Power distribution network power supply radius dynamic calculation method and system based on space-time knowledge graph

Abstract

The present application relates to the field of power distribution network operation management, in particular to a power distribution network power supply radius dynamic calculation method and system based on space-time knowledge graph, the method comprising: obtaining space-time account information data of power consumption equipment from multi-source power grid business systems; according to knowledge graph technology, combing equipment basic account, equipment deployment location, equipment change record, constructing power distribution network space-time knowledge graph; calculating the spatial distance corresponding to the path of all corresponding branch relationships under each entity; calculating the maximum value in the spatial distance corresponding to the path of all corresponding branch relationships under the entity, and outputting the maximum value as the power supply radius of the entity equipment. The present application constructs a knowledge graph based on the space-time data of power grid equipment to obtain a power distribution network structure diagram, and then analyzes the spatial distance of the structure path to obtain the power supply radius of each entity equipment in the power distribution network, thereby providing a low-cost, high-efficiency and high-accuracy power distribution network power supply radius dynamic calculation method.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system distribution network operation and management, and particularly relates to a power distribution network power supply radius dynamic calculation method and system based on a space-time knowledge graph. BACKGROUND

[0002] With the rapid increase in the number of newly planned residential projects, the load center of the power grid is shifted, resulting in a phenomenon of excessively large power supply radius of some distribution lines, which increases the line loss rate. This leads to low voltage phenomenon, which hinders the power supply to residents and has a great impact on their daily use. This is because the excessively large power supply radius will affect the power supply quality, resulting in a decline in voltage quality and affecting customer use.

[0003] Known power supply radius calculation methods usually use a static method to measure and calculate the power supply radius of the distribution network under a state quantity. However, due to the demand for planning and construction, power grid reconstruction, and other work, the users and lines of the distribution network may change, resulting in inaccurate power supply radius calculated based on a certain time in the past. Therefore, it is particularly important to provide a power distribution network power supply radius dynamic calculation method and system to improve the efficiency and accuracy of power supply radius calculation. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a power distribution network power supply radius dynamic calculation method and system based on a space-time knowledge graph. The present application mainly based on multi-source power grid business system data, sorts out the space-time information of power supply equipment of the distribution network, constructs a distribution network structure graph based on a space-time knowledge graph, and dynamically calculates the power supply radius of the power supply equipment, thereby solving the problem of inaccurate power supply radius calculation of the distribution network caused by the inaccuracy of the distribution network structure graph based on single data source and static data quality problems.

[0005] One embodiment of the present application proposes a power distribution network power supply radius dynamic calculation method based on a space-time knowledge graph, which includes the following steps:

[0006] Obtain the space-time account information data of the power consumption equipment from the multi-source power grid business system;

[0007] According to the knowledge graph technology, sort out the equipment basic account, equipment deployment position, and equipment change record, and construct a power distribution network space-time knowledge graph;

[0008] Calculate the spatial distance corresponding to the path of all corresponding branch relationships under each entity;

[0009] Calculate the maximum value in the spatial distance corresponding to the path of all corresponding branch relationships under the entity, and output the maximum value as the power supply radius of the entity equipment.

[0010] Another embodiment of the application provides a power distribution network power supply radius dynamic calculation system based on a space-time knowledge graph, comprising the following modules:

[0011] A data acquisition module is configured to acquire space-time account information data of power consumption equipment from a multi-source power grid business system;

[0012] A graph construction module is configured to sort device basic account, device deployment location and device change record according to a knowledge graph technology, and construct a power distribution network space-time knowledge graph.

[0013] A power supply radius calculation module is configured to calculate the spatial distance corresponding to all corresponding branch relationships of each entity, and calculate the maximum value in the spatial distance corresponding to all corresponding branch relationships of the entity, and output the maximum value as the power supply radius of the entity equipment.

[0014] Compared with the prior art, the application has the following beneficial effects:

[0015] 1. The application acquires space-time account information of power consumption equipment in a multi-source power grid business system, constructs a power distribution network structure graph with high accuracy and dynamics, acquires the longest branch line length of each node based on the constructed power distribution network structure graph, and dynamically calculates the power supply radius of the node, thereby solving the problem of inaccurate power distribution network structure graph caused by single data source and static data quality, and the problem of inaccurate power distribution network power supply radius calculation caused by the inaccurate power distribution network structure graph, improving the efficiency and accuracy of power supply radius overlength determination, and reducing the error and error probability caused by manual determination and system data consistency.

[0016] 2. The application constructs a power distribution network structure graph based on the space-time data of the power grid equipment, analyzes the spatial distance of the structure path, and obtains the power supply radius of each entity equipment of the power distribution network, thereby providing a low-cost, high-efficiency and high-accuracy power distribution network power supply radius dynamic calculation method. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 The figure is a general flowchart of the power distribution network power supply radius dynamic calculation method based on the space-time knowledge graph in the embodiment of the application.

[0018] Figure 2 The figure is a flowchart of acquiring the equipment account data and historical equipment account change information of the power grid system in the embodiment.

[0019] Figure 3 The figure is a specific flowchart of constructing the power distribution network structure graph based on the space-time knowledge graph in the embodiment.

[0020] Figure 4 The figure is a specific flowchart of calculating the node power supply radius based on the power distribution network structure graph in the embodiment.

[0021] Figure 5 This is a structural diagram of the power supply radius calculation system for power distribution networks provided in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0023] Example 1

[0024] like Figure 1 As shown, this embodiment provides a method for dynamically calculating the power supply radius of a distribution network based on a spatiotemporal knowledge graph. By collecting spatiotemporal ledger information of electrical equipment from a multi-source power grid business system, a highly accurate and dynamic distribution network structure graph is constructed. Based on the constructed distribution network structure graph, the longest branch line length of each node is obtained, thereby dynamically calculating the power supply radius of that node. The specific implementation process includes the following steps:

[0025] S100. Obtain the spatiotemporal ledger information data of power-consuming equipment from the multi-source power grid business system.

[0026] This step can obtain multi-source equipment ledger data of electrical equipment from multiple power grid business systems, such as the power grid marketing system, data acquisition system, PMS system, and GIS system. Figure 2 As shown, the details are as follows:

[0027] S101. Identify the data sources related to the basic ledger of distribution network equipment in multiple power grid business systems.

[0028] S102. Apply for data extraction permissions for each power grid business system one by one, and obtain data access and extraction permissions for each data source.

[0029] This system acquires existing equipment ledger data and historical equipment ledger change information from systems such as the power grid marketing system, data acquisition system, PMS system, and GIS system. In this embodiment, the existing equipment ledger data and historical equipment ledger change information include information such as "equipment name," "equipment type," "line length," "location information," and "change records."

[0030] S103. Set the data extraction criteria and extraction range. In this embodiment, the extraction criteria include existing data and historical status data, and the extraction range is all target data.

[0031] S104, extract relevant data of substations, distribution lines, transformers, feeder lines, users, etc. The data extracted in this embodiment includes existing equipment account data and historical equipment account change information, such as "equipment name", "equipment type", "line length", "location information" and "change record" and the like.

[0032] S105, establish a data monitoring and incremental extraction mechanism, and when the target data of the target system changes, the changed data is incrementally extracted.

[0033] S200, according to the knowledge graph technology, comb the equipment basic account, equipment deployment location, equipment change record and the like, and construct the power distribution network space-time knowledge graph.

[0034] Referring to Figure 3 , the implementation process of this step specifically includes:

[0035] S201, extract equipment account data and relationship data.

[0036] S202, identify ontology and relationship, and configure the graph structure belonging to each power grid business system. In this embodiment, the ontology includes substations and transformers, and is represented as nodes; the relationship includes the lines between substations and transformers and between transformers and users, and is represented as paths; the system includes power grid marketing systems, collection systems, PMS systems and GIS systems and the like. By configuring the substation nodes, transformer nodes, paths between substations and transformers, and paths between transformers and users, the power distribution network structure graph of the power grid marketing system, the collection system, the PMS system and the GIS system is configured respectively.

[0037] S203, compare the inconsistent graph structures in each power grid business system. Each power grid business system has a counterpart business department, so the power distribution network structure graph in the power grid business system will change due to the development of business, and due to data synchronization and other problems, the data in each system will be inconsistent, which will lead to inconsistency in the power distribution network structure graph generated based on the data of each system. The inconsistent places of the graph structure of each power grid business system are marked by comparison.

[0038] S204, according to the time sequence data of the change record, the inconsistent graph structure is discriminated and corrected.

[0039] S205, according to the spatial data of the location information, the inconsistent graph structure is discriminated and corrected.

[0040] S206, output the corrected power distribution network unique dynamic graph structure based on space-time data.

[0041] In this step of constructing the power distribution network space-time knowledge graph, the time series data and the spatial data are integrated into the knowledge graph, and the triple is expanded to a five-tuple (s, p, o, t, g) to form a space-time knowledge graph. At time t i and space g k The space-time knowledge graph is defined as:

[0042]

[0043] Wherein, p is the relationship in the power distribution network structure graph, which is represented as a path in this embodiment, such as a certain line between a substation A and a transformer. s is an entity in the power distribution network structure graph, and o is a tail entity in the power distribution network structure graph. In this embodiment, both the entity and the tail entity are represented as nodes, such as a substation or a transformer, which is the target device for calculating the power supply radius. t is the timestamp information, which can be obtained from the change time in the power distribution network device change record in this embodiment. g is the spatial information, which can be obtained from the latitude and longitude information contained in the deployment location in the power distribution network device profile in this embodiment. The five-tuple h = (s, p, o, t, g) represents a space-time knowledge graph event, and the same five-tuple will be removed to reduce redundancy. represents a state of the multi-layer power distribution network structure graph at time t i and in space g k represents The five-tuple corresponding to the time and space of each element in h.

[0044] S207, store the dynamic graph structure in the graph database to improve the calculation efficiency when traversing.

[0045] S300, calculate the spatial distance corresponding to the path of each entity under all corresponding branch relationships.

[0046] Referring to Figure 4 , the implementation process of this step includes:

[0047] S301, select a target node. In this embodiment, the target node can be a certain substation, a certain transformer, or a certain transformer area.

[0048] S302, based on the unique dynamic graph structure, all power supply paths under the target node are sorted out by the traversal method. In this embodiment, the substation has a plurality of “substation-line-transformer-line-user” power supply paths, and the transformer has one or more “transformer-line-user” power supply paths.

[0049] S303, add the preset length of each segment path under each power supply path to obtain the power supply length of all power supply paths under the target node. For example, the length of the i-th power supply path under the target node A is​ may be expressed as:

[0050]

[0051] wherein, is the length of the power supply path between node A and node B in the i-th power supply path under node A, which can be obtained by the preset line length between node A and node B, is the preset length of the first segment path under the i-th power supply path under node A; is the length of the power supply path between node B and node C in the i-th power supply path under node A, is the preset length of the second segment path under the i-th power supply path under node A; is the preset length of the n-th segment path under the i-th power supply path under node A; the other parts are derived in the same way.

[0052] S304, calculate the maximum value of the length in all power supply paths under the target node as the power supply radius of the target node. For example, the power supply path length R A may be expressed as:

[0053]

[0054] wherein is the length of the first power supply path under the target node A, is the length of the second power supply path under the target node A, is the length of the i-th power supply path under the target node A.

[0055] S400, calculate the maximum value of the spatial distance corresponding to the path of all corresponding branch relationships of the entity, and output the maximum value as the power supply radius of the entity device.

[0056] Embodiment 2

[0057] As Figure 5 shown, based on the same inventive concept as embodiment 1, the embodiment provides a power supply radius dynamic calculation system of a power distribution network based on a space-time knowledge graph, which runs in a power grid company data platform, and specifically includes the following modules:

[0058] 1) a data acquisition module 51 for acquiring space-time account information data of power consumption devices from multi-source power grid business systems such as a collection marketing system, a collection system, a PMS system and a GIS system, including existing device account data and historical device account change record information;

[0059] 2) the graph construction module 52 is configured to extract entity and relationship information from the device basic account, device deployment location and device change record according to the knowledge graph technology, generate node and path information, and construct a power distribution network spatio-temporal knowledge graph based on a spatio-temporal knowledge graph;

[0060] 3) the power supply radius calculation module 53 is configured to calculate the spatial distance corresponding to all branch relationships of each entity, and calculate the maximum value in the spatial distance corresponding to all branch relationships of the entity, and output the maximum value as the power supply radius of the entity device;

[0061] 4) the data sharing module 54 is configured to generate a data wide table of the power supply radius calculation result of the power supply device, and share it with other systems in the form of API.

[0062] Specifically, the data acquisition module 51 comprises:

[0063] 1) the data acquisition sub-module 511 acquires existing device account data and historical device account change information of the power grid marketing system, the acquisition system, the PMS system and the GIS system.

[0064] 2) the data monitoring sub-module 512 is configured to monitor the changes of the device account data of the power grid marketing system, the acquisition system, the PMS system and the GIS system. The monitoring sub-module automatically starts the graph construction module 52, the power supply radius calculation module 53 and the data sharing module 54 when the source data changes, and realizes dynamic calculation of the power supply radius of the power distribution network.

[0065] 3) the data storage sub-module 513 stores the acquired existing device account data and historical device account change information of each system in the power grid company data center.

[0066] The graph construction module 52 comprises:

[0067] 1) the knowledge extraction sub-module 521 is configured to extract device account data and relationship data, and identify ontology and relationship.

[0068] 2) the graph construction sub-module 522 is configured to extract entity and relationship information, generate node and path information, and configure the graph structure belonging to each system.

[0069] 3) the graph correction sub-module 523 is configured to fuse time sequence information such as account change records and spatial information such as device deployment location data, and correct the unique dynamic graph structure of the power distribution network based on spatio-temporal data in combination with the graph structure of each system.

[0070] The power supply radius calculation module 53 comprises:

[0071] 1) path traversal submodule 531, for traversing all power supply paths under the target node;

[0072] 2) radius calculation submodule 532, for calculating the final power supply radius value from all power supply paths under the target node.

[0073] The data sharing module 54 comprises:

[0074] 1) result storage submodule 541, for storing the power supply radius information of the power distribution network calculated by the module 53;

[0075] 2) data output submodule 542, for providing the power supply radius output of the power distribution network in the form of API to other systems and users.

[0076] The embodiment of the present application obtains the time and space account information of the power consumption equipment of the multi-source system, constructs a dynamic power distribution network structure graph with high accuracy, obtains the longest branch line length of each node based on the power distribution network structure graph, and thus obtains the power supply radius of the node. The specific implementation process of the data collection module 51, the graph construction module 52 and the power supply radius calculation module 53 is described in Embodiment 1 and will not be repeated here.

[0077] The present application can effectively solve the problem of inaccurate power supply radius calculation caused by calculating the power supply radius of the dynamically changing power distribution network structure based on single state or single system data, effectively improve the accuracy of power supply radius calculation of the power distribution network, and dynamically output the power supply radius result of the power distribution network according to the change of the power distribution network structure in time, reduce the occurrence of low voltage phenomenon caused by too large power supply radius, and avoid the obstruction of the power supply of residents. The above is only a preferred embodiment of the present application and does not limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for dynamically calculating the power supply radius of a distribution network based on a spatiotemporal knowledge graph, characterized in that, Includes the following steps: Obtain spatiotemporal ledger information data of electrical equipment from the multi-source power grid business system; Based on knowledge graph technology, we sort out the basic equipment ledger, equipment deployment location, and equipment change records to construct a spatiotemporal knowledge graph of the power distribution network. Calculate the spatial distance of all corresponding branch relationships under each entity; Calculate the maximum spatial distance among all paths corresponding to the branch relationships under an entity, and output the maximum value as the power supply radius of the entity device; The process of constructing a spatiotemporal knowledge graph of a power distribution network includes: Extract equipment ledger data and relationship data; Identify entities and relationships, and configure the graph structure belonging to each power grid business system; entities include substations and transformers, represented as nodes; relationships include lines between substations and transformers, and between transformers and users, represented as paths; Compare the inconsistent graph structures in various power grid business systems and mark the inconsistencies in the graph structures of various power grid business systems. Based on the time-series data of the change records, inconsistent spectral structures are identified and corrected; Based on spatial data of location information, inconsistent spectral structures are identified and corrected; Output the corrected, spatiotemporal data-based unique dynamic map structure of the distribution network; Calculate the spatial distances of all corresponding branch relationships under each entity, including: Select the target node, which is a substation or a transformer; Based on the unique dynamic graph structure, all power supply paths under the target node are identified by traversal. Add up the preset lengths of the segmented paths under each power supply path to obtain the power supply length of all power supply paths under the target node; The maximum length among all power supply paths under the target node is calculated as the power supply radius of the target node.

2. The method for dynamically calculating the power supply radius of a distribution network according to claim 1, characterized in that, Obtain spatiotemporal ledger information data of power-consuming equipment from the multi-source power grid business system, including: We have identified the data sources related to the basic ledger of distribution network equipment in multiple power grid business systems. Apply for data extraction permissions for each power grid business system one by one, and obtain data access and extraction permissions for each data source; Set data extraction criteria and extraction range; Extract relevant data from substations, distribution lines, transformers, feeders, and users. The extracted data includes existing equipment ledger data and historical equipment ledger change information. Establish a data monitoring and incremental extraction mechanism to perform incremental extraction of the changed data when the target data of the target system changes.

3. The method for dynamically calculating the power supply radius of a distribution network according to claim 1, characterized in that, When constructing a spatiotemporal knowledge graph for a power distribution network, temporal and spatial data are integrated into the knowledge graph, and triples are expanded to quintuples (s, p, o, t, g) to form a spatiotemporal knowledge graph. At time t... i and space g k The spatiotemporal knowledge graph is defined as: Where p represents the relationship in the distribution network structure diagram, including the lines between substations and transformers, and between transformers and users, and is represented as a path; s represents the entity in the distribution network structure diagram; o represents the tail entity in the distribution network structure diagram. Both the entity and the tail entity include substations and transformers, and are represented as nodes, which are the target devices for calculating the power supply radius. t represents timestamp information, derived from the change time in the power distribution network equipment change record; g represents spatial information, derived from the latitude and longitude information contained in the deployment location in the power distribution network equipment file; the quintuple h = (s, p, o, t, g) represents a spatiotemporal knowledge graph event; The diagram of the multi-layer distribution network structure at t i Time, in space g k One of the states, express Each element in the equation corresponds to a quintuple of time and space.

4. The method for dynamically calculating the power supply radius of a distribution network according to claim 1, characterized in that, Length of the i-th power supply path under target node A Represented as: in, The length of the power supply path between node A and node B in the i-th power supply path under the target node A is obtained by using the preset line length between node A and node B. It is the preset length of the first segment path under the i-th power supply path under node A. The length of the power supply path between node B and node C in the i-th power supply path under node B is the preset length of the second segment path under the i-th power supply path under target node A; Preset the length of the nth segment path under the i-th power supply path of the target node A; The power supply path length R of target node A A Represented as: in The length of the first power supply path subordinate to target node A. The length of the second power supply path subordinate to target node A. Let be the length of the i-th power supply path under the target node A.

5. The method for dynamically calculating the power supply radius of a distribution network according to claim 2, characterized in that, Existing equipment ledger data and historical equipment ledger change information include "equipment name", "equipment type", "line length", "location information" and "change record".

6. A dynamic calculation system for the power supply radius of a distribution network based on a spatiotemporal knowledge graph, characterized in that, Includes the following modules: The data acquisition module is used to obtain spatiotemporal ledger information data of electrical equipment from the multi-source power grid business system; The knowledge graph construction module is used to organize equipment basic ledgers, equipment deployment locations, and equipment change records based on knowledge graph technology to construct a spatiotemporal knowledge graph of the power distribution network. The power supply radius calculation module is used to calculate the spatial distance corresponding to the path of all corresponding branch relationships under each entity, and to calculate the maximum value among the spatial distances corresponding to the path of all corresponding branch relationships under the entity, and output the maximum value as the power supply radius of the entity device. The process of constructing a spatiotemporal knowledge graph of a power distribution network includes: Extract equipment ledger data and relationship data; Identify entities and relationships, and configure the graph structure belonging to each power grid business system; entities include substations and transformers, represented as nodes; relationships include lines between substations and transformers, and between transformers and users, represented as paths; Compare the inconsistent graph structures in various power grid business systems and mark the inconsistencies in the graph structures of various power grid business systems. Based on the time-series data of the change records, inconsistent spectral structures are identified and corrected; Based on spatial data of location information, inconsistent spectral structures are identified and corrected; Output the corrected, spatiotemporal data-based unique dynamic map structure of the distribution network; The power supply radius calculation module calculates the spatial distance corresponding to the paths of all corresponding branch relationships under each entity, including: Select the target node, which is a substation or a transformer; Based on the unique dynamic graph structure, all power supply paths under the target node are identified by traversal. Add up the preset lengths of the segmented paths under each power supply path to obtain the power supply length of all power supply paths under the target node; The maximum length among all power supply paths under the target node is calculated as the power supply radius of the target node.

Images