Branch relation checking method and device based on voltage records among breadth-first search users

Abstract

The invention discloses a branch relation checking method based on a voltage record between breadth-first search users. The branch relation checking method comprises the following steps: acquiring a rising and falling record between phase-crossing voltage effective values of the users in a transformer area; constructing a voltage change incidence matrix based on the record; constructing a distance matrix between users based on the voltage change incidence matrix; constructing an adjacent matrix based on the distance matrix; executing a breadth-first search algorithm on the adjacent matrix to obtain a branch result; and performing branch relation checking on the branch result comparison file. According to the method, the high-frequency change record of the user voltage is utilized, the voltage correlation characteristics between the users are combined, the correlation matrix and the adjacent matrix of the user voltage change are constructed, branch checking is conducted on the direct or indirect connection relation between the users through the breadth-first search algorithm, the identification efficiency is high, the branch relation identification task can be rapidly completed, and the user experience is improved. And the method has relatively high robustness for diversified branch relations and dynamic changes in the power grid.

Description

Technical Field

[0001] The present invention belongs to the technical field of power systems, and particularly relates to a method for checking branch relationships in low-voltage substations. Background Art

[0002] With the rapid development of the power system and the continuous improvement of the intelligent level, the checking of branch relationships in low-voltage substations has become one of the important technologies in the operation and maintenance of the power system. Branch relationship checking, also known as branch restoration, is a key technology for clarifying the branches to which each user belongs by analyzing the connection relationships between users and branch boxes or meter boxes in the low-voltage distribution network. It is of great significance in aspects such as low-voltage substation network optimization, fault location, line transformation, and improvement of power supply reliability.

[0003] Currently, the distribution network structure of low-voltage substations is complex, usually arranged in a tree-like or ring network structure. With the popularization of intelligent power consumption equipment and the increase in the number of users, this structure has become more complex and dynamic. Traditional branch relationship checking technologies mainly rely on the power consumption information acquisition system. By collecting basic data of user meters (such as current, voltage, etc.) and combining the layout of branch boxes and meter boxes, the branch belonging of users is inferred. However, these traditional methods have the following problems: 1. Limitations in data collection. Most existing power consumption information acquisition systems focus on the statistics of electricity consumption of user meters, and have insufficient ability to collect high-frequency information such as voltage fluctuations and current changes, making it difficult to capture the small differences in electrical connection relationships between users. In addition, the high cost of collection equipment and the complexity of user distribution in the substation further limit the comprehensive coverage of data.

[0004] Challenges of dynamic changes in low-voltage substations. The branch structure of the low-voltage distribution network is dynamic, and branch relationships may change at any time due to reasons such as new user access, equipment expansion, and line transformation. This dynamic characteristic poses high real-time requirements on traditional checking methods based on fixed models or rules, but existing methods often have difficulty in quickly responding. 3. Insufficient utilization of voltage data. Voltage is one of the important parameters reflecting the state of the distribution network, and there is a certain correlation in voltage changes between different users. Especially during load fluctuations or network faults, the voltage changes between users will show a certain pattern. This correlation can be used as an important basis for branch relationship checking, but traditional methods usually only focus on current or electricity consumption data and have insufficient utilization of voltage data for correlation analysis. 4. Insufficient accuracy and anti-interference ability. The voltage data of low-voltage substations is greatly affected by environmental factors and user load fluctuations, and there is a lot of noise in the data, resulting in a decrease in the accuracy of branch relationship checking. Traditional methods often lack effective anti-interference mechanisms when dealing with this noise, easily leading to incorrect branch judgments. Summary of the Invention

[0005] In view of the above problems, the present invention provides a method for checking branch relationships based on breadth - first search of voltage records between users. This method uses high - frequency change records of user voltages, combines the voltage correlation characteristics between users, constructs an association matrix and an adjacency matrix of user voltage changes, and uses the breadth - first search algorithm to check the direct or indirect connection relationships between users for branches.

[0006] The technical solution adopted by the present invention is as follows: A method for checking branch relationships based on breadth - first search of voltage records between users, including:

[0007] Obtain the rising and falling records of the effective values of the voltages of the users in the sub - areas across phases;

[0008] Construct a voltage change association matrix based on the records;

[0009] Construct a distance matrix between users based on the voltage change association matrix;

[0010] Construct an adjacency matrix based on the distance matrix;

[0011] Execute the breadth - first search algorithm on the adjacency matrix to obtain the branch result;

[0012] Compare the branch result with the file to check the branch relationship.

[0013] Obtain the rising and falling records of the effective values of the voltages of the users in the sub - areas across phases, specifically:

[0014] According to the high - frequency recording data of the actual three - phase voltage values of each user in the same sub - area, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value is greater than or equal to 0.1 volt within 0.1 second, and form a voltage change moment record of the user. The record includes: the time point that meets the condition and the corresponding user number; the time point format is year - month - day - hour - minute - second, and the time is in 24 - hour format;

[0015] Select the voltage change moment records of each user to be checked within 1 day of any working day, and record the record of the rise of phase B as Record the record of the fall of phase A as where i represents the user number; A, B, and C are the default phase sequences.

[0016] Construct a voltage change association matrix based on the records, specifically:

[0017] According to the number of users n for which the branch relationship needs to be checked, construct an n×n matrix M; all initial elements of M are 0;

[0018] Assign different numbers to the users to be checked, and the numbers are 1, 2, 3,..., n;

[0019] Assign values to the elements in the voltage incidence matrix M:

[0020] Specifically That is, the voltage change correlation strength between user i and user j. The Count(*) function is used to count the number of elements in the record set, and t represents the time point; Represents the set of time points when the voltage of user i rises and the voltage of user j drops.

[0021] Construct the distance matrix between users, specifically:

[0022] For any two users i and j, their correlation vectors are respectively represented as x i =(x i1 , x i2 ,..., x in}) and x j =(x j1 , x j2 ,..., x jn} The calculation formula for the Euclidean distance d(x i , x j ) is

[0023] where x ik represents the k-th eigenvalue of user i in the correlation vector, and x jk represents the k-th eigenvalue of user j in the correlation vector, and n represents the dimension of the correlation vector;

[0024] The distance matrix D is an N×N symmetric matrix, where N is the total number of users. Calculate the Euclidean distance d(x i , x j ) for all users i, j ∈ {1, 2,..., N} and fill it into the distance matrix D.

[0025] Construct the adjacency matrix based on the distance matrix; perform a breadth-first search algorithm on the adjacency matrix to obtain the branching result; compare the branching result with the archive for branching relationship verification, specifically:

[0026] Construct an N×N adjacency matrix A, initialize all elements to 0, where N is the total number of users; according to the distance matrix, for each pair of users i and j, if the Euclidean distance between user i and j is less than 0.5, set A[i][j] = 1 and A[j][i] = 1. If i = j, then A[i][i] = 1;

[0027] Create an empty set B to store the visited nodes;

[0028] Create a queue Q for storing nodes to be visited, so as to visit each node and its associated nodes layer by layer according to the breadth - first search algorithm, add the starting node u0 to the queue, mark the starting node u0 as visited and add it to set B;

[0029] When the queue Q is not empty, repeat the following steps: Take out a node u from the queue Q; Add the node u to set B; Traverse all neighbor nodes a connected to u in the adjacency matrix A. If A[u][a] = 1, it means there is a direct connection between node u and node a; For each unvisited neighbor node a, mark it as visited and add it to the queue Q; When the queue is empty, return set B, which contains the starting node u0 and all nodes directly or indirectly connected to it, forming a complete branch;

[0030] Compare the result with the system file to screen out abnormal users.

[0031] The present invention also provides a branch relationship verification device based on breadth - first search for voltage records between users, including:

[0032] A record acquisition module, configured to acquire the rising and falling records of the effective voltage values between different phases of users in the power distribution area;

[0033] An association matrix construction module, configured to construct a voltage change association matrix based on the records;

[0034] A distance matrix construction module, configured to construct a distance matrix between users based on the voltage change association matrix;

[0035] A clustering verification module, configured to construct an adjacency matrix based on the distance matrix, perform a breadth - first search algorithm on the adjacency matrix to obtain a branch result, and verify the branch relationship by comparing the branch result with the file.

[0036] The record acquisition module is specifically configured to:

[0037] According to the high - frequency recording data of the three - phase actual voltage values of each user in the same power distribution area, extract and record the corresponding moments when the absolute value of the decrease and increase within 0.1 second of the effective value is greater than or equal to 0.1 volt, form a voltage change time record of the user, and the record includes: the time point that meets the condition and the corresponding user number; The time point format is year - month - day - hour - minute - second, and the time is in 24 - hour format;

[0038] Select the voltage change time records of each user to be verified within 1 day of any working day, record the record of the B - phase rise as Record the record of the A - phase drop as where i represents the user number; A, B, and C are the default phase sequences.

[0039] The association matrix construction module is specifically used for:

[0040] Construct an n×n matrix M according to the number n of users who need to check the branch relationship; all initial elements of M are 0;

[0041] Assign different numbers to the users to be checked, and the numbers are 1, 2, 3,..., n;

[0042] Assign values to the elements in the voltage association matrix M:

[0043] Specifically That is, the voltage change association strength between user i and user j. The Count(*) function is used to count the number of elements in the record set, and t represents the time point; Represents the set of time points when the voltage of user i rises and the voltage of user j drops.

[0044] The distance matrix construction module is specifically used for:

[0045] For any two users i and j, their association vectors are respectively represented as x i =(x i1 ,x i2 ,...,x in}) and x j =(x j1 ,x j2 ,...,x jn}). The calculation formula of the Euclidean distance d(x i ,x j ) is

[0046] where x ik represents the k-th eigenvalue of user i in the association vector, and x jk represents the k-th eigenvalue of user j in the association vector, and n represents the dimension of the association vector;

[0047] The distance matrix D is an N×N symmetric matrix, where N is the total number of users. For all users i, j ∈ {1, 2,..., N}, calculate the Euclidean distance d(x i ,x j ) and fill it into the distance matrix D.

[0048] The clustering verification module is specifically used for:

[0049] Construct an N×N adjacency matrix A, and all elements are initialized to 0, where N is the total number of users; according to the distance matrix, for each pair of users i and j, if the Euclidean distance between user i and j is less than 0.5, set A[i][j] = 1 and A[j][i] = 1. If i = j, then A[i][i] = 1;

[0050] Create an empty set B to store the visited nodes;

[0051] Create a queue Q for storing the nodes to be visited so as to access each node and its associated nodes layer by layer according to the breadth - first search algorithm, add the starting node u0 to the queue, mark the starting node u0 as visited and add it to the set B;

[0052] When the queue Q is not empty, repeat the following steps: Take out a node u from the queue Q; Add the node u to the set B; Traverse all the neighbor nodes a connected to u in the adjacency matrix A. If A[u][a]=1, it means there is a direct connection between node u and node a; For each unvisited neighbor node a, mark it as visited and add it to the queue Q; When the queue is empty, return the set B, which contains the starting node u0 and all the nodes directly or indirectly connected to it, forming a complete branch;

[0053] Compare the result with the system file to screen out abnormal users.

[0054] On the other hand, the present invention provides an electronic device, characterized in that: it includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the branch relationship verification method based on breadth - first search of user - to - user voltage records as described in any one of claims 1 - 5.

[0055] On the other hand, the present invention provides a non - transitory computer - readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the branch relationship verification method based on breadth - first search of user - to - user voltage records as described in any one of claims 1 - 5.

[0056] The present invention has the following beneficial effects:

[0057] 1) No additional hardware investment is required. By analyzing the records of the user voltage change moments, the present invention can achieve high - precision identification only relying on the data of the existing power consumption information acquisition system. Compared with the traditional method, there is no need to install additional hardware devices such as sensors or monitoring devices, significantly reducing the system deployment and operation and maintenance costs.

[0058] 2) Efficient algorithm design. The algorithm of the present invention is designed simply and efficiently, with a short calculation cycle, and can quickly complete the branch relationship identification task, meeting the real - time requirements of the dynamic changes in the low - voltage distribution network. Especially, the design combining hierarchical clustering and breadth - first algorithm greatly improves the identification efficiency.

[0059] 3) The present invention can handle complex tree-like or ring network structures in low-voltage distribution networks and has strong robustness to diverse branch relationships and dynamic changes in the power grid. This enables the method to also exhibit stability and accuracy in complex network structures.

[0060] 4) Through accurate identification of branch relationships, the present invention optimizes the operation and management mode of low-voltage distribution networks, improving the digital and transparent levels of distribution networks. At the same time, the present invention provides important technical support for the intelligent operation and maintenance of low-voltage distribution networks, contributing to improving the power supply reliability and management efficiency of distribution networks. Detailed implementation manners

[0061] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0062] According to the network topology and circuit principle of low-voltage distribution networks, when a user consumes electricity, its current increases, and its own effective voltage value will decrease accordingly. At the same time, the large voltage drop generated by the power consumption in this phase will cause a small increase in the voltage of other phases, and this cross-phase transmission is more obvious within the same branch box. Therefore, the branch relationship verification of the low-voltage substation area can be realized based on this. The first aspect of the present invention provides a branch relationship verification method based on breadth-first search of voltage records between users, and the specific steps are as follows.

[0063] Step 1: Record the rise and fall records of the effective voltage across phases

[0064] Step 1.1: According to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation area, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value within 0.1 second is greater than or equal to 0.1 volt, forming the voltage change moment record of the user. The record includes: the time point that meets the conditions and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format.

[0065] Step 1.2: Screen all records of the decrease of phase A and the increase of phase B between different users according to the recorded three-phase voltages of all users, and screen the records of the decrease of phase A and the increase of phase B of the same user. Where A, B, and C are the default phase sequences.

[0066] Step 2: Construct a voltage change correlation matrix

[0067] Step 2.1: Construct an n×n matrix M according to the number of users n for which the branch relationship needs to be verified. All initial elements of M are 0.

[0068] Step 2.2: Assign different numbers to the users to be verified, numbered 1, 2, 3,..., n.

[0069] Step 2.3: Select the voltage change records of the users to be verified within 1 day on any working day. The records with the B-phase rising are Record the records with the A-phase falling as where i represents the user number.

[0070] Step 2.4: Assign values to the elements in the voltage correlation matrix M.

[0071] Specifically That is, the voltage change correlation strength between user i and user j; The Count(*) function is used to count the number of elements in the record set, and t represents the time point. Represents the set of time points when user i's voltage rises and user j's voltage falls.

[0072] Step 3: Construct a distance matrix D between users using the Euclidean distance:

[0073] For any two users i and j, their correlation vectors are respectively expressed as x i =(x i1 , x i2 ,..., x in} and x j =(x j1 , x j2 ,..., x jn} The calculation formula for the Euclidean distance d(x i , x j ) is

[0074] where x ik represents the kth eigenvalue of user i in the correlation vector, x jk represents the kth eigenvalue of user j in the correlation vector, and n represents the dimension of the correlation vector. The distance matrix D is an N×N symmetric matrix, where N is the total number of users. Calculate the Euclidean distance d(x i , x j ) for all users i, j ∈ {1, 2,..., N} and fill it into the matrix D.

[0075] Step 4: Verification of branch relationship based on breadth-first algorithm

[0076] Step 4.1: Construct an N×N adjacency matrix A, with all elements initialized to 0, where N is the number of users. For each pair of users i and j according to the matrix D calculated above: If the Euclidean distance between users i and j is less than 0.5, set A[i][j] = 1 and A[j][i] = 1; if i = j, then A[i][i] = 1.

[0077] Step 4.2: Create an empty set B to store the visited nodes.

[0078] Step 4.3: Create a queue Q for storing the nodes to be visited so as to visit each node and its associated nodes layer by layer according to the breadth-first search algorithm, and add the starting node u0 to the queue, marking the starting node u0 as visited (i.e., adding it to set B). The starting node u0 is the starting point of the breadth-first search, usually selected by the input user number or a node number in the adjacency matrix. The selection of the starting node can be a representative node in the cluster. For example, select the center point of the cluster, or any node within the cluster as the starting node.

[0079] When the queue Q is not empty, repeat the following steps:

[0080] 1. Take out a node u from the queue Q.

[0081] 2. Add the node u to set B.

[0082] 3. Traverse all the neighbor nodes a connected to u in the adjacency matrix. If A[u][a] = 1, it means there is a direct connection between node u and node a. For each unvisited neighbor node a, mark it as visited and add it to the queue Q.

[0083] When the queue is empty, return set B, which contains the starting node u0 and all the nodes directly or indirectly connected to it, forming a complete branch.

[0084] Step 5: Compare the results with the system files to screen out abnormal users.

[0085] Specific example: The electricity consumption data of 11 branches in 3 substations in a certain community in Nanjing were selected as the experimental data to verify this method. The experimental data were the voltage change time records and 96-point voltage data of 11 branches within 15 consecutive days of a certain month. The acquisition method of the 96-point voltage data was as follows: Each user had the same initial sampling time and was sampled once every 15 minutes. Each user could obtain a total of 96 time-point voltage effective value data per day.

[0086] Table Test Results of the Branch Relationship Recognition Algorithm

[0087]

[0088]

[0089] For the 11 branches in a total of 3 substations included in the test set, the branch relationships are identified respectively, and compared with the prior topological structure. The correct rate of branch relationship identification is 95%. It can be seen from the table that the proposed method can accurately identify the branch relationships, and has good effects over multiple days with high stability.

[0090] An embodiment of the present invention further provides a low-voltage distribution network household-transformer relationship verification device, including:

[0091] A branch relationship verification device based on breadth-first search of voltage records between users, including:

[0092] A record acquisition module, configured to acquire the rising and falling records between the effective values of the voltages of the users in the substation across phases;

[0093] An association matrix construction module, configured to construct a voltage change association matrix based on the records;

[0094] A distance matrix construction module, configured to construct a distance matrix between users based on the voltage change association matrix;

[0095] A clustering verification module, configured to construct an adjacency matrix based on the distance matrix, perform a breadth-first search algorithm on the adjacency matrix to obtain a branch result, and verify the branch relationship by comparing the branch result with the archive.

[0096] The record acquisition module is specifically configured to:

[0097] According to the high-frequency recording data of the actual three-phase voltage values of each user in the same substation, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value is greater than or equal to 0.1 volt within 0.1 second, and form a voltage change moment record of the user. The record includes: the time point that meets the condition and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format;

[0098] Select the voltage change moment records of each user to be verified within 1 day of any working day, and record the record of the rise of phase B as Record the record of the drop of phase A as where i represents the user number; A, B, and C are the default phase sequences.

[0099] The association matrix construction module is specifically configured to:

[0100] Construct an n×n matrix M according to the number n of users for which the branch relationship needs to be verified; all initial elements of M are 0;

[0101] Assign different numbers to the users to be verified, with the numbers being 1, 2, 3,..., n;

[0102] Assign values to the elements in the voltage incidence matrix M:

[0103] Specifically That is, the voltage change correlation strength between user i and user j. The Count(*) function is used to count the number of elements in the record set, and t represents the time point; Represents the set of time points when the voltage of user i rises and the voltage of user j drops.

[0104] The distance matrix construction module is specifically used for:

[0105] For any two users i and j, their correlation vectors are respectively represented as x i =(x i1 , x i2 ,..., x in ) and x j =(x j1 , x j2 ,..., x jn ). The calculation formula for the Euclidean distance d(x i , x j ) is

[0106] where x ik represents the k-th eigenvalue of user i in the correlation vector, and x jk represents the k-th eigenvalue of user j in the correlation vector, and n represents the dimension of the correlation vector;

[0107] The distance matrix D is an N×N symmetric matrix, where N is the total number of users. For all users i, j ∈ {1, 2,..., N}, calculate the Euclidean distance d(x i , x j ) and fill it into the distance matrix D.

[0108] The clustering verification module is specifically used for:

[0109] Construct an N×N adjacency matrix A, with all elements initialized to 0, where N is the total number of users; according to the distance matrix, for each pair of users i and j, if the Euclidean distance between user i and j is less than 0.5, set A[i][j]=1 and A[j][i]=1. If i = j, then A[i][i]=1;

[0110] Create an empty set B to store the visited nodes;

[0111] Create a queue Q for storing nodes to be visited, so as to visit each node and its associated nodes layer by layer according to the breadth-first search algorithm. Add the starting node u0 to the queue, mark the starting node u0 as visited, and add it to set B;

[0112] When the queue Q is not empty, repeat the following steps: Take out a node u from the queue Q; Add the node u to set B; Traverse all neighbor nodes a connected to u in the adjacency matrix A. If A[u][a]=1, it means there is a direct connection between node u and node a; For each unvisited neighbor node a, mark it as visited and add it to the queue Q; When the queue is empty, return set B, which contains the starting node u0 and all nodes directly or indirectly connected to it, forming a complete branch;

[0113] Compare the result with the system file to screen out abnormal users.

[0114] On the other hand, the present invention provides an electronic device, characterized in that it includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the branch relationship verification method based on breadth-first search of user voltage records as described in any one of claims 1-5.

[0115] On the other hand, the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the branch relationship verification method based on breadth-first search of user voltage records as described in any one of claims 1-5.

[0116] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0117] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in one or more flows of the flowchart and / or one or more blocks of the block diagram.

[0118] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction means for implementing the functions specified in one or more flows of the flowchart and / or one or more blocks of the block diagram.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flows of the flowchart and / or one or more blocks of the block diagram.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement without departing from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A branch relationship checking method based on breadth - first search of voltage records between users, characterized in that: Including: Obtain the rising and falling records of the effective voltage between different phases of the users in the substation area; Construct a voltage change correlation matrix based on the records; Construct a distance matrix between users based on the voltage change correlation matrix; Construct an adjacency matrix based on the distance matrix; Execute a breadth-first search algorithm on the adjacency matrix to obtain a branching result; Compare the branching result with the archive for checking the branching relationship.

2. The method for verifying branch relationships based on breadth-first search of voltage records between users according to claim 1, characterized in that: Obtain the rising and falling records of the effective voltage between different phases of the users in the substation area. Specifically: According to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation area, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value within 0.1 second is greater than or equal to 0.1 volt, forming a voltage change moment record of the user. The record includes: the time point meeting the condition and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format; Select the voltage change time records of each user to be verified within one day of any working day, and record the records of the rising of phase B as Record the records of the falling of phase A as where i represents the user number; A, B, and C are the default phase sequences.

3. The branch relationship verification method based on breadth-first search for user voltage records according to claim 1, wherein: Construct a voltage change correlation matrix based on the records. Specifically: Construct an n×n matrix M according to the number n of users whose branching relationship needs to be checked; all initial elements of M are 0; Assign different numbers to the users to be checked, and the numbers are 1, 2, 3,..., n; Assign values to the elements in the voltage correlation matrix M: Specifically That is, the voltage change correlation strength between user i and user j. The Count(*) function is used to count the number of elements in the record set, and t represents the time point; It represents the set of time points when the voltage of user i rises and the voltage of user j drops.

4. The method for verifying branch relationships based on breadth-first search of voltage records between users according to claim 1, characterized in that: Construct a distance matrix between users. Specifically: For any two users i and j, their association vectors are respectively represented as x i =(x i1 , x i2 ,..., x in}) and x j =(x j1 , x j2 ,..., x jn}), and the calculation formula for the Euclidean distance d(x i , x j ) is where x ik represents the k-th eigenvalue of user i in the association vector, and x jk represents the k-th eigenvalue of user j in the association vector, and n represents the dimension of the association vector; The distance matrix D is an N×N symmetric matrix, where N is the total number of users. For all users i, j ∈ {1, 2,..., N}, the Euclidean distance d(x i , x j ) is calculated and filled into the distance matrix D.

5. The method for verifying branch relationship based on breadth - first search of voltage records between users according to claim 1, wherein: Construct an adjacency matrix based on the distance matrix; execute a breadth-first search algorithm on the adjacency matrix to obtain a branching result; compare the branching result with the archive for checking the branching relationship. Specifically: Construct an N×N adjacency matrix A, and initialize all elements to 0, where N is the total number of users; according to the distance matrix, for each pair of users i and j, if the Euclidean distance between users i and j is less than 0.5, set A[i][j] = 1 and A[j][i] = 1, and if i = j, then A[i][i] = 1; Create an empty set B for storing the visited nodes; Create a queue Q for storing the nodes to be visited to access each node and its associated nodes layer by layer according to the breadth-first search algorithm, and add the starting node u0 to the queue, and mark the starting node u0 as visited and add it to the set B; When the queue Q is not empty, repeat the following steps: take out a node u from the queue Q; add the node u to the set B; traverse all neighbor nodes a connected to u in the adjacency matrix A. If A[u][a] = 1, it means there is a direct connection between node u and node a; for each unvisited neighbor node a, mark it as visited and add it to the queue Q; when the queue is empty, return the set B, which contains the starting node u0 and all nodes directly or indirectly connected to it, forming a complete branch; Compare the result with the system archive to screen out abnormal users.

6. A branch relationship checking device based on breadth-first search for voltage records between users, characterized in that: Including: A record acquisition module for obtaining the rising and falling records of the effective voltage between different phases of the users in the substation area; A correlation matrix construction module for constructing a voltage change correlation matrix based on the records; A distance matrix construction module for constructing a distance matrix between users based on the voltage change correlation matrix; A clustering verification module, configured to construct an adjacency matrix based on the distance matrix, perform a breadth-first search algorithm on the adjacency matrix to obtain a branching result, and verify the branching relationship by comparing the branching result with an archive.

7. The branch relationship verification device for user voltage records based on breadth-first search according to claim 6, characterized in that: A record acquisition module, specifically configured to: Extract and record corresponding moments when the absolute value of the decrease and increase of the effective value within 0.1 second is greater than or equal to 0.1 volt according to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same power distribution area, form a record of the voltage change moments of the users, and the record includes: the time points that meet the conditions and the corresponding user numbers; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format; Select the voltage change time records of each user to be verified within 1 day of any working day, and record the records of the B-phase rise as Record the records of the A-phase drop as where i represents the user number; A, B, and C are the default phase sequences.

8. The branch relationship verification device based on breadth-first search for user voltage records according to claim 6, characterized in that: An association matrix construction module, specifically configured to: Construct an n×n matrix M according to the number n of users whose branching relationship needs to be verified; all initial elements of M are 0; Assign different numbers to the users to be verified, and the numbers are 1, 2, 3,..., n; Assign values to the elements in the voltage association matrix M: Specifically That is, the voltage change correlation strength between user i and user j. The Count(*) function is used to count the number of elements in the record set, and t represents the time point; It represents the set of time points when the voltage of user i rises and the voltage of user j falls.

9. The branch relationship verification device based on breadth-first search for user voltage records according to claim 6, characterized in that: A distance matrix construction module, specifically configured to: For any two users i and j, their association vectors are respectively represented as x i =(x i1 , x i2 ,..., x in}) and x j =(x j1 , x j2 ,..., x jn}), and the calculation formula for the Euclidean distance d(x i , x j ) is where x ik represents the k-th eigenvalue of user i in the association vector, and x jk represents the k-th eigenvalue of user j in the association vector, and n represents the dimension of the association vector; The distance matrix D is an N×N symmetric matrix, where N is the total number of users. For all users i, j ∈ {1, 2,..., N}, the Euclidean distance d(x i , x j ) is calculated and filled into the distance matrix D.

10. The branch relationship checking device based on breadth-first search for user voltage records according to claim 6, characterized in that: A clustering verification module, specifically configured to: Construct an N×N adjacency matrix A, with all elements initialized to 0, where N is the total number of users; according to the distance matrix, for each pair of users i and j, if the Euclidean distance between users i and j is less than 0.5, set A[i][j]=1 and A[j][i]=1, and if i = j, then A[i][i]=1; Create an empty set B for storing the visited nodes; Create a queue Q for storing the nodes to be visited so as to visit each node and its associated nodes layer by layer according to the breadth-first search algorithm, add the starting node u0 to the queue, and mark the starting node u0 as visited and add it to the set B; When the queue Q is not empty, repeat the following steps: take out a node u from the queue Q; add the node u to the set B; traverse all neighbor nodes a connected to u in the adjacency matrix A, if A[u][a]=1, it means there is a direct connection between node u and node a; for each unvisited neighbor node a, mark it as visited and add it to the queue Q; when the queue is empty, return the set B, which contains the starting node u0 and all nodes directly or indirectly connected to it, forming a complete branch; Compare the result with the system archive to screen out abnormal users.

11. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the branching relationship verification method based on the breadth-first search of the voltage records between users as described in any one of claims 1-5.

12. A non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the branching relationship verification method based on the breadth-first search of the voltage records between users as described in any one of claims 1-5.

Images