information uploaded by the nodes.

[0062] By constructing a network description matrix and a fault information matrix, the complex power grid structure and fault information can be converted into a matrix form that is easy to calculate, simplifying the fault location process.

[0063] Network Description Matrix:

[0064] The network description matrix is ​​an n-order square matrix, where n represents the number of nodes in the distribution network. In the process of forming the network description matrix, the line parameters mentioned usually refer to the electrical characteristics of each line in the distribution network, including but not limited to resistance (R), reactance (X), conductance (G) and susceptance (B). These parameters reflect the energy loss characteristics of the line when transmitting electric energy and the degree of obstruction of the line to the flow of current. For example, for a given distribution network, if there is a line between node i and node j, then the corresponding elements in the network description matrix will reflect certain electrical characteristics of the line, such as the admittance or impedance of the line. If there is no direct connection between nodes i and j, the corresponding matrix element may be 0.

[0065] For example, the power flow direction from the main power source to the load and DG in the distribution network is defined as positive. The network description matrix A is an n-order square matrix, where n is the number of nodes in the distribution network. When a fault occurs in the distribution network section, if node j detects a forward overcurrent, the j-th diagonal element of the fault information matrix F is 1; if node j detects a reverse overcurrent, the j-th diagonal element of the fault information matrix F is -1; if node j detects no overcurrent, the j-th diagonal element of the fault information matrix F is 0.

[0066] Fault information matrix:

[0067] The information uploaded by each node (in this embodiment, the feeder terminal unit) mainly includes the fault information detected at each node. Specifically, the information may include but is not limited to overcurrent detection information, voltage level, phase information and switch status.

[0068] For example, overcurrent detection information: When a fault occurs in the distribution network section, the feeder terminal unit detects an overcurrent condition. If node j detects a forward overcurrent, the jth diagonal element of the fault information matrix is ​​1; if node j detects a reverse overcurrent, the jth diagonal element of the fault information matrix is ​​-1; if node j detects no overcurrent, the jth diagonal element of the fault information matrix is ​​0.

[0069] The information uploaded by each node is combined to form the basic data of the fault information matrix. With these data, a fault information matrix can be constructed, which is used for subsequent fault location analysis. Specifically, the construction of the fault information matrix depends on the flow direction from the main power supply to the load and distributed power supply in the distribution network, as well as the fault information uploaded by each node. For example, when a fault occurs in the distribution network section, if node j detects a forward overcurrent, the jth diagonal element of the fault information matrix is ​​1; if node j detects a reverse overcurrent, the jth diagonal element of the fault information matrix is ​​-1; if node j detects no overcurrent, the jth diagonal element of the fault information matrix is ​​0.

[0070] In addition, due to the presence of distributed power sources in the distribution network, the line parameters also include the impact of these power sources on the electrical characteristics of the network, especially how they change the direction and size of the power flow in the network. In the presence of distributed power sources, the power flow of the distribution network is no longer unidirectional, but may present a bidirectional flow, so the positive direction of the distribution network power flow can also be specified to adapt it to the distribution network containing distributed power sources.

[0071] S103: Obtain a fault discrimination matrix according to the network description matrix and the fault information matrix.

[0072] The fault discrimination matrix is ​​obtained by multiplying the network description matrix and the fault information matrix. By calculating the fault discrimination matrix, the suspected fault section can be preliminarily determined. In particular, when further fault location is required, candidate areas can be provided.

[0073] S104: Determine that the sections corresponding to the rows of the fault discrimination matrix that only contain element 1 and element 0 are the suspected fault section set.

[0074] By multiplying the network description matrix with the fault information matrix to obtain the fault discrimination matrix, the operation of the conventional matrix algorithm is simplified, and a relatively simple fault judgment criterion is proposed, that is, if a row of the matrix contains only 1 and 0 elements, it is considered that the section corresponding to the row has a fault. Among them, the suspected fault section set can include multiple fault sections.

[0075] Therefore, further, after obtaining the set of suspected fault sections, all candidate fault sections in the set of suspected fault sections can be obtained by exhaustive enumeration; determine whether the node information of the candidate fault sections is distorted; if the node information of the candidate fault sections is not distorted, determine the corresponding candidate fault section as the final fault section; if the node information of the candidate fault sections is distorted, use the candidate fault section as the fault hypothesis of the optimization model, and determine the final fault section by finding the optimal solution.

[0076] For example, assuming that the suspected faulty sections initially determined are S1, S2, and S3, all possible candidate faulty sections of the initially determined suspected faulty sections are exhaustively enumerated, such as {S1}, {S2}, {S3}, {S1, S2}, {S1, S3}, {S2, S3}, {S1, S2, S3}. If the node information of the candidate faulty section is not distorted, the corresponding candidate faulty section is determined to be the final faulty section.

[0077] In this embodiment, by exhaustively verifying the suspected fault section, all possible fault section combinations can be comprehensively evaluated to improve the accuracy of fault location. By judging whether the information is distorted or not, the final fault section without distortion can be directly determined, simplifying the process and improving efficiency.

[0078] In addition, if only one combination has an evaluation function of 0, the combination is directly determined as the final fault section. If multiple combinations have evaluation functions of 0, the actual situation can be further analyzed or an optimization algorithm can be used to determine the most reasonable fault section combination.

[0079] Furthermore, the evaluation function of the candidate fault section can be calculated to determine whether the node information of the candidate fault section is distorted. Specifically, if the result of the evaluation function is 0, it means that the combination can fully explain all the fault information, and it is determined that the node information of the candidate fault section is not distorted; if the result of the evaluation function is not 0, it means that the combination cannot fully explain all the fault information, and it is determined that the node information of the candidate fault section is distorted. The evaluation function satisfies:

[0080]

[0081] Where f is the fitness value of the suspected fault section; D1 is the number of sections in the area; I jis the fault current information of node j in the area; I j (s) represents the switching function of regional node j. The calculation method of the fitness value usually depends on the specific application scenario and optimization goal. In the distribution network fault location, the calculation of the fitness value can be based on the fault section matching degree, information consistency, etc. For example, it is evaluated whether the candidate solution is consistent with the topological structure and electrical characteristics of the distribution network. If they are consistent, the fitness value is higher.

[0082] In this embodiment, by calculating the evaluation function, the rationality of the candidate fault section can be quantitatively evaluated to determine whether the information is distorted. In addition, the consistency of the information can also be determined by the result of the evaluation function to ensure the reliability of fault location.

[0083] Furthermore, it is also possible to verify that the node information of the candidate fault section is distorted, including:

[0084] Determine the regional type of the candidate fault section, which includes load area and other areas: Since the distributed power source is connected to the distribution network, the distribution network branch can be divided into load area and other areas according to the flow direction of the current. The load area means that only the load is connected to the end. No matter what kind of fault occurs in the distribution network section, the node can never detect the reverse current. Other areas refer to the distribution network failure, and the nodes in the area may detect the reverse current.

[0085] If the region type of the candidate fault section is a load region, a load region switch function is used for verification; the load region switch function satisfies:

[0086]

[0087] If the area type of the suspected fault section is other areas, other area switch functions are used for verification; the other area switch functions satisfy:

[0088]

[0089] Among them, I j (s) represents the switching function of node j; S j,Su Represents the node j to the upstream initial node, S j,Sd represents the state of the segment between node j and the downstream electrical terminal node; S j,d represents the status of all downstream segments of node j, S j,u Represents the status of all upstream segments of node j.

[0090] The ideal state value of the candidate fault segment is obtained by the switching function corresponding to the regional type of the candidate fault segment; the ideal state value is compared with the actual state value of the fault segment node; if the actual state value is consistent with the ideal state value, it is determined that the node information of the candidate fault segment is not distorted; if the actual state value is inconsistent with the ideal state value, it is determined that the node information of the candidate fault segment is distorted.

[0091] In this embodiment, by dividing the distribution network according to regional types (load areas and other areas), further information consistency assessment is taken, the section information within the area can be analyzed independently, and calculation through partitioning can effectively reduce computing time, further improving the accuracy and efficiency of fault judgment.

[0092] Furthermore, if the node information of the candidate fault section is distorted, the candidate fault section is used as the fault hypothesis of the optimization model, and the final fault section is determined by finding the optimal solution, including: if the node of the suspected fault section is distorted, the switch function is used as the optimization model; the optimal solution of the switch function is calculated by the binary whale algorithm; if the element 1 appears in the i-th column of the optimal solution, it is determined that the i-th section of the suspected fault section has a fault. The switch function satisfies:

[0093]

[0094] Among them, I ju (s) and I jd (s) are the switch functions of the upstream and downstream segments of node j, respectively; I j is the switch function of node j; s j,us 、s j,ds M is the segment status from node j to the upstream power source and the downstream power source; s 、N s is the number of power sources upstream and downstream of node j; M x 、N x is the total number of upstream and downstream sections of node j; sj,u and sj,d are the section states from node j to upstream and downstream; k u , k d Indicates the access status of DG. It is 1 when DG is connected, otherwise it is 0.

[0095] In this implementation, the optimal solution of the switch function refers to the result obtained by solving the optimization algorithm (such as the improved binary whale algorithm), which is the most likely combination of faulty sections. The optimal solution is usually a vector, each element of which corresponds to a section, indicating whether the section has a fault. If an element in the optimal solution is 1, it means that the section has a fault; if it is 0, it means that the section has not a fault. In the fault location process, the optimal solution of the switch function can be used to determine which sections have a fault. The optimal solution helps us identify the faulty section.

[0096] In this embodiment, when there is distortion in the node information of the candidate fault section, an optimization algorithm needs to be further used to solve it. The optimization algorithm can help find the optimal solution; even if there is information distortion, the error can be minimized.

[0097] Furthermore, the optimal solution of the switch function may be determined by a fitness function; wherein the solution corresponding to the highest value of the fitness function is determined as the optimal solution of the switch function;

[0098] The fitness function satisfies:

[0099] Among them, I j is the actual fault current information of the jth switch node; I ej (s) is the expected fault current value of the jth switch node; sj is the state of the jth section; N is the total number of switch nodes in the distribution network; M is the total number of sections in the distribution network; θ is the weight coefficient.

[0100] The fitness function is mainly used in evolutionary algorithms (such as genetic algorithms, particle swarm optimization algorithms, whale optimization algorithms, etc.) to evaluate the quality of candidate solutions. Its main functions include: Evaluating the quality of the solution: The fitness function is used to measure the quality of candidate solutions (i.e. possible solutions). Guiding the search process: Through the value of the fitness function, the algorithm can guide the search direction and gradually approach higher quality solutions. In this embodiment, the fitness function is mainly used to select the optimal solution: at the end of the algorithm switch function, the solution with the highest fitness value is usually considered to be the optimal solution.

[0101] The following is a further introduction to a method for locating a fault section in a distribution network provided by the present application in conjunction with a specific embodiment. Figures 2 to 3 , Figure 2 FIG. 1 is a flow chart A of another method for locating a fault section of a distribution network provided in an embodiment of the present application. Figure 3 This is a flow chart B of the method for locating a fault section in a distribution network provided in an embodiment of the present application.

[0102] The method for locating a fault section of a distribution network comprises the following steps:

[0103] S201: Determine the connection relationship between each node of the distribution network and generate a network description matrix.

[0104] The devices without feeder terminal units are ignored and the FTU is regarded as a node, which simplifies the topology of the distribution network and generates a network description matrix A according to the topology of the distribution network. The network description matrix A is an n-order square matrix, where n is the number of nodes in the distribution network. In this embodiment, the power flow direction from the main power source to the load and DG in the distribution network is regarded as the positive direction, so the matrix elements are determined as follows:

[0105]

[0106] For example, Figure 4 The schematic diagram of the 9-node distribution network with distributed power sources shown in FIG. 1 is a schematic diagram of the 9-node distribution network with distributed power sources, and the network description matrix A of the 9-node distribution network with distributed power sources is as follows: Figure 5 shown.

[0107] S202: Generate a fault information matrix using the fault information uploaded by the FTU.

[0108] Due to the existence of distributed power sources, the power flow of the distribution network is bidirectional, so it is stipulated that any system power source is selected as the main power source, and the direction from the main power source to the load or other power sources is considered to be the positive direction. If it is the same as the fault current direction, it is set to 1, otherwise it is -1. If there is no fault, it is set to 0. The specific representation is as follows:

[0109]

[0110] Assume that DG is put into operation. If section 5 fails, the fault information matrix F = diag[1 1 1 1 1 0 0 -1-1]. If sections 7 and 8 fail at the same time, the fault information matrix F is: F = diag[1 1 1 1 1 1 1 1 -1].

[0111] S203: Obtain a set of suspected fault sections according to the fault discrimination matrix.

[0112] The fault discrimination matrix P is obtained by multiplying the network description matrix and the fault information matrix on the FTU, and the fault location is preliminarily determined according to the fault judgment criteria to form a set of suspected fault sections. The principle for judging suspected fault sections is: if the i-th row elements of the matrix P contain only 1 and 0 elements, it is considered that a fault has occurred in section i. Figure 4 In the distribution network shown, it is assumed that section 5 fails and the information uploaded by the FTU is not distorted.

[0113] At this time, the fault information matrix F uploaded by FTU is F=diag[1 1 1 1 1 0 0 -1-1], so the fault discrimination matrix P is as follows: Figure 6And, it can be concluded that section 5 is suspected to have a fault.

[0114] S204: Determine the area where the suspected fault section is located according to the topological structure and partitions of the distribution network.

[0115] Since DG is connected to the distribution network, the distribution network branch can be divided into load area and other areas according to the flow direction of the power flow. The load area means that only the load is connected to the end. No matter what kind of fault occurs in the distribution network section, the node can never detect the reverse current. The other area refers to the distribution network failure, and the nodes in the area may detect the reverse current.

[0116] S205: Determine whether the information uploaded by the FTU in the area where the suspected fault set is located is distorted by an exhaustive method.

[0117] If the information uploaded by the FTU in the area is not distorted, the fault section is output. If distortion occurs, further analysis is performed.

[0118] Specifically, since the association matrix A contains the association constraints between the segment and node alarm information, even if the distortion of individual node information affects the decision-making ability of the algorithm and increases the number of elements in the fault set, the real fault must be included in it. Therefore, the search dimension of the optimization algorithm can be further reduced through an exhaustive method.

[0119] First, determine the area where the suspected fault section is located and whether the fault area is a load branch or other branch. If the area belongs to the load area, the load area switch function is used to calculate the ideal state of the node. The load area switch function is as follows:

[0120]

[0121] If it belongs to other areas, the ideal state of the node is calculated using the switch function of other areas. The switch function of other areas is as follows:

[0122]

[0123] In the above formula, I j (s) represents the switching function of node j; S j,Su Represents the node j to the upstream initial node, S j,Sd represents the state of the segment between node j and the downstream electrical terminal node; S j,d represents the status of all downstream segments of node j, S j,u Represents the status of all upstream segments of node j.

[0124] The ideal state value of the candidate fault section is obtained through the switching function corresponding to the regional type of the candidate fault section; the ideal state value is compared with the actual state value of the node of the fault section; if the actual state value is consistent with the ideal state value, it is determined that the node information of the candidate fault section is not distorted; if the actual state value is inconsistent with the ideal state value, it is determined that the node information of the candidate fault section is distorted.

[0125] The evaluation function of the fault section satisfies:

[0126] Where f is the fitness value of the suspected fault section; D1 is the number of sections in the area; I j is the fault current information of node j in the area; I j (s) represents the switch function of regional node j. The calculation function formula of fitness value satisfies:

[0127]

[0128] Among them, I j is the actual fault current information of the jth switch node; I ej (s) is the expected fault current value of the jth switch node, which can be calculated by the fault discrimination matrix P; s j is the state of the jth section; N is the total number of switch nodes in the distribution network; M is the total number of sections in the distribution network; θ is the weight coefficient, which can be 0.5 in this embodiment.

[0129] For example, assume that sector 5 fails and the FTU information becomes distorted.

[0130] That is, F = diag[1-111100-11]. The matrix algorithm preliminarily determines that faults occur in sections 5 and 9, extracts the section information of areas 2 and 3 respectively, and exhaustively calculates the fitness values ​​of the above two faults. If a fault occurs in section 5, the ideal state of each node is calculated as:

[0131] I j (s) = [1(4)1(5)0(6)0(7)]

[0132] According to the information actually uploaded by FTU, j =[1(4)1(5)0(6)0(7)], the evaluation function is 0, so the fault section can be directly output. Similarly, when section 9 fails, the ideal state of the node in area 2 is I j (s) = [1(8)1(9)], the evaluation function is 2, so the fault section 9 needs to be re-diagnosed as the fault hypothesis of the optimization algorithm.

[0133] S206: The suspected fault section is used as a fault hypothesis of the optimization model to be solved.

[0134] A switch function model suitable for distributed power sources is established, and the suspected fault section is used as the fault hypothesis of the optimization model for solution through the improved binary whale algorithm. Among them, the solution can be obtained through the switch function. In this embodiment, the switch function is a mathematical model used to describe the relationship between the section state and the node state. Its function is to convert the section state into the state of whether the node has a fault current passing through. The specific formula satisfies:

[0135]

[0136] Among them, I ju (s) and I jd (s) are the switch functions of the upstream and downstream segments of node j, respectively; I j is the switch function of node j; s j,us 、s j,ds M is the segment status from node j to the upstream power source and the downstream power source; s 、N s is the number of power sources upstream and downstream of node j; M x 、N x is the total number of upstream and downstream sections of node j; sj,u and sj,d are the section states from node j to upstream and downstream; k u , k d Indicates the access status of DG. It is 1 when DG is connected, otherwise it is 0.

[0137] The idea of ​​the whale optimization algorithm comes from the foraging behavior of whales. When solving the problem through the improved binary whale algorithm, the whale's position update methods mainly include three methods: surrounding prey, random search, and spiral search. The specific update method is: first, randomly generate a random number between (0,1). If the random number is less than 0.5 and |A| is less than 1, choose to surround the prey to update its position; if the random number is less than 0.5 and |A| is greater than or equal to 1, choose to update its position by random search; if the random number is greater than or equal to 0.5, use spiral search to update the position.

[0138] (1) Surrounding the prey

[0139] When a whale finds prey, it uses echolocation to share the location information of the prey with other whales. Then other whales find the optimal position through iterative updates and update the position towards the prey. The formula is as follows:

[0140]

[0141] Where: t is the number of iterations; X(t) is the position of the whale; X*(t) is the current global optimal whale position; A and C are coefficient matrices.

[0142]

[0143] Where r1 and r2 are random numbers in (0,1); a is a convergence operator varying between 0 and 2; tmax is the maximum number of iterations.

[0144] (2) Random Search Process

[0145]

[0146] Where: Xrand(t) is the whale at a random position.

[0147] (3) Spiral search process

[0148]

[0149] Where: and denote the positions of the ith whale at the tth iteration and the t+1th iteration respectively; represents the whale at the best position of the t-th iteration; b is a constant of the logarithmic spiral shape, usually 1; l is a random constant between [-1,1].

[0150] (4) Binarization of position information

[0151] After the whale position information is updated, it becomes non-binary position information, so each time its position is updated, it needs to be converted to binary through a transfer function. The conversion transfer function satisfies:

[0152]

[0153] S207: Determine the fault area through the local optimal solution.

[0154] If a "1" element appears in the i-th column of the optimal solution, it means that a fault has occurred in the i-th segment.

[0155] Furthermore, through the distribution network fault interval location method of the present application, in the case Figure 7 The simulation experiment is carried out in the 33-node distribution network with distributed generation as shown in Figure 1. The simulation results are shown in Table 1. When a single-point fault, multiple-point fault, or fault information distortion occurs in the distribution network containing DG, the fault location can be accurately performed, and the fault section location can be quickly and effectively performed.

[0156] Fault section Distortion position Matrix algorithm results Optimization algorithm results Fault section L7 none L7 none L7 L24 L10 L24, L10 L24 L24 L31 L7, L19 L7, L31 L31 L31 L20, L23 none L9, L20 none L20, L23 L12, L33 L18, L30 L12, L18, L33 L12, L33 L12, L33

[0157] Table 1

[0158] In summary, the technical solution provided by the present application reduces the amount of calculation in each partition by partitioning the distribution network and simplifying its topological structure, thereby reducing the overall calculation complexity. This means that the section information is analyzed independently in each area, which reduces the calculation time. Then, by multiplying the network description matrix with the fault information matrix to obtain the fault discrimination matrix, the operation of the conventional matrix algorithm is simplified, and a relatively simple fault judgment criterion is proposed. Then, the suspected fault section is further diagnosed by the exhaustive method to determine whether there is information distortion. In this way, the fault section can be directly located without distorted information, thereby avoiding redundant calculations using the optimization algorithm. If there is information distortion, the optimization algorithm is used to solve it, which further reduces the search dimension of the optimization algorithm, that is, the suspected fault section is used as the fault hypothesis of the optimization model, and by establishing a switch function model suitable for distributed power sources and using the improved binary whale algorithm to solve it, it can still maintain good fault tolerance when the information is distorted. Thus, the large amount of data involved in fault location is reduced in dimension, and the fault section can be identified and located more quickly and effectively to ensure the safe and stable operation of the distribution network.

Claims

1. A method for locating a fault section of a distribution network, characterized in that: include: Determine the topological structure of the distribution network according to the nodes of the distribution network; Obtaining a network description matrix according to the topology structure and line parameters between nodes, and establishing a fault information matrix according to information uploaded by the nodes; Obtaining a fault discrimination matrix according to the network description matrix and the fault information matrix; The sections corresponding to the rows of the fault discrimination matrix containing only element 1 and element 0 are determined as a set of suspected fault sections.

2. The method according to claim 1, characterized in that Determine that the sections corresponding to the rows of the fault discrimination matrix containing only element 1 and element 0 are the suspected fault section set, and then further include: Obtain all candidate fault sections in the suspected fault section set by exhaustive method; Determine whether the node information of the candidate fault section is distorted; If the node information of the candidate fault section is not distorted, the corresponding candidate fault section is determined to be the final fault section; If the node information of the candidate fault section is distorted, the candidate fault section is used as the fault hypothesis of the optimization model, and the final fault section is determined by finding the optimal solution.

3. The method according to claim 2, characterized in that If the node information of the candidate fault section is distorted, the candidate fault section is used as the fault hypothesis of the optimization model, and the final fault section is determined by finding the optimal solution, including: If the node of the suspected fault section is distorted, a switch function is used as an optimization model; Calculating the optimal solution of the switching function by a binary whale algorithm; If the element 1 appears in the i-th column of the optimal solution, it is determined that the i-th section of the suspected fault section has a fault.

4. The method according to claim 3, characterized in that The switching function satisfies: Among them, I ju (s) and I jd (s) are the switch functions of the upstream and downstream segments of node j, respectively; I j is the switch function of node j; s j,us 、s j,ds M is the segment status from node j to the upstream power source and the downstream power source; s 、N s is the number of power sources upstream and downstream of node j; M x 、N x is the total number of upstream and downstream sections of node j; sj,u and sj,d are the section states from node j to upstream and downstream; k u , k d Indicates the access status of DG. It is 1 when DG is connected, otherwise it is 0.

5. The method according to claim 2, characterized in that: Determining whether the node information of the candidate fault section is distorted includes: Calculating the evaluation function of the candidate fault section; If the result of the evaluation function is 0, it is determined that the node information of the candidate fault section is not distorted; if the result of the evaluation function is not 0, it is determined that the node information of the candidate fault section is distorted.

6. The method according to claim 5, characterized in that Also includes: Verify that the node information of the candidate fault section is distorted, including: Determine the region type of the candidate fault section, the region type includes load region and other regions; if the region type of the candidate fault section is the load region, verify using the load region switch function; the load region switch function satisfies: If the area type of the suspected fault section is other areas, other area switch functions are used for verification; the other area switch functions satisfy: Among them, I j (s) represents the switching function of node j; S j,Su Represents the node j to the upstream initial node, S j,Sd represents the state of the segment between node j and the downstream electrical terminal node; S j,d represents the status of all downstream segments of node j, S j,u Represents the status of all upstream segments of node j.

7. The method according to claim 6, characterized in that Verify that the node information of the candidate fault section is distorted, including: Obtaining an ideal state value of the candidate fault section through a switch function corresponding to the region type of the candidate fault section; Comparing the ideal state value with the actual state value of the faulty segment node; If the actual state value is consistent with the ideal state value, it is determined that the node information of the candidate fault section is not distorted; if the actual state value is inconsistent with the ideal state value, it is determined that the node information of the candidate fault section is distorted.

8. The method according to claim 4, characterized in that include: Determine the optimal solution of the switch function through the fitness function; wherein the solution corresponding to the highest value of the fitness function is determined as the optimal solution of the switch function; The fitness function satisfies: Among them, I j is the actual fault current information of the jth switch node; I ej (s) is the expected fault current value of the jth switch node; sj is the state of the jth section; N is the total number of switch nodes in the distribution network; M is the total number of sections in the distribution network; θ is the weight coefficient.

9. The method according to claim 5, characterized in that The evaluation function satisfies: f is the fitness value of the suspected fault section; D1 is the number of sections in the area; I j is the fault current information of node j in the area; Ij(s) represents the switching function of node j in the area.

10. The method according to claim 1, characterized in that According to the nodes of the distribution network, the topological structure of the distribution network is determined, including: Determine the node connection relationship of the distribution network; Devices without feeder terminal units are ignored, the feeder terminal units are considered as nodes, and the topology of the distribution network is determined.

Images