A method for dynamic identification of low voltage distribution area topology based on characteristic current

Abstract

The present invention discloses a method for dynamically identifying the topology of a low-voltage distribution station area based on characteristic current, which comprises the following steps: determining all nodes of a low-voltage distribution station area, and forming all nodes into an area node set; randomly selecting two nodes from the area node set in a permutation and combination manner, and injecting characteristic current into the randomly selected nodes in a synchronous and then asynchronous order; collecting identification signals of all nodes when injecting characteristic current synchronously and asynchronously, screening identification signals corresponding to the attributes of characteristic current in all nodes of the low-voltage distribution station area, and determining the topological hierarchical relationship between the screened nodes; eliminating branch nodes or user loads with unique identities from the area node set to generate a reduced dimension set, and always retaining the root node in the reduced dimension set; repeatedly performing node identification on the reduced dimension set until all nodes are traversed to generate a topological relationship diagram; the present invention reduces the number of permutations and combinations, and reduces the difficulty of identifying the node hierarchical relationship.

Description

Technical Field

[0001] The invention relates to the technical field of low-voltage distribution station area maintenance, and in particular to a method for dynamically identifying the topology of a low-voltage distribution station area based on characteristic current. Background Art

[0002] There is a problem of unclear connection relationship between household transformers in distribution rooms and low-voltage substations, especially in newly built substations, which makes it impossible to directly obtain some topologies. It is necessary to take manual surveys, which is a huge workload and wastes manpower and material resources. At present, most current injection methods are used to obtain topological relationships. This method adds a characteristic current (small current) injection circuit to the carrier module of the user's meter and each branch circuit breaker (or low-voltage branch monitoring unit). At the same time, the branch circuit breaker (or low-voltage branch monitoring unit) is equipped with a core board. By controlling the carrier module of the user's meter and the branch circuit breaker (or low-voltage branch monitoring unit) in turn to inject characteristic current, the core board detects whether the characteristic current flows through the circuit breaker (or low-voltage branch monitoring unit) to determine the hierarchical relationship between the user's meter and each branch.

[0003] At present, in order to realize the low-voltage distribution network with known topology structure and timely monitor the changes of the power supply network topology structure, most of the current injection methods are used to online identify and monitor the topological relationship of the low-voltage distribution network, such as the invention patent with application number 201911263075.8, a method and system for identifying the topology of a low-voltage distribution network station area, and the invention patent with application number 201911169845.2, an online identification and monitoring method for the topological relationship of a low-voltage distribution network, respectively, through "each branch node sends a branch node identification signal to each subordinate node layer by layer, and the subordinate node of the branch node receives and analyzes the identification signal", and "sends activation instructions to the identification current generators at the ends of other users one by one in time-sharing" to obtain the identification signal. The first patent determines the topological network according to the known node hierarchy, and the second patent traverses all nodes in turn to determine the topological network.

[0004] For newly built substations and uncertain node hierarchy divisions, the existing technology for determining the topological network based on known node hierarchies is not applicable. The method of determining the topological network by traversing all nodes in sequence is highly complex and cannot quickly determine the node distribution of the topological network. Summary of the invention

[0005] The purpose of the present invention is to provide a method for dynamically identifying the topology of a low-voltage distribution station area based on characteristic current, so as to solve the technical problems in the prior art.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] A method for dynamically identifying low-voltage distribution station area topology based on characteristic current includes the following steps:

[0008] Step 100, determining all nodes in the low-voltage distribution station area, the nodes including the root node, each branch node and all user loads of the low-voltage distribution station area, and forming all nodes of the low-voltage distribution station area into an area node set;

[0009] Step 200: randomly select two nodes from the area node set in a permutation and combination manner and mark them as selection nodes, and mark all nodes in the area node set as screening nodes, and then inject characteristic current into the two selection nodes in a synchronous and then asynchronous order, so that the current signals of all nodes located in the same branch have high signal similarity, and the current signals of all nodes located in different branches have low signal similarity;

[0010] Step 300: When injecting characteristic current synchronously and asynchronously, identification signals of all screening nodes are collected, and the substation terminal performs similarity analysis on the identification signals to determine the topological hierarchical relationship between the screening nodes in the substation node set and the node identities of the selection nodes and the screening nodes;

[0011] Step 400: based on the topological hierarchical relationship between the screening nodes and the node identities of the selection nodes and the screening nodes, the branch nodes or user loads with determined unique node identities are removed from the station area node set to generate a reduced dimension set, and the root node is always retained in the reduced dimension set;

[0012] Step 500, replace the substation node set with the reduced dimensionality set, repeat steps 200 to 400 for the reduced dimensionality set, until all nodes in the reduced dimensionality set are traversed, and the topological hierarchical relationship between the screening nodes in the reduced dimensionality set and the node identities of the selection nodes and the screening nodes are determined in turn, and the substation terminal connects all nodes according to the node identities and topological hierarchical relationships of all nodes in turn to generate a topological structure diagram of the low-voltage distribution station area.

[0013] As a preferred solution of the present invention, in step 200, the attribute parameters of the characteristic current injected at the two selected nodes remain consistent, and the attribute parameters of the characteristic current include switching frequency, duty cycle and amplitude.

[0014] As a preferred solution of the present invention, in step 300, the identification signals of all screening nodes are collected when the characteristic current is injected synchronously and asynchronously respectively, and the substation terminal performs similarity analysis on the identification signals, including: injecting the characteristic current synchronously into two selection nodes, the substation terminal collects the identification signals of all screening nodes, and calculates the signal similarity of the identification signal and the characteristic current of each screening node in turn based on the attribute parameter, and the calculation formula of the signal similarity is:

[0015]

[0016] Where P i It is characterized by the signal similarity between the identification signal and the characteristic current of the i-th screening node, X O The component X, X characterized as the characteristic current attribute parameter i The components X, x1, x2, and x3 are respectively represented as the three components of the attribute parameters of the identification signal of the i-th screening node: switching frequency, duty cycle, and amplitude. X The component weight of component X represented as the attribute parameter, i is a measurement constant;

[0017] If the signal similarity of the screening node exceeds a preset similarity threshold, the screening node is identified as a user load, and a characteristic current is synchronously injected into the two selection nodes;

[0018] If the signal similarity of the screening nodes does not exceed a preset similarity threshold and the number of the screening nodes exceeds 2, two selection nodes are identified as root nodes and / or branch nodes, and the screening nodes are subjected to a self-loop identification operation;

[0019] If the signal similarity of the screening nodes does not exceed the preset similarity threshold and the number of the screening nodes does not exceed 2, the two selection nodes are identified as user loads, and the process goes to step 400 for execution;

[0020] Preferably, the determination of the component weight of the component X of the attribute parameter includes:

[0021] The principal component analysis is performed on the identification signal of the screening node to obtain the eigenvalues ​​of the three components x1, x2, and x3 in the component X of the attribute parameter, and the eigenvalues ​​of x1, x2, and x3 are normalized to obtain the component weights of x1, x2, and x3. The calculation formula of the component weights is:

[0022]

[0023] Where, X PCA Characterizes the eigenvalue of component X of the attribute parameter.

[0024] As a preferred solution of the present invention, the self-loop identification operation is implemented as follows:

[0025] Synchronously injecting characteristic current into the screening node, and the station terminal acquires and analyzes the identification signal of the screening node;

[0026] The signal similarities of the identification signals of the screening nodes are arranged in descending order to respectively determine the topological levels of the screening nodes, wherein the node with the largest signal similarity of the identification signals of the screening nodes is the user load, and when the signals of the identification signals of the screening nodes decrease successively, the node levels of the screening nodes are arranged in descending order.

[0027] As a preferred solution of the present invention, in step 300, characteristic currents are asynchronously injected into two selection nodes to determine the number of nodes where characteristic currents appear synchronously to determine whether the two selection nodes include root nodes and / or branch nodes, and the topological hierarchical relationship of the screening nodes in the two selection nodes is determined based on the number of times characteristic currents appear synchronously in the identification signals of all nodes.

[0028] As a preferred solution of the present invention, the analysis steps of the station terminal when selecting a node to inject characteristic current in an asynchronous order include:

[0029] Inject the characteristic current into any one of the two selection nodes, and the station terminal collects the identification signals at all the screening nodes, and calculates the signal similarity between each screening node and the characteristic current;

[0030] If the number of screening nodes whose signal similarity does not exceed the similarity threshold does not exceed 1, the corresponding selection node is the user load;

[0031] If the number of screening nodes whose signal similarity does not exceed the similarity threshold exceeds 1, the corresponding selection node is a root node or a branch node.

[0032] As a preferred solution of the present invention, in step 300, the implementation method of determining whether the selected node is a root node or a branch node includes:

[0033] Counting the specific nodes whose number of screening nodes with signal similarity higher than the similarity threshold exceeds 1, and when the number of specific nodes is the total number of nodes in the station area node set, the corresponding node is selected as the root node;

[0034] A specific node is selected when the number of the selected nodes whose statistical signal similarity is higher than the similarity threshold exceeds 1, and when the number of the specific nodes is less than the total number of nodes in the station area node set, the node is selected as a branch node.

[0035] As a preferred solution of the present invention, in step 300, when the characteristic current is asynchronously injected into the selected node, the number of occurrences of the screening node of the identification signal whose signal similarity does not exceed the similarity threshold in the two operations of asynchronously injecting the characteristic current is counted to determine the user load in the substation node set, and when the number of occurrences of the screening node of the identification signal whose signal similarity does not exceed the similarity threshold is twice, the node is determined to be a user load.

[0036] As a preferred solution of the present invention, in the first two selection nodes screening all nodes in the low-voltage distribution station area, among the identification signals whose signal similarity does not exceed the similarity threshold, the levels of multiple branch nodes cannot be accurately and uniquely determined in the first round, and the branch nodes to which the user load belongs are marked according to the affiliation of the screening nodes, and the identified branch nodes and root nodes are marked in the dimensionality reduction set.

[0037] As a preferred solution of the present invention, in step 500, two nodes are repeatedly selected from the dimensionality reduction set to eliminate all user loads and branch nodes, and when the characteristic signal is injected into the branch node, and there is no identification signal with the same partial attributes as the characteristic current in all nodes of the low-voltage distribution station area, the topology identification is completed, the root node that is always retained is incorporated into the topological hierarchy relationship, and all nodes are spliced ​​according to all topological hierarchy relationships to generate a topological structure diagram of the low-voltage distribution station area.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention is applied to newly built low-voltage distribution station areas and working scenarios where the node hierarchy division is unclear. Two nodes are randomly selected by permutation and combination, and characteristic current is injected into the nodes at the same time. At this time, the relationship and identity between the two randomly selected nodes are determined according to the identification signal of the currently screened node that changes synchronously with the current, that is, whether the characteristic current amplitude is superimposed. The two nodes are randomly selected as a hierarchical relationship or a parallel relationship, or the two nodes are randomly selected as user loads. The preliminary identities of the two randomly selected nodes can be quickly pre-determined, and then the topological hierarchy of the two randomly selected nodes and the screening hierarchy of the screened nodes that change synchronously with the current are determined by asynchronously injecting characteristic current and self-circulating identification of the screened nodes that change synchronously with the current. The nodes of the determined hierarchy are eliminated, the number of permutations and combinations is reduced, and the difficulty of identifying the node hierarchy relationship is reduced, thereby achieving a fast topological dynamic identification effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0041] Figure 1 A schematic diagram of a flow chart of a method for dynamically identifying distribution station area topology provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] like Figure 1 As shown, the present invention provides a method for dynamically identifying the topology of a low-voltage distribution station based on characteristic current. This embodiment is applied to newly built low-voltage distribution stations and working scenarios where the node hierarchy division is unclear. At this time, the node level of the low-voltage distribution station cannot be determined (i.e., the tree diagram layering). Therefore, the topological association diagram cannot be determined by injecting characteristic current into each node in turn according to the hierarchy. At the same time, the method of determining the topological relationship by injecting characteristic current into each node in turn cannot quickly determine the node level and topological relationship.

[0044] In order to solve the above problems, the present embodiment randomly selects two nodes by permutations and combinations, and simultaneously injects characteristic current into the nodes. At this time, the relationship and identity between the two randomly selected nodes are determined according to the identification signal of the currently screened node that changes synchronously with the current, that is, whether the characteristic current amplitude is superimposed. The two nodes are randomly selected as a hierarchical relationship or a parallel relationship, or the two nodes are randomly selected as user loads. The preliminary identities of the two randomly selected nodes can be quickly pre-determined, and then the topological hierarchy of the two randomly selected nodes and the screening hierarchy of the screened nodes that change synchronously with the current are determined by asynchronously injecting characteristic current and self-looping identification of the screened nodes that change synchronously with the current. The nodes of the determined hierarchy are eliminated, the number of permutations and combinations is reduced, and the difficulty of identifying the node hierarchy relationship is reduced, thereby achieving a fast topological dynamic identification effect.

[0045] The method for dynamically identifying the topology of a low-voltage distribution station area includes the following steps:

[0046] Step 100: determine all nodes of the low-voltage distribution station area, including the root node, each branch node and all user loads of the low-voltage distribution station area, and form all nodes of the low-voltage distribution station area into an area node set.

[0047] According to the actual application of the low-voltage distribution substation, there is only one root node in the low-voltage distribution substation, each branch node is greater than or equal to one, and the user load of each branch node is greater than one.

[0048] Step 200, randomly select two nodes from the substation node set in the permutation and combination manner and mark them as selection nodes, and mark all nodes in the substation node set as screening nodes, and then inject characteristic current into the two selection nodes in a synchronous and then asynchronous order, so that the current signals of all nodes located in the same branch have high signal similarity, and the current signals of all nodes located in different branches have low signal similarity, and the node hierarchical relationship or identity can be distinguished by signal similarity.

[0049] The attribute parameters of the characteristic current injected at the two selected nodes are kept consistent, and the attribute parameters of the characteristic current include switching frequency, duty cycle and amplitude. In this way, for the topological relationship tree diagram, the current signal of the node at the lower layer is affected by the current signals of multiple nodes at the upper layer. When two branch nodes on the same branch line inject characteristic current at the same time, the amplitude of the current signal on the user load on the branch line is the superposition value of the amplitude of the circuit signal on the two branch nodes after circuit attenuation. Therefore, through the above method, this embodiment can quickly determine whether the two nodes have a hierarchical relationship of the same branch line and the user load under the same node in the node. The nodes with a hierarchical relationship can then inject characteristic currents in an asynchronous sequence to preliminarily determine the stratification of two nodes with a hierarchical relationship of the same branch line, thereby achieving the purpose of quickly realizing the recognition of the topological structure diagram.

[0050] Step 300, collect identification signals of all screening nodes when injecting characteristic current synchronously and asynchronously respectively, and the substation terminal performs similarity analysis on the identification signals to determine the topological hierarchical relationship between the screening nodes in the substation node set and the node identities of the selection nodes and screening nodes. After analyzing the identification signals, the substation terminal knows that the purpose of synchronously injecting characteristic currents into two randomly selected nodes is to identify the user load. The substation terminal counts the identification signals of all nodes in the substation node set, and identifies the user load based on the matching of the identification signals of all nodes with the attribute of the characteristic current. The synchronous and asynchronous injection of characteristic currents to determine the topological hierarchical relationship between the screening nodes is divided into the following three aspects:

[0051] In the first aspect, the identification signals of all screening nodes are collected when the characteristic current is injected synchronously and asynchronously, and the station terminal performs similarity analysis on the identification signals, including:

[0052] The characteristic current is injected into the two selection nodes synchronously, and the station terminal collects the identification signals of all the screening nodes, and calculates the signal similarity between the identification signal of each screening node and the characteristic current in turn based on the attribute parameter. The calculation formula of the signal similarity is:

[0053]

[0054] Where P i It is characterized by the signal similarity between the identification signal and the characteristic current of the i-th screening node, X O The component X, X characterized as the characteristic current attribute parameter i The components X, x1, x2, and x3 are respectively represented as the three components of the attribute parameters of the identification signal of the i-th screening node: switching frequency, duty cycle, and amplitude. X The component weight of component X represented as the attribute parameter, i is a measurement constant;

[0055] If the signal similarity of the screening node exceeds the preset similarity threshold, the screening node is identified as a user load, and the characteristic current is synchronously injected into the two selected nodes. In this case, the screening node is a user load, and there is a hierarchical relationship between the two randomly selected nodes. The two randomly selected nodes are subjected to the next step of asynchronous injection of characteristic current. The purpose of asynchronous injection of characteristic current is to determine the specific hierarchical division of the two randomly selected nodes.

[0056] If the signal similarity of the screening nodes does not exceed a preset similarity threshold and the number of the screening nodes exceeds 2, two selection nodes are identified as root nodes and / or branch nodes, and the screening nodes are subjected to a self-loop identification operation;

[0057] The self-loop recognition operation is implemented as follows:

[0058] Synchronously injecting characteristic current into the screening node, and the station terminal acquires and analyzes the identification signal of the screening node;

[0059] The signal similarities of the identification signals of the screening nodes are arranged in descending order to respectively determine the topological levels of the screening nodes, wherein the node with the largest signal similarity of the identification signals of the screening nodes is the user load, and when the signals of the identification signals of the screening nodes decrease successively, the node levels of the screening nodes are arranged in descending order.

[0060] If the signal similarity of the screening nodes does not exceed the preset similarity threshold and the number of the screening nodes does not exceed 2, the two selection nodes are identified as user loads, and the process goes to step 400 for execution;

[0061] In reality, when the characteristic current is injected synchronously, there are different situations for the identification signal of the screening node. The first identification signal is 0, that is, the identification signal is identified, and the identity and level of this screening node cannot be determined. The second identification signal is not 0, which is the characteristic current generated after circuit attenuation or circuit attenuation and superposition. Therefore, when the signal similarity is closest to 1, it means that the characteristic current is obtained only after circuit amplitude attenuation. When the number of screening nodes exceeds 2, the identity of the selection node can be identified. When the signal similarity is closest to 2, it means that the characteristic current is obtained only after circuit amplitude attenuation and superposition. When the number of screening nodes does not exceed 2, the identity of the selection node can be identified.

[0062] When the characteristic current is injected synchronously, the identification signal is theoretically consistent in switching frequency and duty cycle. When setting the similarity threshold, only the amplitude can be considered, that is, theoretically:

[0063] If the switching frequency and duty cycle of the screening node are the same as the characteristic current, and the amplitude of the identification signal is nearly twice the characteristic current, the screening node is identified as a user load, and the characteristic current is injected into the two selection nodes synchronously;

[0064] If the switching frequency and duty cycle of the screening node are the same as the characteristic current, the amplitude of the identification signal is nearly twice the characteristic current, and the number of the screening nodes exceeds 2, the two selection nodes are identified as root nodes and / or branch nodes, and the screening nodes are subjected to a self-loop identification operation;

[0065] If the switching frequency and duty cycle of the screening node are the same as the characteristic current, the identification signal amplitude is close to twice the characteristic current, and the number of the screening nodes does not exceed 2, the two selection nodes are identified as user loads, and the similarity threshold is set accordingly. It can also be set by the user.

[0066] Although the identification signal is theoretically consistent in switching frequency and duty cycle when the characteristic current is injected synchronously, attenuation and offset are always inevitable in the actual current. Therefore, the characteristic information is used to set the weights of the components of the attribute parameters to achieve accurate calculation of signal similarity.

[0067] Preferably, the determination of the component weight of the component X of the attribute parameter includes:

[0068] The principal component analysis is performed on the identification signal of the screening node to obtain the eigenvalues ​​of the three components x1, x2, and x3 in the component X of the attribute parameter, and the eigenvalues ​​of x1, x2, and x3 are normalized to obtain the component weights of x1, x2, and x3. The calculation formula of the component weights is:

[0069]

[0070] Where, X PCA The eigenvalue of the component X represented as the attribute parameter, X PCA It is the eigenvalue of component X obtained by the PCA algorithm (Principal Component Analysis algorithm).

[0071] Since the randomly selected node is a combination of two random elements between {root node, branch node, user load}, when the characteristic signal is injected asynchronously and the node with an identification signal similar to the switching frequency, duty cycle and amplitude of the characteristic current appears twice in the screening node, the node is determined to be a user load. Therefore, the asynchronous injection characteristic signal can be used to identify the user load twice.

[0072] When the signal similarity of the screening nodes does not exceed the preset similarity threshold and the number of the screening nodes does not exceed 2, two randomly selected nodes are used as root nodes and / or branch nodes, and the screening nodes are subjected to a self-loop identification operation. The hierarchy of the screening nodes is determined through the self-loop identification operation, so that only two randomly selected nodes are used to implement a consistent hierarchy confirmation operation, thereby reducing the number of topology identification operations, reducing the complexity, and distinguishing topology hierarchies is relatively quick and simple.

[0073] Since it is not clear whether the node whose current signal changes synchronously with a node after the characteristic current is injected into it is a branch node or a user load, a self-loop identification operation is required to perform a secondary node identification operation. The implementation method of the self-loop identification operation is as follows:

[0074] The characteristic current is synchronously injected into the screening node, and the substation terminal obtains and analyzes the identification signal of the screening node.

[0075] The amplitudes of the identification signals of the screening nodes are arranged in descending order to determine the topological levels of the screening nodes, wherein the node with the largest amplitude of the identification signal of the screening node is the user load, and when the amplitudes of the identification signals of the screening nodes decrease successively, the node levels of the screening nodes are arranged from bottom to top.

[0076] When the signal similarity of the screening nodes does not exceed the preset similarity threshold and the number of the screening nodes exceeds 2, it can only be explained that the two randomly selected nodes are parallel nodes, that is, they are respectively on different branch lines, but the hierarchical division between the randomly selected nodes and the screening nodes is not clear. Therefore, this embodiment can determine the hierarchical division between the randomly selected nodes and the screening nodes through self-loop identification operation, and accurately and quickly identify the user load. The two randomly selected nodes are synchronously injected with characteristic currents to identify the user load, but the hierarchical relationship of the two randomly selected nodes cannot be determined. Therefore, on the second aspect, this embodiment asynchronously injects characteristic currents into the two randomly selected nodes to determine the number of nodes where the characteristic current appears synchronously to determine whether the randomly selected nodes contain root nodes and / or branch nodes, and the topological hierarchical relationship of the screening nodes in the randomly selected nodes is determined according to the number of characteristic currents that appear synchronously in the identification signals of all nodes.

[0077] The characteristic current is injected into the two selection nodes asynchronously to determine the number of nodes where the characteristic current appears synchronously to determine whether the two selection nodes contain root nodes and / or branch nodes, and the topological hierarchical relationship of the screening nodes in the two selection nodes is determined based on the number of times the characteristic current appears synchronously in the identification signals of all nodes.

[0078] The analysis steps of the station area terminal when selecting a node to inject characteristic current in an asynchronous order include:

[0079] Inject the characteristic current into any one of the two selection nodes, and the station terminal collects the identification signals at all the screening nodes, and calculates the signal similarity between each screening node and the characteristic current;

[0080] If the number of screening nodes whose signal similarity does not exceed the similarity threshold does not exceed 1, the corresponding selection node is the user load;

[0081] If the number of screening nodes whose signal similarity does not exceed the similarity threshold exceeds 1, the corresponding selection node is a root node or a branch node.

[0082] The implementation method for determining that the selected node is a root node or a branch node includes: counting specific nodes whose number of screening nodes whose signal similarity is higher than the similarity threshold is greater than 1, and when the number of specific nodes is the total number of nodes in the station area node set, the corresponding selected node is the root node;

[0083] A specific node is selected when the number of the selected nodes whose statistical signal similarity is higher than the similarity threshold exceeds 1, and when the number of the specific nodes is less than the total number of nodes in the station area node set, the node is selected as a branch node.

[0084] When the characteristic current is asynchronously injected into the selected node, the number of times the screening node of the identification signal whose signal similarity does not exceed the similarity threshold in the two operations of asynchronously injecting the characteristic current is counted to determine the user load in the station node set, and when the number of times the screening node of the identification signal whose signal similarity does not exceed the similarity threshold appears is twice, the node is determined to be a user load.

[0085] Step 400, based on the topological hierarchical relationship between the screening nodes and the node identities of the selection nodes and the screening nodes, the branch nodes or user loads with determined unique node identities are removed from the station node set to generate a reduced dimension set, and the root node is always retained in the reduced dimension set.

[0086] Since there are a large number of user loads in a low-voltage distribution station area, this embodiment uses the above-mentioned steps 200 and 300 to confirm and clear branch nodes or user loads whose identity information has been determined, thereby reducing the number of elements in the station area node set and generating a reduced-dimensionality set. This can greatly reduce the number of combinations of subsequent topological level confirmation operations and the implementation steps of determining the level, reduce a lot of computing time, and improve the accuracy and efficiency of topological identification operations.

[0087] In step 400, in the first two selection nodes screening all nodes in the low-voltage distribution station area, among the identification signals whose signal similarity does not exceed the similarity threshold, the levels of multiple branch nodes cannot be accurately and uniquely determined in the first round, and the branch nodes to which the user load belongs are marked according to the affiliation of the screening nodes, and the identified branch nodes and root nodes are marked in the dimensionality reduction set.

[0088] Step 500, replace the substation node set with the reduced dimensionality set, repeat steps 200 to 400 for the reduced dimensionality set, until all nodes in the reduced dimensionality set are traversed, and the topological hierarchical relationship between the screening nodes in the reduced dimensionality set and the node identities of the selection nodes and the screening nodes are determined in turn, and the substation terminal connects all nodes according to the node identities and topological hierarchical relationships of all nodes in turn to generate a topological structure diagram of the low-voltage distribution station area.

[0089] In step 500, two nodes are repeatedly selected from the dimensionality reduction set to eliminate all user loads and branch nodes, and when the characteristic signal is injected into the branch node, and there is no identification signal with the same partial attributes as the characteristic current in all nodes of the low-voltage distribution station area, the topology identification is completed, the root node that is always retained is incorporated into the topological hierarchy relationship, and all nodes are spliced ​​according to all topological hierarchy relationships to generate a topological structure diagram of the low-voltage distribution station area.

[0090] This implementation method randomly extracts two nodes by permutations and combinations, and simultaneously injects characteristic currents into the nodes. At this time, the relationship and identity between the two randomly extracted nodes are determined based on the identification signals of the currently screened nodes that change synchronously with the current, that is, whether the characteristic current amplitudes are superimposed. The two randomly extracted nodes are hierarchical or parallel, or the two randomly extracted nodes are user loads. The preliminary identities of the two randomly extracted nodes can be quickly pre-determined, and then the topological levels of the two randomly extracted nodes and the screening levels of the screened nodes that change synchronously with the current are determined by asynchronously injecting characteristic currents and self-looping identification of the screened nodes that change synchronously with the current. The nodes of the determined levels are eliminated to reduce the number of permutations and combinations, and the above operations are repeated to achieve a fast topological dynamic identification effect.

[0091] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for dynamically identifying low-voltage distribution area topology based on characteristic current, characterized in that: The following steps are involved: Step 100, determining all nodes in the low-voltage distribution station area, the nodes including the root node, each branch node and all user loads of the low-voltage distribution station area, and forming all nodes of the low-voltage distribution station area into an area node set; Step 200: randomly select two nodes from the area node set in a permutation and combination manner and mark them as selection nodes, and mark all nodes in the area node set as screening nodes, and then inject characteristic current into the two selection nodes in a synchronous and then asynchronous order, so that the current signals of all nodes located in the same branch have high signal similarity, and the current signals of all nodes located in different branches have low signal similarity; In step 200, the attribute parameters of the characteristic current injected at the two selected nodes remain consistent, and the attribute parameters of the characteristic current include switching frequency, duty cycle and amplitude; Step 300: When injecting characteristic current synchronously and asynchronously, identification signals of all screening nodes are collected, and the substation terminal performs similarity analysis on the identification signals to determine the topological hierarchical relationship between the screening nodes in the substation node set and the node identities of the selection nodes and the screening nodes; In step 300, the identification signals of all screening nodes are collected when the characteristic current is injected synchronously and asynchronously respectively, and the station terminal performs similarity analysis on the identification signals, including: The characteristic current is injected into the two selection nodes synchronously, and the station terminal collects the identification signals of all the screening nodes, and calculates the signal similarity between the identification signal of each screening node and the characteristic current in turn based on the attribute parameter. The calculation formula of the signal similarity is: Where Pi represents the signal similarity between the identification signal of the i-th screening node and the characteristic current, X O The component X represented by the characteristic current attribute parameter, Xi represents the component X of the identification signal attribute parameter of the i-th screening node, x1, x2, x3 represent the three components of the attribute parameter: switching frequency, duty cycle and amplitude, A X The component weight of component X represented as the attribute parameter, i is a measurement constant; If the signal similarity of the screening node exceeds a preset similarity threshold, the screening node is identified as a user load, and a characteristic current is synchronously injected into the two selection nodes; If the signal similarity of the screening nodes does not exceed a preset similarity threshold and the number of the screening nodes exceeds 2, two selection nodes are identified as root nodes and / or branch nodes, and the screening nodes are subjected to a self-loop identification operation; If the signal similarity of the screening nodes does not exceed the preset similarity threshold and the number of the screening nodes does not exceed 2, the two selection nodes are identified as user loads, and the process goes to step 400 for execution; The determination of the component weight of the component X of the attribute parameter includes: The principal component analysis is performed on the identification signal of the screening node to obtain the eigenvalues ​​of the three components x1, x2, and x3 in the component X of the attribute parameter, and the eigenvalues ​​of x1, x2, and x3 are normalized to obtain the component weights of x1, x2, and x3. The calculation formula of the component weights is: Where, X PCA Characterized as the eigenvalue of component X of the attribute parameter; The self-loop recognition operation is implemented as follows: Synchronously injecting characteristic current into the screening node, and the station terminal acquires and analyzes the identification signal of the screening node; The signal similarities of the identification signals of the screening nodes are arranged in descending order to determine the topological levels of the screening nodes respectively, wherein the node with the largest signal similarity of the identification signals of the screening nodes is the user load, and when the signal similarity of the identification signals of the screening nodes decreases in sequence, the node levels of the screening nodes are arranged in descending order; In step 300, the characteristic current is asynchronously injected into the two selection nodes to determine the number of nodes where the characteristic current appears synchronously to determine whether the two selection nodes include a root node and / or a branch node, and the topological hierarchical relationship of the screening nodes in the two selection nodes is determined according to the number of characteristic currents that appear synchronously in the identification signals of all nodes; The analysis steps of the station area terminal when selecting a node to inject characteristic current in an asynchronous order include: Inject the characteristic current into any one of the two selection nodes, and the station terminal collects the identification signals at all the screening nodes, and calculates the signal similarity between each screening node and the characteristic current; If the number of screening nodes whose signal similarity does not exceed the similarity threshold does not exceed 1, the corresponding selection node is the user load; If the number of screening nodes whose signal similarity does not exceed the similarity threshold exceeds 1, the corresponding selection node is a root node or a branch node; In step 300, the implementation method of determining whether the selected node is a root node or a branch node includes: Counting the specific nodes whose number of screening nodes with signal similarity higher than the similarity threshold exceeds 1, and when the number of specific nodes is the total number of nodes in the station area node set, the corresponding node is selected as the root node; Counting the specific nodes whose number of selected nodes with signal similarity higher than the similarity threshold exceeds 1, and when the number of specific nodes is less than the total number of nodes in the station area node set, selecting the node as a branch node; In the step 300, when the characteristic current is asynchronously injected into the selected node, the number of occurrences of the screening node of the identification signal whose signal similarity does not exceed the similarity threshold in two operations of asynchronously injecting the characteristic current is counted to determine the user load in the station node set, and when the number of occurrences of the screening node of the identification signal whose signal similarity does not exceed the similarity threshold is two, the node is determined to be a user load; Step 400: based on the topological hierarchical relationship between the screening nodes and the node identities of the selection nodes and the screening nodes, the branch nodes or user loads with determined unique node identities are removed from the station area node set to generate a reduced dimension set, and the root node is always retained in the reduced dimension set; Step 500, replace the substation node set with the reduced dimensionality set, repeat steps 200 to 400 for the reduced dimensionality set, until all nodes in the reduced dimensionality set are traversed, and the topological hierarchical relationship between the screening nodes in the reduced dimensionality set and the node identities of the selection nodes and the screening nodes are determined in turn, and the substation terminal connects all nodes according to the node identities and topological hierarchical relationships of all nodes in turn to generate a topological structure diagram of the low-voltage distribution station area.

2. According to claim 1, a method for dynamically identifying low-voltage distribution station area topology based on characteristic current is characterized in that: In the first round, when the two selection nodes screen all the nodes in the low-voltage distribution station area and the identification signals whose signal similarity does not exceed the similarity threshold, the levels of multiple branch nodes cannot be accurately and uniquely determined in the first round, and the branch nodes to which the user load belongs are marked according to the affiliation of the screening nodes, and the identified branch nodes and root nodes are marked in the dimensionality reduction set.

3. According to claim 2, a method for dynamically identifying low-voltage distribution station area topology based on characteristic current is characterized in that: In step 500, two nodes are repeatedly selected from the dimensionality reduction set to eliminate all user loads and branch nodes, and when the characteristic signal is injected into the branch node, and there is no identification signal with the same partial attributes as the characteristic current in all nodes of the low-voltage distribution station area, the topology identification is completed, the root node that is always retained is incorporated into the topological hierarchy relationship, and all nodes are spliced ​​according to all topological hierarchy relationships to generate a topological structure diagram of the low-voltage distribution station area.

Images