KR102631020B1 - Distribution System Relationship-set-based Data Matching Method and Integrated DB System for Distribution System - Google Patents

Abstract

The distribution system relationship set-based data matching method of the present invention includes the steps of collecting data for matching; graph modeling the collected data; Performing node matching on graph modeled data; If there are nodes that do not match, measuring the similarity between the facility sets of the nodes; And it may include performing matching according to the measured similarity between the sets of facilities.

Description

Distribution System Relationship-set-based Data Matching Method and Integrated DB System for Distribution System}

The present invention relates to a relationship set-based data matching device and method that ensures consistency for unifying distribution system information. In particular, a relationship set-based data matching device and method that can unify distribution facility information data and distribution operation information. It's about.

The currently operating Total Distribution Automation System (TDAS) is a system developed with Korea Electric Power Corporation (KEPCO)'s own technology and has been verified as having high reliability as a result of actual distribution system operation, stably managing the nationwide distribution system centered on a total of 41 distribution centers. It's in progress. However, with the advent of the 4th Industrial Revolution, rapidly growing information and communication technologies are being applied to various fields such as the economy and industry, and there is a need to incorporate the latest ICT technology into the operating comprehensive distribution automation system.

In particular, with the announcement of the Renewable Energy 3020 Implementation Plan, the spread of renewable energy began in earnest, and the resulting increase in the acceptance rate of distributed power and the emergence of numerous information devices such as IoT sensors are increasing the complexity and scale of the distribution network and the operating capacity within the system. . Based on these diverse requirements, we are facing a new challenge of improving the distribution intelligence system to accommodate and efficiently manage changes in the distribution system.

In accordance with this trend, KEPCO is conducting research and development on a next-generation intelligent distribution management system (ADMS) that can actively respond to changes in the distribution environment with the aim of completely reorganizing the comprehensive distribution automation system currently in operation.

The next-generation power distribution intelligence system under development is a system that guarantees interoperability, standardization, openness, functional modularity, and integrity. It efficiently analyzes massive amounts of data and suggests optimization measures to efficiently operate the increasingly complex distribution system and generates electricity from the system. Failure prediction and state estimation technology can be advanced to respond to various phenomena in real-time.

Meanwhile, Sales & Marketing and Distribution Information System 40 (SDIS40) is an integrated management system that supports overall distribution tasks such as distribution design and facility operation by applying GIS-based technology. It is an integrated management system that supports distribution tasks such as distribution technology calculations and system analysis. It is a system that improves reliability and provides various management information in the distribution field, such as facility and construction status and power outage management.

Currently, the grid relational database (DB) of SDIS40 and ADMS, which were built by KEPCO to operate the actual distribution system, has distribution facility information data and distribution operation information data built in different DBs and operated independently. SDIS40 system data handles and stores distribution facility data such as customer information (number of units, contracted power), distribution facility information (geungjang, wire specifications, pole information, etc.), design/completion information, distributed power cumulative capacity, etc. for power outage duties and protection. The purpose is to provide services for smooth business processing, such as cooperation and device operation review. On the other hand, ADMS grid data stores distribution operation information data such as real-time grid connection information, section load, and real-time power outage section.

However, in response to various improvement requirements such as power outage management and high-voltage customer management, attempts are being made to improve work and provide services based on information linkage between SDIS40 and ADMS. However, the distribution system information DB is operated and managed in a dual manner, improving user work efficiency and service. There is a problem of quality deterioration. This is because the system DBs of SDIS40 and ADMS are designed to be stored in different schemas according to the respective purposes of distribution facility management and system operation.

Figure 1 is a graph showing the status of continuous information linkage between distribution system DBs.

Looking at it specifically, first, when equipment information in the SDIS40 system is changed (additional, deleted) due to distribution facility construction, etc., updating and automatic synchronization of the information linked to the ADMS system is required at certain intervals. , Due to the lack of a specific solution, information linking is being done manually.

Second, the number of facilities that cannot be matched with each other is gradually increasing due to the absence of an organic management system and consistency/validity verification solution for the data managed in the system DBs of SDIS40 and ADMS, respectively.

Third, due to the above reasons, human errors occur and a lot of time is required due to manual information linking of the added, deleted, and changed information generated in SDIS40 to the system DB of ADMS.

Currently, the grid relational database (DB) of SDIS40 and ADMS, which were built by KEPCO to operate the actual distribution system, has distribution facility information data and distribution operation information data built in different DBs and operated independently.

However, in response to various improvement requirements such as power outage management and high-voltage customer management, attempts are being made to improve work and provide services based on information linkage between SDIS40 and ADMS. However, the distribution system information DB is operated and managed in a dual manner, improving user work efficiency and service. There is a problem of quality deterioration.

To solve these problems, the method of determining whether distribution equipment data matches simply by using SQL queries between pre-built databases considering computerized information, line information, etc. has the limitation of not guaranteeing the accuracy and performance of the results. exist. There is no common data management system between systems that are dually operated, and there are many cases where equipment IDs are inconsistent due to information input and equipment add/delete commands performed by users without verification of consistency and validity.

Therefore, in order to efficiently search for equipment data that cannot be matched between dual distribution system DBs, a data matching solution that guarantees validity and consistency based on the relationship between facilities is needed.

Republic of Korea Registered Publication No. 10-1656730

The present invention seeks to provide a data matching method and a distribution system integrated DB system based on a distribution system relationship set that can conveniently and accurately match distribution facility information data collected independently from each other and distribution operation information.

A data matching method based on a distribution system relationship set according to an aspect of the present invention includes collecting data for matching; graph modeling the collected data; Performing node matching on graph modeled data; If there are nodes that do not match, measuring the similarity between the facility sets of the nodes; And it may include performing matching according to the measured similarity between the sets of facilities.

Here, in the step of collecting data for matching, data collected for operation of the power system and data collected for managing facilities for constructing the power system may be collected.

Here, in the graph modeling step, the data collected for the operation of the power system and the data collected for managing the facilities that build the power system are modeled as a directed acyclic graph (DAG). You can.

Here, the DAG establishes a connection relationship from one node to another node, and may have the characteristic of DAG = (V,E,L), in which there is no path starting from a specific node v and returning to v.

Here, the distribution facility is expressed as a node V, the connection relationship between two nodes is expressed as an edge E, and L is a label with attribute information of the node and edge (e.g. computerized number, line name, line number, etc.) It can be a set.

Here, DAG _S = (V _S , E _S , L _S ), a graph of data collected to manage the facilities that build the power system, is a node set , a set of directed edges , and it may be a directed graph composed of a node label function L _S.

Here, the graph DAG _A = (V _A ,E _A ,L _A ) of the data collected for the operation of the power system is a node set , a set of directed edges And it can be a directed graph composed of node label functions L _A.

Here, in the step of measuring the similarity between the facility sets of the nodes, a similarity measurement between path sets (Path-set similarity check) and a similarity measurement between partial tree nodes (Subtree-based similarity check) are performed targeting a set of nodes that cannot be matched. The degree of similarity can be measured.

Here, the similarity between the partial tree nodes can be obtained by using each node in a specific node set as the root node, using the DFS algorithm to search for nodes located on the load side, and measuring the similarity between the discovered node sets. .

A distribution system integrated DB system according to another aspect of the present invention includes a system management DB in which information collected for the operation of the power system is stored; Facility management DB where information collected to manage facilities that build the power system is stored; A matching device that graph-models data stored in the system management DB and the facility management DB, compares the modeled data, and performs matching on unmatched nodes according to similarity between facility sets; And it may include an integrated DB that stores integrated data generated by matching by the matching device.

Here, the matching device can search and analyze relationships between neighboring facilities, such as power direction facilities and load direction facilities of a specific facility, perform mutual consistency verification, and then perform matching.

Here, the matching device can perform consistency verification using a relationship set-based data matching algorithm for equipment that is the same equipment but stored with different computerized numbers and then perform mutual matching.

Here, the matching device collects data stored in the system management DB and the facility management DB; graph modeling the collected data; Performing node matching on graph modeled data; If there are nodes that do not match, measuring the similarity between the facility sets of the nodes; And a step of performing matching according to the similarity between the measured sets of facilities and storing them in the integrated DB.

Here, in the graph modeling step, the data collected for the operation of the power system and the data collected for managing the facilities that build the power system are modeled as a directed acyclic graph (DAG). You can.

By implementing the distribution system relationship set-based data matching method and/or distribution system integrated DB system according to the spirit of the present invention of the above-described configuration, distribution facility information data and distribution operation information independently collected from SDIS and ADMS are conveniently and There is an advantage in being able to match accurately.

The distribution system relationship set-based data matching method and/or distribution system integrated DB system of the present invention can provide an information linkage function for effective construction of an ADMS integrated DB, distribution environment management, and verified mutual consistency between dualized distribution system data. There is an advantage.

The distribution system relationship set-based data matching method and/or distribution system integrated DB system of the present invention constructs an integrated DB that combines system data with facility and customer information to solve the problems of dualized distribution system DB operation, making it efficient. It has the advantage of being able to create various solutions such as power outage management and customer management.

Figure 1 is a graph showing the status of continuous information connection between distribution system DBs.
Figure 2 is a conceptual diagram showing an outline of the construction of a distribution system integrated DB.
Figure 3 is a flowchart showing an embodiment of a data matching method based on a distribution system relationship set according to the spirit of the present invention.
Figure 4 is a block diagram representing the SDIS40 and ADMS systems that are currently in operation.
Figure 5 is a conceptual diagram illustrating the occurrence of computerized number discrepancy and relationship search between system facilities expressed as a graph.
Figure 6 is an example diagram of DAG.
Figure 7 is an example diagram of SDIS4.0 structure modeled as DAG.
Figure 8 is an example diagram of ADMS lineage data modeled with DAG.
Figure 9 is a diagram showing the results of node matching using computerized numbers between DAG _S and DAG _A as an example of the relationship between SDIS DAG and ADMS DAG.
10 is a graph structure diagram illustrating the results of node matching between SDIS DAG and ADMS DAG.
Figure 11 is a conceptual diagram showing a similarity check process between nodes based on a relationship set to ensure consistency.
Figure 12 is a block diagram showing the connection configuration of a data matching device based on a distribution system relationship set according to the spirit of the present invention.
Figure 13 is a conceptual diagram illustrating a matching method prior to the proposed technique.
Figure 14 is a conceptual diagram illustrating a method for always matching distribution system information using the proposed technique.

In describing the present invention, terms such as first and second may be used to describe various components, but the components may not be limited by the terms. Terms are intended only to distinguish one component from another. For example, a first component may be referred to as a second component, and similarly, the second component may be referred to as a first component without departing from the scope of the present invention.

When a component is mentioned as being connected or connected to another component, it can be understood that it may be directly connected to or connected to the other component, but other components may exist in between. .

The terms used herein are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions may include plural expressions, unless the context clearly indicates otherwise.

In this specification, terms such as include or have are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, including one or more other features or numbers, It can be understood that the existence or addition possibility of steps, operations, components, parts, or combinations thereof is not excluded in advance.

Additionally, the shapes and sizes of elements in the drawings may be exaggerated for clearer explanation.

In order to effectively build an ADMS integrated DB that integrates the system DB of ADMS and the equipment DB of SDIS40, the present invention proposes a relationship set-based data matching technique that guarantees data consistency, and based on this, determines whether equipment with different equipment identifiers are the same equipment. I will explain that you can judge, estimate, and decide.

More specifically, the present invention first introduces a graph data structure to effectively manage each system data, and presents a method of building a system data management model using this. Second, we present a relationship set-based data matching algorithm to verify data consistency based on the modeled SDIS40 and ADMS graphs. Lastly, we present a method for efficient updating in the ADMS graph by using the proposed technique for the changed information set that continuously occurs in the SDIS40 graph over time.

Figure 2 is a conceptual diagram showing an outline of the construction of a distribution system integrated DB.

In order to efficiently search for equipment data that cannot be matched between dual distribution system databases, a data matching solution that guarantees validity and consistency based on the relationship between facilities is needed.

More specifically, mutual consistency verification was performed by exploring and analyzing the relationship between the power direction (substation direction, reverse direction of current flow) of a specific facility and neighboring facilities such as load direction facilities (consumer direction, direction of current flow). A matching solution is needed.

To this end, in the present invention, the systematic data of SDIS40 and ADMS built in the DB is first modeled and expressed as a Directed Acyclic Graph (DAG). This is because the graph is a data structure that can most effectively express and understand the relationships between interconnected objects, and can most effectively express the distribution system in the form of a mesh topology.

Based on the modeled SDIS40 and ADMS graphs, we propose a relationship set-based data matching algorithm that guarantees data consistency verification. The proposed relationship set-based data matching algorithm is a method of performing mutual matching after verifying the consistency of the same equipment but stored with different computerized numbers. To verify consistency, similarity measurement between path sets and subtree-node sets is performed.

Through this, when the similarity between SDIS40 and ADMS facilities is high, it is possible to provide guidance on the estimated results of the same facility and lead to user judgment based on this.

In addition, by utilizing the proposed technique for the set of information that changes (equipment addition, deletion, etc.) that continuously occur in the SDIS40 system DB in a dynamic environment, data updates with guaranteed validity and consistency can be performed in the ADMS system DB. there is.

Table 1 below presents an example of the data matching algorithm proposed in the present invention.

Figure 3 is a flowchart showing an embodiment of a data matching method based on a distribution system relationship set according to the spirit of the present invention.

The illustrated distribution system relationship set-based data matching method includes collecting data for matching; Graph modeling the collected data (S120); A step of performing node matching on graph modeled data (S140); If there are nodes that do not match (S150), measuring the similarity between the facility sets of the nodes (S160); And it may include performing matching according to the measured similarity between the equipment sets (S180).

The step of collecting data for matching is naturally the first process performed, and since there is no difference compared to the prior art, it is omitted from the detailed description and flow chart.

First, we will look at the directed acyclic graph-based systematic data modeling applied in the graph modeling step (S120).

In the domestic distribution system, the power system network from the substation to consumers (customers) is collectively referred to as the distribution system, and the scale of the system is gradually expanding and becoming more complex to ensure smooth demand and power supply. However, as many new and renewable energy sources such as distributed power are operated in connection with the distribution system according to domestic policies and various requirements, the existing unidirectional power flow flow configuration from the substation to the consumer is a loop or mesh network ( Mash Network) form. In addition, there are difficulties in maintaining consistency between each distribution system database, which is divided into continuously changing system information such as distribution facility construction, demolition, and pressurization.

Therefore, in order to operate the distribution system network and perform organic tasks in a dynamic environment where facility information changes periodically, the unique ID (Key) within each system currently in operation is based on computerized number and other information such as line name and line number. Matching is required. In addition, based on previous work, an integrated ADMS DB will be established to aim for efficient information linkage between the dualized and operating distribution system management systems.

Here, the computerized number refers to the unique facility identifier (ID) assigned to each facility, and each system assigns and manages its own unique ID within the system for smooth operation of distribution facilities and systems. (SDIS40 is ' With 'SYSID', ADMS creates 'MRID' as a unique ID and assigns it to each facility.)

Figure 4 shows the SDIS40 and ADMS systems that are currently in operation.

According to this purpose, in order to build an integrated ADMS DB for organic work between distribution system DBs with different purposes, SQL (Standard Query Language) queries are used based on the already established system DB of SDIS40 and ADMS to establish system facilities. After determining whether the computerized number, line name, and line number information match, matching between system IDs is performed. This method of matching system IDs after checking whether facility information matches has the advantage of being able to be performed without structural changes to the established DB and management system.

However, despite the above-mentioned advantages, the method of determining consistency between distribution facility data using SQL queries simply considering computerized information, line information, etc. has limitations in that it does not guarantee the accuracy and performance of the results. do. Due to the absence of a consistency or validation verification solution for manually entered information in each system DB, it is difficult to expect meaningful results when performing simple data matching using facility information. This is because accurate matching between data can only be performed if the original data is verified in advance for errors.

Meanwhile, a match is made between the above-mentioned distribution facility data based on the distribution system DB that expresses the relationship between tables using the Common Information Model (CIM), but it requires very complex queries and long execution times. It takes time. This is because heavy operations such as joins and aggregations between multiple tables occur repeatedly, increasing execution complexity. Therefore, the above method is not an appropriate problem-solving method to provide services such as power outage management and high-voltage customer management based on frequent information linkage between distribution system data.

In order to overcome the above-mentioned limitations (improve accuracy and performance) and efficiently explore facility data that cannot be matched between dual distribution system databases, consistency verification and data matching solutions are required as a result of exploring and analyzing relationships between facilities. do. To this end, the relationship between the power direction (substation direction, reverse direction of current flow) of a specific facility and neighboring facilities such as load direction facilities (consumer direction, direction of current flow) is searched and analyzed to verify mutual consistency and then matched. A solution is needed.

However, the problem of matching between identical facilities for integrated management of dualized unique IDs targeting the system DBs of SDIS40 and ADMS (TDAS), which have been operated for a long time for the purpose of management of distribution systems and facilities, is a difficult problem. There is no common data management system between the dualized operating systems, and there are many cases where equipment IDs are inconsistent as information input and equipment add/delete commands are performed by users without verification of consistency and validity.

For this reason, if there are problems such as missing data, incorrect data entry by the user, or failure to reflect updated information in a specific DB, it is very difficult to identify interdependencies by tracking and analyzing the connection relationships with multiple facilities connected to it. .

In addition, due to the above problem, there is difficulty in identifying data errors and eliminating causes for information that has been updated (added, deleted) in a dynamic environment where distribution construction, grid editing, etc. occur frequently. This is because the construction of an initial data set whose consistency has been verified must be preceded so that updated information in a dynamic environment can be stored in the DB after verifying its consistency in a timely manner.

In order to solve the above-mentioned consistency problem, the distribution system relationship set-based data matching method according to the spirit of the present invention first graph-models the collected data (S120).

Figure 5 is a conceptual diagram illustrating the occurrence of computerized number discrepancy and relationship search between system facilities expressed in a graph.

Graph modeling (representation) is performed in step S120 according to the spirit of the present invention. It is also useful in examining the problems of the conventional method described above, and the efficiency of the present invention is compared with the conventional problem with reference to FIG. 5.

Figure 5(a) shows an example of a computerized number mismatch between SDIS40 and ADMS distribution system facilities due to human error during manual input. The specific equipment d in the SDIS40 system shows a discrepancy in the computerized number with the equipment n in the ADMS system, and as shown in (b) and (c) of Figure 4, d and n are the switch/transformer equipment in the power and load directions, respectively. By comparing and analyzing the relationship, it is possible to recognize and correct computerized number errors. Therefore, in order to solve the above-mentioned limitations and effectively manage the system data of SDIS40 and ADMS built in the DB, it is useful to approach modeling and expressing it as a Directed Acyclic Graph (DAG). .

Graph is a data structure that can most effectively express and understand the relationships between interconnected objects, and is used in various real-world situations such as social networks, road networks, and chemical compounds. Alternatively, the application can be expressed using a graph.

The reasons why the problems presented above can be effectively solved by expressing the distribution system data of SDIS40 and ADMS in DAG are as follows. First, the distribution system is structured starting from the CB (Circuit Breaker) of the substation where power supply begins and ending at the internal/external connection point, branch point, terminal point, or customer. Second, this distribution system is composed of a mesh topology, which consists of complex connections between multiple facilities such as electric poles, transformers, switches, and distributed power sources. Third, graphs can be used to effectively express connection information between facilities and the direction of current flow. Lastly, when using a graph, complex relationship search operations between various facilities can be performed at high speed in memory without a join operation, so more efficient processing is possible when using a disk-based DB.

For this reason, in order to effectively process equipment that is continuously added to and deleted from the system DB of SDIS40 and update it to the system data of ADMS, the present invention performs modeling of each system data with DAG and utilizes this to verify initial data for consistency. Let's do the construction. This is because initial data construction must be preceded in order to update and manage information that continuously changes over time.

DAG = (V,E,L) establishes a connection relationship from one node to another node, and has the characteristic that there is no path starting from a specific node v and returning to v. In other words, DAG is a directed graph with topological sorting. To build the system DB of SDIS40 and ADMS as DAG, distribution facilities such as electric poles are expressed as Node V, and the connection relationship between the two nodes is expressed as Edge E. L is a set of labels containing attribute information (computerization number, line name, line number, etc.) of nodes and edges.

Meanwhile, steps S140 and S140 of FIG. 3 may be performed in a manner similar to FIG. 5(a), and step S160 of FIG. 3 may be performed in a manner similar to FIGS. 5(b) and 5(c). .

Figure 6 is an example of DAG, Figure 7 is an example of SDIS4.0 structure modeled as DAG, and Figure 8 is an example of ADMS lineage data modeled as DAG.

For effective description of the present invention, frequently used basic terms for DAG are explained using FIG. 6. For example, step S160 of FIG. 3 may be performed using DAG.

As explained above, DAG is a graph in which edges have direction, and if there is an edge e ₂ from v ₂ to v ₁ , v ₂ is the source node (source, predecessor node), and v ₁ is the target node (target, successor node). ) is expressed as v ₁ is an out-neighbor of v ₂ and belongs to the out-neighbor node set N ⁺ (v).

Likewise, v ₂ is expressed as an in-neighbor of v ₁ and belongs to the in-neighbor node set N ^- (v).

The set of all neighbors of v is Express it as Here, N ^- (v) refers to the set of equipment in the CB direction (power direction equipment) that supplies power to v, and N ⁺ (v) refers to the set of equipment in the consumer direction (load direction equipment) that receives power from v. means a set of

In addition, in DAG, there is an in-degree (In-Degree) deg ^- (v) and an out-degree (Out-Degree) deg ⁺ (v) depending on the direction of the edge based on a specific node. Here, deg ^- (v) means the number of edges coming into node v, and deg ⁺ (v) means the number of edges going out from node v. For example, deg ^- (v) becomes 2 because e ₂ and e ₃ exist, and deg ⁺ (v) becomes 1 because e ₁ exists. Figure 7 expresses the lineage data of SDIS4.0 and ADMS in DAG.

Next, we describe how to build an initial integrated data set with guaranteed consistency for smooth information linkage between ADMS graphs based on the SDIS40 graph modeled as DAG as dual system graphs. In other words, it explains that DAG can be used to effectively express distribution system data and can be used as a means to solve the problems presented.

SDIS40 graph DAG _S = (V _S ,E _S ,L _S ), which manages the overall distribution facilities built in the actual system, is a set of nodes , a set of directed edges And it is a directed graph composed of node label function L _S.

random node It consists of overhead distribution facilities such as electric poles or ground/underground distribution facilities, and when connected between distribution facilities through a high voltage line, a directional edge Use to logically express the connection relationship between facilities and the direction of current flow.

And L _S (v) is a function that retrieves label information for each node. Node label information consists of SYSID, a unique ID automatically assigned to manage all system facilities in SDIS40, facility computerization number, presence of switch, presence of transformer, line name, and line number.

ADMS graph DAG _A = (V _A ,E _A ,L _A ), which manages only distribution facilities for operating the actual system, is a set of nodes , a set of directed edges And it is a directed graph composed of node label functions L _A.

random node It consists of essential equipment for system operation, such as switches, transformers, and voltage/current value measurement, and when connected between distribution facilities through high-voltage lines, the directional edge Use to logically express the connection relationship between facilities and the direction of current flow.

And L _A (v) is a function that retrieves label information for each node. Node label information consists of MRID, a unique ID automatically assigned to manage all facility objects within the ADMS system, as well as facility computerization number, presence of switch, presence of transformer, line name, and line number.

As mentioned above, the SDIS4.0 system DB stores and overall manages all equipment information existing in the system in order to process tasks related to distribution system design, such as creating or deleting equipment.

On the other hand, the ADMS system DB stores only equipment data required for distribution system operation and does not deal with other equipment and information. In other words, DAG _A is a subset of DAG _S and is defined as DAG _A ⊆ DAG _S , and the relationships V _A ⊆ V _S and E _A ⊆ E _S are necessarily established.

Figures 7 and 8 are examples of the relationship between DAG _S and DAG _A.

However, due to the absence of a common data management system between dualized systems, the absence of consistency and validity verification steps, and data errors in the ADMS system by users, etc., matching between node computerized numbers is impossible (equipment) ) exists in large numbers. A simple way to solve this problem is to perform matching after comparing node label information such as line name, line number, etc., but it is difficult to judge it as a valid result because the consistency and validity of the original data of the ADMS system DB has not been verified first. .

Therefore, in this section, to build an initial data set with higher reliability,

go If?? and similarity between This highest node pair??

Let's solve the problem of exploring . here, silver As a function for measuring and quantifying the degree of similarity between nodes, the similarity between node sets having a connection relationship and the similarity between node sets within the searched path are measured to determine whether v _a and v _s are likely to match.

R _A is a set of nodes not included in DAG _S , and refers to a set of nodes where computerized number information mismatch occurs between distribution system facilities. On the other hand, R _S is a set that includes nodes not covered in distribution system operation and nodes belonging to R _A. thus In this case, because the relationship DAG _A ⊆ DAG _S is established is satisfied.

Figure 9 is a diagram showing the results of node matching using computerized numbers between DAG _S and DAG _A as an example of the relationship between SDIS DAG and ADMS DAG.

Figure 10 is a graph structure diagram illustrating the results of node matching between SDIS DAG and ADMS DAG.

In order to solve the problem defined above, the present invention proposes a technique for measuring mutual similarity based on the relationships discovered between nodes with different facility identifiers and recommending the node with the highest similarity through this. More specifically, the proposed technique consists of an algorithm that measures similarity using all paths that exist from a random node to a CB node and an algorithm that measures similarity using a set of all nodes connected from a random node to a terminal node. .

A method of measuring similarity uses Jaccard Similarity. Jacquard similarity is a method of calculating the similarity between two sets based on the ratio of union to intersection when two sets are given, as shown in Equation 1 below. Jacquard similarity has a value between 0 and 1. If the two sets are identical, it has a value of 1, and if the two sets have no common elements, it has a value of 0.

Algorithm 2 measures the similarity between path sets (Path-set similarity check) and the similarity between partial tree nodes (Subtree-based similarity check) targeting node sets R _{S and} R _A for which computerized number matching between DAG _A and DAG S is impossible. This is a schematic diagram of the process of measuring the degree of similarity. Through Algorithm 2, effective matching can be performed between a random node v _a belonging to R _A and the node v _s of R _S with the highest similarity.

First, the proposed algorithm for measuring similarity between path sets is a set of paths that exist from each node belonging to R _S and R _A to the CB node. As a method of searching, the purpose is to measure the similarity between the node sets that make up the path. For more efficient performance, the proposed path set search algorithm was designed using dynamic programming (DP).

DAG _S and DAG _A have the same CB node ( ) has the property of supplying power currents from terminal nodes, so nodes belonging to R _A and nodes belonging to R _S all share the same CB node. That is, two nodes with different computerization numbers and Assuming that is actually the same node (facility), from the CB node through the node belonging to CN and This means that power current is transmitted to . thus from Power flow supply routes explored to date and, from, Power flow supply routes explored to date liver silver for ?繹rit? The similarity is high. This is because the more nodes shared between facilities, the higher the likelihood that they are the same facility.

From each node v _s belonging to R _S to the CB node of DAG _S The set of all paths that exist until about the path Is A set of power direction nodes based on By exploring and repeating this recursively, Search for nodes connected up to .

Similarly, from each node v _a belonging to R _A to the CB node of DAG _A The set of all paths that exist until about path ?? Is Based on By exploring and repeating this recursively, Search for nodes connected up to . In addition, efficient performance is possible by avoiding duplicate calculations by storing and utilizing the previously searched results in an array.

The similarity between partial tree nodes is searched for nodes located on the load side using the DFS algorithm, a famous algorithm, with each node v _s belonging to R _S as the root node. Likewise, each node v _a belonging to R _A is used as the root node to search for nodes located on the load side, and then measure the similarity between the sets of discovered nodes.

Table 2 below illustrates a relationship set-based similarity measurement algorithm for measuring similarity between nodes. In other words, this is an example of a similarity check algorithm between nodes based on a relationship set to ensure consistency.

Figure 11 shows the similarity check process between nodes based on a relationship set to ensure consistency.

Node F of DAG _S and node Z of DAG _A are actually the same facility, but different computerized numbers are stored in each DB, so they are represented as non-identical facilities in computer terms. Therefore, the path to node A (CB node) is searched based on node F of DAG _S , and the set of nodes that make up this path are found. Similarly, node Z of DAG _A is also found to find the set of nodes that make up the path to node A (CB node), and then the two By measuring the similarity between sets, it is possible to estimate whether the equipment is the same.

As explained above, DAG _S and DAG _A share the same CB node, and have the property of supplying power current from the CB node to the terminal node, so if F and Z are the same node, there must be a connection between the set of nodes constituting the path. This is because it has a high degree of similarity. Similarly, based on each of node F of DAG _S and node Z of DAG _A , a set of nodes discovered through the DFS algorithm up to the terminal nodes can be formed, and the similarity between the two sets can be compared to estimate whether they are the same equipment.

Figure 12 is a block diagram showing a grid-integrated DB system with a connection configuration of a data matching device based on a distribution grid relationship set according to the spirit of the present invention.

The grid integrated DB system shown includes a grid management DB 200 in which information collected for power system operation is stored; a facility management DB (100) in which information collected to manage facilities that build a power system is stored; A matching device 400 that graph-models the data stored in the system management DB 200 and the equipment management DB 100, compares the modeled data, and performs matching on unmatched nodes according to similarity between equipment sets. ); And it may include an integrated DB 600 that stores integrated data generated by matching by the matching device 400.

According to the spirit of the present invention, the matching device 400 can search and analyze the relationship between neighboring facilities such as power direction equipment and load direction equipment of a specific equipment, perform mutual consistency verification, and then perform matching.

In addition, the matching device 400 can perform consistency verification using a relationship set-based data matching algorithm for the same equipment but stored with different computerized numbers and then perform mutual matching.

The matching device 400 can perform a data matching method based on a distribution system relationship set according to the spirit of the present invention as shown in FIG. 3.

In other words, the matching device 400 includes the following steps: collecting data stored in the system management DB 200 and the facility management DB 100; Graph modeling the collected data (S120); A step of performing node matching on graph modeled data (S140); If there are nodes that do not match (S150), measuring the similarity between the facility sets of the nodes (S160); And a step (S180) of performing matching according to the measured similarity between the equipment sets and storing them in the integrated DB (600).

Figure 13 is a conceptual diagram illustrating the matching method prior to the proposed technique.

Figure 14 is a conceptual diagram illustrating a method for always matching distribution system information using the proposed technique.

In the existing method as shown in Figure 13, information received from users was manually entered and stored in each system. In this method, user work efficiency is reduced due to the dual operation of the system (ADMS) that calculates the measured voltage, power, and load values in the distribution system and the system (SDIS40) that inputs and approves the calculated information.

On the other hand, when introducing the proposed technique as shown in Figure 14, the validity and consistency of the input information can be pre-verified and only valid input values can be stored and reflected. Through this, it is possible to determine whether there is a match between the same equipment but stored with different computerized numbers.

Those skilled in the art to which the present invention pertains should understand that the present invention can be implemented in other specific forms without changing its technical idea or essential features, and that the embodiments described above are illustrative in all respects and not restrictive. Just do it. The scope of the present invention is indicated by the claims described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

100: Facility management DB
200: System management DB
400: Data matching device based on distribution system relationship set
600: Integrated DB

Claims (14)

Collecting distribution operation information data for operating a power system and distribution facility information data for managing facilities constructing a power system;
graph modeling the collected data;
Performing node matching on graph modeled data;
If there are nodes that do not match, measuring the similarity between the facility sets of the nodes; and
Performing matching according to the similarity between the measured sets of facilities
Including,
In the graph modeling step,
The distribution operation information data and the distribution facility information data are modeled as a directed acyclic graph (DAG),
The DAG establishes a connection relationship from one node to another node, and is a directed graph with topology that starts from a specific node v and has no path back to v, and has the characteristics of DAG = (V, E, L). A data matching method based on a set of distribution grid relationships.
(Here, the electric pole is expressed as a node V, the connection relationship between two nodes is expressed as an edge E, and the attribute information of the node and edge (including the computerized number as a unique facility identifier (ID) assigned to each facility) (represented by a set of labels L with
delete delete delete delete According to paragraph 1,
DAG _S = (V _S , E _S , L _S ), a graph of data collected to manage the facilities that build the power system, is a node set , a set of directed edges , and a data matching method that is a directed graph composed of node label functions L _S.
According to paragraph 1,
DAG _A = (V _A ,E _A ,L _A ), a graph of data collected for the operation of the power system, is a set of nodes , a set of directed edges And a data matching method that is a directed graph composed of node label functions L _A.
According to paragraph 1,
In the step of measuring the similarity between the facility sets of the nodes,
A data matching method that measures the degree of similarity through similarity measurement between path sets (Path-set similarity check) and similarity measurement between partial tree nodes (Subtree-based similarity check) targeting a set of nodes that cannot be matched.
According to clause 8,
A data matching method in which the similarity between the partial tree nodes is obtained by using each node in a specific node set as the root node, using the DFS algorithm to search for nodes located on the load side, and measuring the similarity between the discovered node sets.
System management DB where information collected for power system operation is stored;
Facility management DB where information collected to manage the facilities that build the power system is stored;
A matching device that graph-models data stored in the system management DB and the facility management DB, compares the modeled data, and performs matching on unmatched nodes according to similarity between facility sets; and
Integrated DB that stores the integrated data created by matching the matching device
Including,
The matching device,
Collecting data stored in the system management DB and the facility management DB;
graph modeling the collected data;
Performing node matching on graph modeled data;
If there are nodes that do not match, measuring the similarity between the facility sets of the nodes; and
Performing matching according to the similarity between the measured sets of facilities and storing them in the integrated DB
performs,
In the graph modeling step,
The data collected for the operation of the power system and the data collected for managing the facilities that build the power system are modeled as a Directed Acyclic Graph (DAG),
The DAG establishes a connection relationship from one node to another node, and is a directed graph with topology that starts from a specific node v and has no path back to v, and has the characteristics of DAG = (V, E, L). It has an integrated distribution system DB system.
(Here, the electric pole is expressed as a node V, the connection relationship between two nodes is expressed as an edge E, and the attribute information of the node and edge (including the computerized number as a unique facility identifier (ID) assigned to each facility) (represented by a set of labels L with
According to clause 10,
The matching device,
An integrated distribution system DB system that explores and analyzes relationships with neighboring facilities such as power direction facilities and load direction facilities of a specific facility, performs mutual consistency verification, and then matches them.
According to clause 10,
The matching device,
An integrated distribution system DB system that performs mutual matching after verification of consistency using a relationship set-based data matching algorithm for the same equipment but stored with different computerized numbers.
delete delete

Images