A low-voltage area undisturbed topology identification method and system, medium and terminal

Abstract

This invention relates to the field of smart grid technology, specifically a method and system for non-disruptive topology identification of low-voltage distribution areas, including a medium and a terminal. The method comprises: S10, cleaning and filtering raw communication and measurement data; S20, converting the RSSI between devices into distances between nodes and forming an undirected weighted graph; S30, using a clustering algorithm to determine the affiliation between meter boxes and meters; S40, constraining and matching transformers, lines, and meter box nodes based on electricity consumption information, and optimizing the topology using the law of conservation of energy and Kirchhoff's current law; and S50, using adjusted RAND exponents and normalized mutual information to evaluate the accuracy of the identification results. This invention features a simple process and convenient operation. Without adding extra hardware, interrupting power supply, or injecting feature signals, it gradually constructs a complete topology based on existing smart meters, communication signals, and power data from electricity consumption information collection systems, achieving high-precision and automated identification of distribution area topology.

Description

Technical Field

[0001] This invention belongs to the field of smart grid technology, and in particular relates to a method and system for non-disruptive topology identification of low-voltage distribution areas, as well as a medium and terminal. Background Technology

[0002] The low-voltage distribution network topology is fundamental information for distribution network operation analysis and management, and is crucial for the safe and reliable operation of the power grid. Accurately understanding the hierarchical relationships between transformers, branch boxes, meter boxes, and meters within a distribution area is a prerequisite for advanced applications such as line loss analysis, load forecasting, and fault location. However, many medium- and low-voltage distribution areas currently suffer from inaccurate or missing topology data. The main reasons include: firstly, the diverse and widespread installation of meters, coupled with frequent adjustments to actual power supply lines due to expansion, renovation, and temporary connections, leading to discrepancies between actual wiring and records; secondly, existing topology identification methods, such as those based on injected current and those based on big data analysis, have limitations and struggle to operate stably in large-scale, real-time-critical scenarios.

[0003] Patent CN115842342B discloses a method and apparatus for topology identification of distributed distribution networks, applied in a distribution network dispatching platform. Based on node information in the distribution network, it divides the network into topology identification regions, including a first identification region, a second identification region, and a third identification region. The topology structure of the first identification region is identified using high-frequency power line carrier communication, and the first identification region includes at least one generating node. The second identification region is identified using a micro-current injection method, and the second identification region includes at least one control node. The third identification region is identified using load parameter metering, and the nodes at the end of the third identification region are load nodes. The topological relationships of edge nodes in the distribution network are identified. Based on the identification results within the regions and the topological relationships of the edge nodes, the distribution network topology is constructed. This patent uses a micro-current injection method for topology identification, which makes it difficult to guarantee the accuracy of the identification results, exhibiting the same drawbacks as existing technologies.

[0004] Therefore, how to provide a high-precision method for identifying the topology of low-voltage distribution areas is a problem that urgently needs to be solved by those in this technical field. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide a non-disruptive topology identification method for low-voltage distribution areas, thereby solving the problem of low accuracy in low-voltage distribution area topology identification in existing technologies. In addition, this invention also provides a non-disruptive topology identification system, medium, and terminal for low-voltage distribution areas.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a method for non-disruptive topology identification of low-voltage distribution areas, comprising the following steps:

[0008] S10. Clean and filter the raw communication and measurement data;

[0009] S20. Convert the RSSI between devices into the distance between nodes and form an undirected weighted graph;

[0010] S30. Use clustering algorithms to determine the ownership relationship between meter boxes and electricity meters;

[0011] S40. Based on the electricity consumption information, perform constraint matching on the transformer, line and meter box nodes, and optimize the topology using the law of conservation of energy and Kirchhoff's current law.

[0012] S50. Use the adjusted RAND index and normalized mutual information to evaluate the accuracy of the identification results.

[0013] Furthermore, in S10, the total number of times each node appears in the communication list is counted and regarded as the degree of the graph. A threshold k is set. For nodes with a degree less than k, they are marked as low-connectivity nodes and removed or processed separately from the subsequent adjacency matrix construction and clustering.

[0014] Furthermore, in S20, each meter or meter box in the power network topology is considered as a node in a graph, and the communication connections between nodes are considered as undirected edges. Let the graph... It is an undirected graph, where Represents a set of nodes, where Indicates the number of nodes. Represents the set of edges, adjacency matrix for Matrix, matrix elements Represents a node With nodes The connection weights between nodes are determined by the fact that if there is no effective communication between nodes, they are considered disconnected, and the corresponding weights are reset to infinity. Defined as:

[0015]

[0016] For nodes that can receive signals, their weights are represented as follows:

[0017]

[0018] in Represents a node With nodes Between signal strength, It is an environmental attenuation factor that controls the signal attenuation rate.

[0019] Furthermore, the adjacency matrix construction process is as follows:

[0020] S201. Initialize the adjacency matrix: Construct the adjacency matrix based on the denoised node list. Set the diagonal elements to 0 and the rest to 0. ;

[0021] S202, Signal strength conversion: For each communication record ,like If the value is not 0, calculate the distance using the formula above. ;

[0022] S203, Adjacency Matrix Filling: Fill the adjacency matrix according to the above definition;

[0023] S204, Symmetry: Since the topological graph is undirected, therefore .

[0024] Furthermore, in step S30, a meter box and a set of electricity meters are provided. The corresponding similarity matrix The algorithm execution process is as follows:

[0025] S301. Initialization: Treat each node as an independent cluster. Initial number of clusters ;

[0026] S302. Distance Calculation: The distance between all clusters is calculated using the adjacency matrix. The distance between clusters is defined as the average distance between all pairs of nodes in two clusters, as shown in the following formula:

[0027]

[0028] in, and Indicates two different clusters. and Representing clusters and The number of samples in Point With point The distance between them is represented by the Euclidean distance;

[0029] S303. Merge Clusters: Based on the calculated distance, select the two most similar clusters to merge and update the cluster set;

[0030] S304. Termination condition: Continue merging until the number of clusters reaches a preset value. ;

[0031] S305, Cluster Center Allocation: For each cluster... The node with the largest communication weight and the average weight of the meters in the cluster is selected as the candidate parent node of the meter box.

[0032] Furthermore, a post-processing mechanism is introduced to address low-connectivity nodes whose cluster affiliation is not explicitly defined. Based on its connection relationships with its already assigned nodes, analyze the table cluster to which it most likely belongs. The specific steps are as follows:

[0033] S311, Node sorting: Count the number of neighbors of each low-connectivity node that have been assigned to a cluster, sort them in descending order, and prioritize the processing of nodes with richer information;

[0034] S312, Find Connected Neighbors: For each low-connectivity node Find all its neighboring nodes in the communication data. Sort the nodes in descending order based on whether the neighboring nodes belong to a cluster.

[0035] S313. Extract Neighbor Attribution: Obtain the cluster label of the node whose attribution has been determined among the above neighbor nodes. The frequency of each cluster label was counted and recorded as the number of matches. ;

[0036] S314. Calculate the affiliation score: For each possible cluster... Combined with connection weights and number of matches Calculate the attribution score As shown in the following formula:

[0037]

[0038] in, Represents a node With already belonged to neighboring nodes The connection weights, , Indicates the weighting coefficient;

[0039] S315. Determine the final cluster: Find the cluster with the highest score. As shown in the following formula:

[0040]

[0041] and low connection nodes Belong to this cluster;

[0042] S316. Exception Handling: If none of the neighbors belong to a cluster, then retain... The status is unrecognized, and a log is recorded.

[0043] Furthermore, the specific steps of S40 are as follows:

[0044] S401. For each potential parent node, candidate selection is performed based on electricity consumption information data to determine a set of possible child nodes;

[0045] S402. For each parent node, enumerate all possible combinations of its child nodes;

[0046] S403. Calculate the line loss and leakage current according to the law of conservation of energy and Kirchhoff's current law:

[0047] Law of Conservation of Energy: Therefore, line loss ,in Represents the set of all child nodes;

[0048] in, Indicates the parent node's electrical energy value. This represents the electrical energy value of each child node;

[0049] Kirchhoff's Current Law: Therefore, leakage current ;

[0050] in, Indicates the current value of the parent node. This represents the current value of each child node;

[0051] S404, Requirements for line loss Leakage current ;

[0052] in, This indicates the maximum allowable line loss threshold. This represents the minimum allowable line loss threshold. This indicates the maximum permissible leakage current threshold. Indicates the minimum allowable leakage current threshold;

[0053] S405. Define the comprehensive evaluation indicators as follows:

[0054]

[0055] in, , These are weighting parameters used to control the importance of line loss and leakage current;

[0056] S406. For each parent node, select the combination of child nodes that results in the best overall performance and update the topology.

[0057] Secondly, the present invention also provides a low-voltage distribution area uninterrupted topology identification system, comprising:

[0058] The data preprocessing module is used to clean and filter the raw communication and measurement data;

[0059] The adjacency matrix construction module is used to convert the RSSI between devices into the distance between nodes and form an undirected weighted graph;

[0060] The meter box-household identification module is used to determine the ownership relationship between the meter box and the electricity meter based on a clustering algorithm;

[0061] The transformer-line-box identification module is used to constrain and match the nodes of transformers, lines and meter boxes according to power consumption information, and optimize the topology using the law of conservation of energy and Kirchhoff's current law.

[0062] The results evaluation module is used to evaluate the accuracy of the identification results based on the adjusted RAND index and normalized mutual information.

[0063] Thirdly, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described above.

[0064] Fourthly, the present invention also provides an electronic terminal, comprising: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the terminal to perform the method described above.

[0065] Compared with existing technologies, the low-voltage distribution area non-disruptive topology identification method, system, medium, and terminal provided by this invention have at least the following advantages:

[0066] Currently, many medium and low-voltage distribution areas suffer from inaccurate or missing topology data, making it difficult to guarantee high accuracy in topology identification. This invention offers a simple and convenient process, achieving deep coupling between RSSI / SNR communication characteristics and energy metering characteristics. It constructs a multi-source data fusion framework for topology identification, overcoming the limitations of single-signal dependence. By replacing signal injection with algorithmic modeling, it achieves uninterrupted topology inference throughout the entire process, enabling real-time operation under energized conditions and demonstrating engineering deployment feasibility. For environments with sparse or weak connections, a dynamic completion algorithm is proposed, effectively compensating for the identification defects of isolated nodes and significantly improving topology integrity. Using the energy conservation equation and Kirchhoff's current law as criteria, the results conform to statistical and electrical physics laws, ensuring the rationality and reliability of the identification results. In summary, this invention, without adding extra hardware, interrupting power, or injecting feature signals, relies solely on existing smart meters, communication signals, and power data from electricity consumption information collection systems to gradually construct a complete topology structure, achieving high-precision and automated identification of distribution area topology and ensuring high accuracy in low-voltage distribution area topology identification. Attached Figure Description

[0067] To more clearly illustrate the solution of the present invention, a brief introduction will be given to the drawings used in the description of the embodiments below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0068] Figure 1 A flowchart of a method for non-disruptive topology identification of low-voltage distribution areas provided in an embodiment of the present invention;

[0069] Figure 2 This is a schematic diagram of a clustering algorithm in a low-voltage transformer area non-disruptive topology identification method provided in an embodiment of the present invention;

[0070] Figure 3 This is a topology diagram of an experimental station area in a low-voltage station area non-disruptive topology identification method provided in an embodiment of the present invention.

[0071] Figure 4 This is a tree diagram of the clustering process for box-to-household identification based on a clustering algorithm in a low-voltage distribution area non-disruptive topology identification method provided in an embodiment of the present invention;

[0072] Figure 5 This is a clustering confusion heatmap for box-household identification based on a clustering algorithm in a low-voltage transformer area non-disruptive topology identification method provided in an embodiment of the present invention.

[0073] Figure 6 This invention provides a connection matrix diagram for identifying transformer-line-box connection relationships based on an electricity consumption information matching algorithm in a low-voltage distribution area non-disruptive topology identification method.

[0074] Figure 7 The diagram shows the output topology identification results of the experimental station area in a low-voltage station area non-disruptive topology identification method provided in an embodiment of the present invention. Detailed Implementation

[0075] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0076] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0077] This invention provides a non-disruptive topology identification method for low-voltage distribution areas, applied in the process of identifying the topology of low-voltage distribution areas. The non-disruptive topology identification method for low-voltage distribution areas includes the following steps:

[0078] S10. Clean and filter the raw communication and measurement data; S20. Convert the RSSI between devices into distances between nodes and form an undirected weighted graph; S30. Use clustering algorithms to determine the affiliation between meter boxes and meters; S40. Perform constraint matching on transformers, lines and meter box nodes based on electricity consumption information, and optimize the topology using the law of conservation of energy and Kirchhoff's current law; S50. Use adjusted Rand index and normalized mutual information to evaluate the accuracy of the identification results.

[0079] The invention has a simple process and is easy to operate. Without adding extra hardware, without interrupting power supply, and without injecting feature signals, it relies on the power data of existing smart meters, communication signals, and power consumption information collection systems to gradually build a complete topology structure, thereby achieving high-precision and automated identification of the transformer area topology structure.

[0080] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0081] This invention provides a non-disruptive topology identification method for low-voltage distribution areas, applied in the process of identifying the topology of low-voltage distribution areas, combined with... Figures 1 to 7 In this embodiment, the low-voltage distribution area non-disruptive topology identification method includes the following steps:

[0082] S10. Data preprocessing: Cleaning and filtering the raw communication and measurement data.

[0083] Specifically, in this embodiment, data preprocessing includes:

[0084] Data loading: Read all communication records between devices and count the number of connections (neighbors) of each node. To improve the accuracy of subsequent identification, low-connection nodes with very few connections can be filtered out immediately after data loading to prevent them from affecting the clustering and other algorithm effects as noise.

[0085] Low-connectivity node filtering: Low-connectivity nodes refer to nodes in the communication record that are connected to only a few other nodes. These nodes may be isolated boxes or terminal devices, and are easily considered noise during clustering.

[0086] Filtering strategy: Count the total number of times each node appears in the communication list (considered as the degree of the graph), and set a threshold k. For nodes with a degree less than k, mark them as "low-connectivity nodes" and remove them from subsequent adjacency matrix construction and clustering, or process them separately.

[0087] Parameter settings: The threshold k is an adjustable parameter, and the cluster size should be set empirically based on the network size and density. If the filtering is too strict, valid nodes will be missed; if it is too lenient, noise will still be retained. Optimization can be performed based on historical data from the distribution area.

[0088] S20. Adjacency Matrix Construction: Convert the RSSI (Signal Strength Indicator) between devices into distances between nodes and form an undirected weighted graph.

[0089] Specifically, in this embodiment, the adjacency matrix is ​​defined as follows: In the power network topology, each meter or meter box is considered a node in a graph, and the communication connections between nodes are considered undirected edges. Let the graph... It is an undirected graph, where Represents a set of nodes, where Indicates the number of nodes. Represents the set of edges. Adjacency matrix. for Matrix, matrix elements Represents a node With nodes The connection weights between nodes are set such that if there is no effective communication between nodes, they are considered disconnected, and the corresponding weights can be set to infinity. It can be defined as:

[0090]

[0091] For nodes that can receive signals, their weights are represented as follows:

[0092]

[0093] in Represents a node With nodes Between signal strength, It is an environmental attenuation factor that controls the signal attenuation rate.

[0094] Furthermore, in this embodiment, the adjacency matrix construction process is as follows:

[0095] S201. Initialize the adjacency matrix: Construct the adjacency matrix based on the denoised node list. Set the diagonal elements to 0 and the rest to 0. ;

[0096] S202, Signal strength conversion: For each communication record ,like If the value is not 0, calculate the distance using the formula above. ;

[0097] S203, Adjacency Matrix Filling: Fill the adjacency matrix according to the above definition;

[0098] S204, Symmetry: Since the topological graph is undirected, therefore .

[0099] S30. Meter-household identification based on cluster analysis: Using clustering algorithms to determine the ownership relationship between meter boxes and electricity meters.

[0100] Specifically, in this embodiment, clustering analysis is a typical unsupervised learning method, suitable for grouping problems when samples are unlabeled. Its goal is to maximize intra-cluster similarity and minimize inter-cluster differences. Considering the natural similarity of communication characteristics between meter nodes, this embodiment uses a hierarchical clustering algorithm for initial meter-household identification. The basic idea is to calculate the distance (similarity) between nodes and gradually merge the most similar nodes or clusters until a set number of clusters is met, such as... Figure 2 As shown.

[0101] Assume there is a meter box and a set of meters. The corresponding similarity matrix The algorithm execution process is as follows:

[0102] S301. Initialization: Treat each node as an independent cluster. Initial number of clusters ;

[0103] S302. Distance Calculation: The distance between all clusters is calculated using the adjacency matrix. The distance between clusters is defined as the average distance between all pairs of nodes in two clusters, as shown in the following formula:

[0104]

[0105] in, and Indicates two different clusters. and Representing clusters and The number of samples in Point With point The distance between them is represented by the Euclidean distance;

[0106] S303. Merge Clusters: Based on the calculated distance, select the two most similar clusters to merge and update the cluster set;

[0107] S304. Termination condition: Continue merging until the number of clusters reaches a preset value. This value is generally obtained based on prior knowledge;

[0108] S305, Cluster Center Allocation: For each cluster... The node with the largest communication weight and the average weight of the meters in the cluster is selected as the candidate parent node of the meter box.

[0109] Furthermore, in this embodiment, in actual communication scenarios, some nodes, due to poor signal quality or unstable communication, only establish connections with a small number of nodes and are identified as low-connectivity nodes. To avoid noise propagation interfering with the main cluster structure, these nodes are usually temporarily removed during the initial data preprocessing stage. To improve the completeness and accuracy of topology identification, this embodiment introduces a post-processing mechanism to optimize and complete the attribution relationships of low-connectivity nodes.

[0110] Specifically, in this embodiment, low-connectivity nodes that do not have a clearly defined cluster affiliation are considered. Based on its connection relationships with its already assigned nodes, analyze the table cluster to which it most likely belongs. The specific steps are as follows:

[0111] S311, Node sorting: Count the number of neighbors of each low-connectivity node that have been assigned to a cluster, sort them in descending order, and prioritize the processing of nodes with richer information;

[0112] S312, Find Connected Neighbors: For each low-connectivity node Find all its neighboring nodes in the communication data. Sort the nodes in descending order based on whether the neighboring nodes belong to a cluster.

[0113] S313. Extract Neighbor Attribution: Obtain the cluster label of the node whose attribution has been determined among the above neighbor nodes. The frequency of each cluster label was counted and recorded as the number of matches. ;

[0114] S314. Calculate the affiliation score: For each possible cluster... Combined with connection weights and number of matches Calculate the attribution score As shown in the following formula:

[0115]

[0116] in, Represents a node With already belonged to neighboring nodes The connection weights, , Indicates the weighting coefficient;

[0117] S315. Determine the final cluster: Find the cluster with the highest score. As shown in the following formula:

[0118]

[0119] and low connection nodes Belong to this cluster;

[0120] S316. Exception Handling: If none of the neighbors belong to a cluster, then retain... The status is unrecognized, and a log is recorded.

[0121] S40. Transformer-Line-Box Identification Based on Electricity Consumption Information Matching: Constraint matching of transformer, line and meter box nodes is performed based on electricity consumption information, and the topology is optimized by using the law of conservation of energy and Kirchhoff's current law.

[0122] Specifically, in this embodiment, the specific steps are as follows:

[0123] S401. Construction of Parent Node and Candidate Child Node Set: For each potential parent node (such as a transformer or line end point), candidates are selected based on electricity consumption information data to determine a set of possible child nodes (lower-level meter boxes or meters).

[0124] S402, Enumeration of Child Node Combinations: For each parent node, enumerate all possible combinations of its child nodes;

[0125] S403. Calculate line losses and leakage current: For a given combination, calculate the sum of the energy and current of the child nodes, as well as the energy and current of the parent node. Based on the law of conservation of energy and Kirchhoff's current law, calculate the line losses and leakage current.

[0126] Law of Conservation of Energy: Therefore, line loss ,in Represents the set of all child nodes;

[0127] in, Indicates the parent node's electrical energy value. This represents the electrical energy value of each child node;

[0128] Kirchhoff's Current Law: Therefore, leakage current ;

[0129] in, Indicates the current value of the parent node. This represents the current value of each child node;

[0130] S404, Validity Verification: Requires line loss verification. Leakage current ;

[0131] in, This indicates the maximum allowable line loss threshold. This represents the minimum allowable line loss threshold. This indicates the maximum permissible leakage current threshold. Indicates the minimum allowable leakage current threshold;

[0132] S405. Evaluation Index Calculation: Define the comprehensive evaluation index as follows:

[0133]

[0134] in, , These are weighting parameters used to control the importance of line loss and leakage current;

[0135] S406. Optimal Topology Determination: For each parent node, select the combination of child nodes that achieves the best comprehensive index and update the topology.

[0136] S50. Results Evaluation: The accuracy of the identification results is evaluated using the Adjusted Rand Index (ARI) and Normalized Mutual Information (NMI).

[0137] Example 1

[0138] A typical low-voltage distribution area was selected as the experimental subject. The area includes one distribution transformer, four meter boxes, and 201 meter nodes. The experiment utilizes the fusion processing of communication and metering data, and employs the method of this invention for topology identification and evaluation. The topology diagram of the experimental distribution area is shown below. Figure 3 As shown.

[0139] The district data contains 201 meter nodes and 4 meter box nodes, totaling 205 nodes. Partial topology data is shown in Table 1 below:

[0140] Table 1

[0141] Serial Number Location address type 1 Table box 3 000002304032 Entity table 2 Table box 3 000002304036 Entity table 3 Table box 3 000002304053 Entity table 4 Table box 3 000002304054 Entity table 5 Table box 3 000002304055 Entity table 6 Table box 3 000002304057 Entity table 7 Table box 3 000002304006 Entity table 8 Table box 3 000002304008 Entity table 9 Table box 3 000002304058 Entity table 10 Table box 3 000002304013 Entity table 11 Table box 3 000002304015 Entity table 12 Table box 3 000002304017 Entity table 13 Table box 3 000002304018 Entity table 14 Table box 3 000002304093 Entity table 15 Table box 4 202404020038 Entity table 16 Table box 4 202404020039 Entity table 17 Table box 4 202404020040 Entity table 18 Table box 4 000002304136 Entity table 19 Table box 4 000002304137 Entity table 20 Table box 4 000002304138 Entity table 21 Table box 4 000002304139 Entity table 22 Table box 4 000002304140 Entity table … .. … Entity table 165 Table box 2 000002304144 Entity table 166 Table box 2 000002304145 Entity table 167 Table box 2 000002304146 Entity table 168 Table box 2 102015070473 Entity table 169 Table box 2 102015070474 Entity table 170 Table box 2 102015070480 Entity table 171 Table box 2 102015070479 Entity table 172 Table box 2 102015070481 Entity table 173 Table box 2 102015070482 Entity table 174 Table box 2 102015070353 Entity table 175 Table box 2 102015070354 Entity table 176 Table box 2 102015070355 Entity table 177 Table box 2 102015070356 Entity table 178 Table box 2 102015070357 Entity table … … … Entity table 198 Table box 1 010000000732 LTU 199 Table box 2 010000000733 LTU 200 Table box 3 010000000734 LTU 201 Table box 4 010000000736 LTU 202 Table box 3 006800991668 Entity table 203 Table box 3 104240250708 Entity table 204 Table box 3 104240250709 Entity table 205 Table box 3 001109000821 Entity table

[0142] The input communication data is shown in Table 2 below:

[0143] Table 2

[0144] Collection time Node address Neighbor address distance HPLC-SNR HPLC-RSSI HRF-SNR HRF-RSSI Equipment type 2025 / 7 / 3 16:56 9 5.11E+11 923 19 4 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 922 10 -6 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 919 11 -11 0 0 Electricity meter 2025 / 7 / 3 16:56 9 13 922 9 2 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 920 10 -16 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 927 8 -6 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 868 24 12 20 -30 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 866 35 23 23 -26 Electricity meter 2025 / 7 / 3 16:56 9 1E+10 866 34 28 20 -16 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 919 8 -4 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 925 9 -7 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 918 6 -17 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 922 6 -17 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 919 9 -6 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 868 24 10 12 -46 Electricity meter 2025 / 7 / 3 16:56 9 1E+10 937 5 -4 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 919 8 -5 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 919 6 -11 0 0 Electricity meter 2025 / 7 / 3 16:56 9 20 919 10 4 0 0 Electricity meter 2025 / 7 / 3 16:56 9 2.02E+11 919 8 -18 0 0 Electricity meter 2025 / 7 / 3 16:56 10 5.11E+11 900 18 4 0 0 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 919 6 -21 0 0 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 919 7 -25 0 0 Electricity meter 2025 / 7 / 3 16:56 10 13 923 11 -17 0 0 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 922 5 -25 0 0 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 919 7 -10 0 0 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 866 21 4 21 -27 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 868 24 11 22 -38 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 868 20 11 19 -32 Electricity meter 2025 / 7 / 3 16:56 10 2.02E+11 865 34 16 22 -17 Electricity meter 2025 / 7 / 3 16:56 10 1E+10 866 20 29 25 -4 Electricity meter … … … … … … … … …

[0145] The node address is the unique identifier of the meter / LTU device; the neighbor address is the identifier of other nodes that have a communication connection with the current node; the distance represents the physical distance between two nodes; HPLC-SNR represents the signal-to-noise ratio of power line carrier communication; HPLC-RSSI represents the received signal strength of power line carrier communication; HRF-SNR represents the signal-to-noise ratio of wireless communication; HRF-RSSI represents the received signal strength of wireless communication.

[0146] Some of the input measurement data are shown in Table 3 below:

[0147] Table 3

[0148] Node address First collection time Initial power level freeze Current collection time Current frozen battery level Time (current) Current Node type Electricity increase 10000000732 2025 / 7 / 2 0:00 218.94 2025 / 7 / 3 0:00 220.98 2025 / 7 / 3 17:01 7.257 LTU 2.04 10000000733 2025 / 7 / 2 0:00 121.51 2025 / 7 / 3 0:00 122.75 2025 / 7 / 3 17:01 4.543 LTU 1.24 10000000734 2025 / 7 / 2 0:00 62.01 2025 / 7 / 3 0:00 63.11 2025 / 7 / 3 17:01 4.395 LTU 1.1 10000000736 2025 / 7 / 2 0:00 59.89 2025 / 7 / 3 0:00 60.57 2025 / 7 / 3 17:01 2.703 LTU 0.68 … … … … … … … … …

[0149] After data preprocessing and analysis, the data file contains 4 meter box nodes and 197 meter nodes. Following step S10, an adjacency matrix is ​​constructed, and unit / household identification is performed based on a clustering algorithm. The clustering process tree diagram is shown below. Figure 4 As shown. Simultaneously, a cluster confusion heatmap is plotted, as shown. Figure 5 As shown, the difference between the household identification results (including 197 meter nodes and 4 LTU nodes) and the topology file is verified to be completely consistent with the household identification results based on communication data.

[0150] Simultaneously, based on the adjusted RAND index and normalized mutual information evaluation of the mutual identification results, when analyzing only the 197 meter nodes and 4 LTU nodes present in the communication data table, the adjusted RAND index is 1 and the normalized mutual information index is 1, indicating that the topology identification result of this embodiment of the invention has a high accuracy. Since the communication data table only collected data from 201 nodes, and the data from the 4 meters was not collected, it is impossible to determine the affiliation of the 4 meters. Subsequently, the transformer-line-box connection relationship is identified based on the electricity consumption information matching algorithm, and its connection matrix is ​​as follows: Figure 6 As shown.

[0151] In the diagram, the horizontal axis represents child nodes, and the vertical axis represents parent nodes. As shown in the table, node 10000000732 is the parent node of nodes 10000000733 and 10000000736, and node 10000000733 is a child node of node 10000000734. This is verified to be consistent with the topology file. The topology identification results for this station area are as follows: Figure 7 As shown.

[0152] This invention also provides a low-voltage distribution area non-disruptive topology identification system, comprising:

[0153] The data preprocessing module is used to clean and filter the raw communication and measurement data;

[0154] The adjacency matrix construction module is used to convert the RSSI between devices into the distance between nodes and form an undirected weighted graph;

[0155] The meter box-household identification module is used to determine the ownership relationship between the meter box and the electricity meter based on a clustering algorithm;

[0156] The transformer-line-box identification module is used to constrain and match the nodes of transformers, lines and meter boxes according to power consumption information, and optimize the topology using the law of conservation of energy and Kirchhoff's current law.

[0157] The results evaluation module is used to evaluate the accuracy of the identification results based on the adjusted RAND index and normalized mutual information.

[0158] This invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the methods in this embodiment.

[0159] This invention also provides an electronic terminal, including: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal performs any of the methods in this embodiment.

[0160] As will be understood by those skilled in the art, the computer-readable storage medium described in this embodiment allows for the implementation of all or part of the steps in the above method embodiments by computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0161] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication between them. The memory is used to store computer programs, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer programs, so that the electronic terminal performs the steps of the above method.

[0162] Compared with existing technologies, the low-voltage distribution area non-disruptive topology identification method, system, medium, and terminal described in the above embodiments address the problem of inaccurate or missing topology data in many medium and low-voltage distribution areas, making it difficult to guarantee high accuracy in topology identification. This invention features a simple process and convenient operation, achieving deep coupling between RSSI / SNR communication characteristics and energy metering characteristics, constructing a multi-source data fusion framework for topology identification, and overcoming the limitations of single signal dependence. By replacing signal injection with algorithmic modeling, it achieves non-disruptive topology inference throughout the entire process, can operate in real-time under energized conditions, and is feasible for engineering deployment. A dynamic completion algorithm is proposed for communication sparse or weakly connected environments, effectively compensating for the identification defects of isolated nodes and significantly improving topology integrity. Using the energy conservation equation and Kirchhoff's current law as discrimination criteria, the results conform to statistical laws and electrical physics laws, ensuring the rationality and reliability of the identification results. In summary, without adding extra hardware, without interrupting power supply, and without injecting feature signals, this invention gradually constructs a complete topology structure solely based on existing smart meters, communication signals, and power data from electricity consumption information collection systems. This enables high-precision and automated identification of transformer area topologies, ensuring high accuracy in low-voltage transformer area topology identification.

[0163] Obviously, the embodiments described above are merely preferred embodiments of the present invention, and not all embodiments. The accompanying drawings illustrate preferred embodiments of the present invention, but do not limit the scope of the patent. The present invention can be implemented in many different forms; rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this invention.

Claims

1. A method for non-disruptive topology identification of low-voltage distribution areas, characterized in that, Includes the following steps: S10. Clean and filter the raw communication and measurement data; S20. Convert the RSSI between devices into the distance between nodes and form an undirected weighted graph; In step S20, each meter or meter box in the power network topology is considered as a node in a graph, and the communication between nodes is considered as an undirected edge. Let the graph... It is an undirected graph, where Represents a set of nodes, where Indicates the number of nodes. Represents the set of edges, adjacency matrix for Matrix, matrix elements Represents a node With nodes The connection weights between nodes are determined by the fact that if there is no effective communication between nodes, they are considered disconnected, and the corresponding weights are reset to infinity. Defined as: For nodes that can receive signals, their weights are represented as follows: in Represents a node With nodes Between signal strength, An environmental attenuation factor that controls the signal attenuation rate; S30. Use clustering algorithms to determine the ownership relationship between meter boxes and electricity meters; S40. Based on the electricity consumption information, perform constraint matching on the transformer, line and meter box nodes, and optimize the topology using the law of conservation of energy and Kirchhoff's current law. S50. Use the adjusted RAND index and normalized mutual information to evaluate the accuracy of the identification results.

2. The method for non-disruptive topology identification of low-voltage distribution areas according to claim 1, characterized in that, In step S10, the total number of times each node appears in the communication list is counted and regarded as the degree of the graph. A threshold k is set. For nodes with a degree less than k, they are marked as low-connectivity nodes and removed or processed separately from the subsequent adjacency matrix construction and clustering.

3. The method for non-disruptive topology identification of low-voltage distribution areas according to claim 1, characterized in that, The adjacency matrix construction process is as follows: S201. Initialize the adjacency matrix: Construct the adjacency matrix based on the denoised node list. Set the diagonal elements to 0 and the rest to 0. ; S202, Signal strength conversion: For each communication record ,like If the value is not 0, calculate the distance using the formula above. ; S203, Adjacency Matrix Filling: Fill the adjacency matrix according to the above definition; S204, Symmetry: Since the topological graph is undirected, therefore .

4. The method for non-disruptive topology identification of low-voltage distribution areas according to claim 3, characterized in that, In step S30, there is a meter box and a set of electricity meters. The corresponding similarity matrix The algorithm execution process is as follows: S301. Initialization: Treat each node as an independent cluster. Initial number of clusters ; S302. Distance Calculation: The distance between all clusters is calculated using the adjacency matrix. The distance between clusters is defined as the average distance between all pairs of nodes in two clusters, as shown in the following formula: in, and Indicates two different clusters. and Representing clusters and The number of samples in Point With point The distance between them is represented by the Euclidean distance; S303. Merge Clusters: Based on the calculated distance, select the two most similar clusters to merge and update the cluster set; S304. Termination condition: Continue merging until the number of clusters reaches a preset value. ; S305, Cluster Center Allocation: For each cluster... The node with the largest communication weight and the average weight of the meters in the cluster is selected as the candidate parent node of the meter box.

5. The method for non-disruptive topology identification of low-voltage distribution areas according to claim 4, characterized in that, A post-processing mechanism is introduced to address low-connectivity nodes whose cluster affiliation is not explicitly defined. Based on its connection relationships with its already assigned nodes, analyze the table cluster to which it most likely belongs. The specific steps are as follows: S311, Node sorting: Count the number of neighbors of each low-connectivity node that have been assigned to a cluster, sort them in descending order, and prioritize the processing of nodes with richer information; S312, Find Connected Neighbors: For each low-connectivity node Find all its neighboring nodes in the communication data. Sort the nodes in descending order based on whether the neighboring nodes belong to a cluster. S313. Extract Neighbor Attribution: Obtain the cluster label of the node whose attribution has been determined among the above neighbor nodes. The frequency of each cluster label was counted and recorded as the number of matches. ; S314. Calculate the affiliation score: For each possible cluster... Combined with connection weights and number of matches Calculate the attribution score As shown in the following formula: in, Represents a node With already belonged to neighboring nodes The connection weights, , Indicates the weighting coefficient; S315. Determine the final cluster: Find the cluster with the highest score. As shown in the following formula: and low connection nodes Belong to this cluster; S316. Exception Handling: If none of the neighbors belong to a cluster, then retain... The status is unrecognized, and a log is recorded.

6. The method for non-disruptive topology identification of low-voltage distribution areas according to claim 5, characterized in that, The specific steps of S40 are as follows: S401. For each potential parent node, candidate selection is performed based on electricity consumption information data to determine a set of possible child nodes; S402. For each parent node, enumerate all possible combinations of its child nodes; S403. Calculate the line loss and leakage current according to the law of conservation of energy and Kirchhoff's current law: Law of Conservation of Energy: Therefore, line loss ,in Represents the set of all child nodes; in, Indicates the parent node's electrical energy value. This represents the electrical energy value of each child node; Kirchhoff's Current Law: Therefore, leakage current ; in, Indicates the current value of the parent node. This represents the current value of each child node; S404, Requirements for line loss Leakage current ; in, This indicates the maximum allowable line loss threshold. This represents the minimum allowable line loss threshold. This indicates the maximum permissible leakage current threshold. Indicates the minimum allowable leakage current threshold; S405. Define the comprehensive evaluation indicators as follows: in, , These are weighting parameters used to control the importance of line loss and leakage current; S406. For each parent node, select the combination of child nodes that results in the best overall performance and update the topology.

7. A low-voltage distribution area non-disruptive topology identification system, characterized in that, include: The data preprocessing module is used to clean and filter the raw communication and measurement data; The adjacency matrix construction module is used to convert the RSSI between devices into the distance between nodes and form an undirected weighted graph; The meter box-household identification module is used to determine the ownership relationship between the meter box and the electricity meter based on a clustering algorithm; The transformer-line-box identification module is used to constrain and match the nodes of transformers, lines and meter boxes according to power consumption information, and optimize the topology using the law of conservation of energy and Kirchhoff's current law. The results evaluation module is used to evaluate the accuracy of the identification results based on the adjusted RAND index and normalized mutual information.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

9. An electronic terminal, characterized in that, include: Processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the terminal to perform the method as described in any one of claims 1 to 6.

Images