CN115829031A - A power outage fault research and judgment method based on knowledge graph and power outage big data analysis - Google Patents

Abstract

The invention discloses a power outage fault research and judgment method based on knowledge map and power outage big data analysis, reconstructs the topological relationship of the power grid through the neo4j knowledge map, forms a triplet of entities, attributes, and relationships, and divides the map into different nodes with switches as nodes. Section; search and determine the outage equipment and outage area through the map search algorithm combined with the recall algorithm; use the backtracking algorithm to query the delay signal of the recall algorithm to correct the outage equipment and outage area. Within the framework of the knowledge map, the traceability of the power grid topology is brought into play, combined with the live signal returned by the recall algorithm, the power grid is checked segment by segment, and dozens of units in a segment can be checked within 3 minutes The live state of the transformer greatly improves the efficiency of power outage investigation, not only saves a lot of labor, but also greatly shortens the power outage response time, and the accuracy is very high. The invention has the advantages of clear thinking, good versatility, high economic value and suitable for popularization and use.

Description

A power outage fault research and judgment method based on knowledge graph and power outage big data analysis

technical field

The invention relates to the technical field of power outage area perception in distribution networks, in particular to a method for researching and judging power outage faults based on knowledge graphs and power outage big data analysis.

Background technique

With the continuous improvement of people's living standards, users have higher and higher requirements for power supply quality and reliability. The distribution network directly serves the majority of users and is related to the user's power consumption experience. Once a distribution network system fault occurs, fast and accurate fault location and isolation, and timely assessment of the impact of the fault are extremely important for minimizing power supply interruptions and improving power supply security and reliability.

As an important link in the production, transmission and use of electric energy, the power distribution system is the key link between the actual user demand side and the power generation and transmission system. Therefore, after the distribution network fails, fast and reasonable positioning to effectively guide the maintenance work is a key part of the safe and sustainable development of the current urban power grid. However, there are many factors that lead to difficulties in the research and judgment of distribution network faults: ①The harsh environment of network communication devices affects the fluency of information; ②Inconsistent development of distribution automation results in a large number of uncertain factors in fault information , affecting the integrity of information; ③When multiple complex faults occur, there will be a large number of faulty components and non-faulty components in the power-off area, and the switch protection will refuse to operate or malfunction, which will interfere with fault research and judgment. The above factors will lead to the expansion of the scope of research and judgment and the distortion of information uploading, which will affect the safe and stable operation of the power distribution system. With the increasingly complex wiring forms and the increasing types of equipment components, it is of great significance for the comprehensive development of the power system to study the big data algorithm for power outage fault judgment and improve the accuracy of fault judgment.

This method proposes a power outage research and judgment method that uses the knowledge graph to reconstruct the searchable power grid topology and combines the big data analysis algorithm. It can not only automatically judge the power outage signal in a short time, but also improves the efficiency of power outage investigation and allows power outage users to timely Feedback can be obtained, and at the same time, the human resources of a large number of investigators can be liberated. ,

Contents of the invention

In order to solve the above-mentioned deficiencies in the prior art, the present invention provides a power outage fault research and judgment method based on knowledge graph and power outage big data analysis, which can quickly respond to power outage signals without manual intervention, and relies solely on electrical signals to detect power outage equipment. Make research and judgments to increase the speed of power restoration in case of power outages, stabilize the emotions of electricity customers, and improve the quality of power grid services. Concretely, the present invention is realized like this:

A power outage fault research and judgment method based on knowledge graph and power outage big data analysis, comprising the following steps,

Step S1, embedding the big data information of the power grid equipment in the transmission line that needs to be studied and judged into the neo4 knowledge map, and performing thinning processing on the original map to form single-line data of the line, and obtain the constructed neo4 knowledge map;

Step S2, numbering and marking the power grid equipment in the constructed neo4j knowledge map, and dividing them into several different sections with switches as nodes, and numbering them;

Step S3, obtaining and filtering the transformer outage signal data of the power grid equipment in each line;

Step S4, use cypher language to construct neo4j map search algorithm, combined with the transformer call test algorithm, search for the power outage area in the direction of the upstream substation in the map through the power outage signal, if the call test signal is charged, stop the search, merge the same power outage data and return Power outage area equipment; if the calling test signal is not charged, then enter step S5;

Step S5. After completing the search of a section, retrospectively check the delay signal of the previous section. If there is no delay signal returned, search for the blackout area of the next section in the direction of the upstream substation in the map and execute step S4. If there is a delayed signal return, stop the next stage of search and return to the previous section;

Step S6, stop searching when finding a powered area, modify or merge the blackout area through the neo4j map structure, and return to the blackout area device.

Further, the filtering step in the step S3 includes: taking the processed line single-line diagram as a node, discretizing the line, filtering the power failure signal data, and using the line clustering algorithm to analyze the short-term The power outage signals at intervals are clustered and analyzed, and the power outage signals with low reliability are not deleted.

Further, in the step S1, before embedding the big data information of the power grid equipment in the transmission line that needs to be studied and judged into the neo4 knowledge map, use the transfomer model lac encapsulated by the Baidu paddle framework to identify entities and relationships, and also include: power grid The GIS data of the device is extracted and processed to make it into triplet data of entity nodes, entity attributes, and relationships between entities, and embedded in the neo4j graph database.

Further, the step S6 also includes: if the backtracking delay signal has electricity, then according to the section number identified in the backtracking algorithm, all the sections after this section are marked as having electricity, and the previous sections are then Still flagged as outages, and return devices in those segments.

Further, in the step S4, the algorithm is written in Java language. When the algorithm receives the transformer outage alarm signal, it first uses the alarm transformer as a starting point to search for the direction of the substation of the line through the cypher neo4j graph search language, and The line between the transformer and the substation is divided into sections with switches as nodes.

Further, in the step S4, if the number of transformers in the section exceeds 10, 10 transformers are randomly selected for testing in order to prevent channel blockage of the testing algorithm.

Introduction to the working principle and beneficial effects of the present invention: the present invention reconstructs the topological relationship of the power grid through the neo4j knowledge graph, forms a triplet of entities, attributes, and relationships, and divides the graph into different sections with switches as nodes; through the graph search algorithm Combined with the recall algorithm to search and determine the outage equipment and outage area; through the backtracking algorithm to query the delay signal of the recall algorithm to correct the outage equipment and outage area. Within the framework of the knowledge map, the traceability and searchability of the power grid topology is brought into play, and combined with the live signal returned by the recall algorithm, the power grid is checked section by section, and dozens of units in a section can be checked within 3 minutes. The live state of the transformer greatly improves the efficiency of power outage investigation, not only saves a lot of labor, but also greatly shortens the power outage response time, and the accuracy is very high. The present invention can quickly, accurately and automatically find all relevant power outage equipment in the current power outage event, save human resources to check lines, speed up fault repair, and improve service quality of the power grid. The invention has the advantages of clear thinking, good versatility, high economic value and suitable for popularization and use.

Description of drawings

Fig. 1 is a flow chart of a power outage fault research and judgment method based on knowledge map and power outage big data analysis of the present invention;

Fig. 2 is a schematic diagram of the data storage structure in the backtracking algorithm of the present invention;

Fig. 3 is a study and judgment result of the present invention in a practical application of a power grid, wherein the dark lines are live lines, the light lines are power outage lines, and the junction points of live lines and power outage lines are fault points.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are exemplary only, and are not intended to limit the scope of the present invention. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concept of the present invention.

Example 1: A power outage fault research and judgment method based on knowledge graph and power outage big data analysis

Extract and process the GIS data of power grid equipment to turn it into triplet data of entity nodes, entity attributes, and relationships between entities, and embed the triplets into the neo4j graph database through the neo4j-admin import method that comes with neo4j. Form a knowledge map, and then thin the knowledge map to remove some irrelevant entity nodes and the relationship between related entities, simplify the knowledge map, and form a single-line diagram of each line to make it easier to operate;

The single-line diagram of the neo4j map of the processed line uses the switch as a node, and discretizes the line to form a segmented single-line diagram;

Filter the power outage signal data, use the line clustering method to cluster and analyze the short time interval power outage signals on the same line, and delete the power outage signals with low reliability without processing;

Through the graph search algorithm written in the cypher language that comes with neo4j, combined with the transformer call test algorithm, after analyzing the return information of the call test, if the transformer signal in this section does not return, it means that there is a high probability of no electricity, and then go upstream in the map In the direction of the substation, search for the power outage area in the next section, and if the return signal has power, stop the search for the next stage. The cypher search algorithm is as follows:

Search for all devices in a section1:

MATCH p=(ns{dev_id:'%s'})-[*]-(dev)

where apoc.coll.duplicates(nodes(p))=[]

and all(x in nodes(p)where x.feederId='%s')

and all(x in nodes(p)where not x.dev_id='%s')

and all(x in relationships(p)where x.feederid='%s')

return dev.dev_id AS DeviceID,labels(dev)AS DeviceLabel,dev.feederIdAS FeederID,dev.classId AS ClassID,dev.buro AS Buro,dev.subBuro AS SubBur,dev.trans_id AS TransformerID,dev.fromDev AS fromDevID",

Search for all devices in a section2:

MATCH p=shortestpath((s{dev_id:'%s'})-[*]-(e{dev_id:'%s'}))

where apoc.coll.duplicates(nodes(p))=[]

with[head(relationships(p))]as h_r

match pp=(ns{dev_id:'%s'})-[*]-(dev)

where apoc.coll.duplicates(nodes(pp))=[]

and all(x in nodes(pp)where x.feederId='%s')

and all(x in relationships(pp)where not x in h_r and x.feederid='%s')

and all(x in nodes(pp) where not x.dev_id='%s')

return dev.dev_id AS DeviceID,labels(dev)AS DeviceLabel,dev.feederIdAS FeederID,dev.classId AS ClassID,dev.buro AS Buro,dev.subBuro AS SubBur,dev.trans_id AS TransformerID,dev.fromDev AS fromDevID"

Search for the shortest path:

MATCH(dev_start{dev_id:'%s'}),(dev_end{dev_id:'%s'}),

path=shortestpath((dev_start)-[*]-(dev_end))

WHERE apoc.coll.duplicates(nodes(path))=[]

RETURN length(path) AS PathLength

The closest upstream switch on the search path:

MATCH(dev_start{dev_id:'%s'}),(dev_end{dev_id:'%s'}),

path=shortestpath((dev_start)-[*]-(dev_end))

WHERE apoc.coll.duplicates(nodes(path))=[]

WITH path

MATCH(dev)

WHERE dev IN nodes(path)

AND(dev:%s OR dev:%s OR dev:%s)

AND NOT dev.dev_id='%s'

RETURN dev.dev_id AS DeviceID;

After completing a stage of search, backtrack to check the delay signal of the previous section. If there is no delay signal return of the transformer signal in the previous section, it means that there is a high probability that there is no power, and then proceed to the upstream substation in the map. Search for a power outage area in one section, if the return signal has power, stop the next stage of search;

When the live area is searched, the search is stopped and the equipment in the power-off area is returned. If the backtracking delay signal has power, all the sections from this section onwards will be marked as having electricity according to the section number identified in the backtracking algorithm. , the previous section is still marked as a power outage, combined with the neo4j map structure to correct or merge the outage area, and return the equipment in these sections;

The present invention can quickly, accurately and automatically find all relevant power outage equipment in the current power outage event, save human resources to check lines, speed up fault repair, and improve service quality of the power grid. Main features of the present invention are:

1. Reconstruct the topological relationship of the power grid through the neo4j knowledge graph, form a triplet of entities, attributes, and relationships, and divide the graph into different sections with switches as nodes;

2. Through neo4j's cypher language, the map search algorithm is combined with the call-to-test algorithm to search and determine the power-off equipment and power-off area;

3. Correct the outage equipment and merge the outage area through the constructed neo4j knowledge map structure.

The beneficial effect of the present invention is that within the framework of the neo4j knowledge map, the traceable searchability of the power grid topology is brought into play, and an easier-to-handle graph search algorithm is written through the convenience of the cypher language, and combined with the electrification returned by the recall algorithm Signal, check the power grid section by section, and check the live status of dozens of transformers in a section within 3 minutes, which greatly improves the efficiency of power outage investigation, not only saves a lot of labor, but also greatly shortens the response to power outages time, and the accuracy is very high. The invention has the advantages of clear thinking, good versatility, high economic value and suitable for popularization and use.

Embodiment 2: Example

A power outage fault research and judgment method based on knowledge graph and power outage big data analysis, based on the operation research and judgment process and results of a certain area under a certain power grid, and the power grid will be analyzed for nearly a year from January 2021 to early December 2021 The specific steps of a power outage research and judgment are as follows:

(1) Convert the latest GIS data of the power grid into triplets of entity, attribute, and relationship, and embed the triplet into the neo4j map database through the neo4j-admin import method that comes with neo4j. The specific statement is: neo4j-adminimport-- database=neo4j --nodes "import/sto.csv" --nodes "import/con.csv" --relationships "import/rel.csv" --ignore-empty-strings --skip-bad-relationships --skip -duplicate-nodes=true, construct the single-line diagram knowledge graph of transmission lines.

(2) The power failure signal filtering algorithm is written in Java language. After receiving the transformer power failure alarm signal, it is judged by the power failure signal filtering algorithm that the power failure alarm is true, and the signal is sent to the power failure equipment search and judgment algorithm.

(3) The search and judgment algorithm for outage equipment is written in Java language. When the algorithm receives the transformer outage alarm signal, it first takes the alarm transformer as a starting point, uses the cypherneo4j graph search language to find the direction of the substation of the line, and compares the transformer with the The lines between substations are divided into sections with switches as nodes.

(4) Use the graph search algorithm written in the cypher language that comes with neo4j to search for the transformer in the next section of the line, and conduct a call test on the search results. If the number of transformers in this section exceeds 10, in order to prevent call test algorithm channel For blockage, 10 transformers were randomly selected for testing. The cypher search algorithm is as follows:

MATCH p=(ns{dev_id:'%s'})-[*]-(dev)

where apoc.coll.duplicates(nodes(p))=[]

and all(x in nodes(p)where x.feederId='%s')

and all(x in nodes(p)where not x.dev_id='%s')

and all(x in relationships(p)where x.feederid='%s')

return dev.dev_id AS DeviceID,labels(dev)AS DeviceLabel,dev.feederIdAS FeederID,dev.classId AS ClassID,dev.buro AS Buro,dev.subBuro AS SubBur,dev.trans_id AS TransformerID,dev.fromDev AS fromDevID",

MATCH p=shortestpath((s{dev_id:'%s'})-[*]-(e{dev_id:'%s'}))

where apoc.coll.duplicates(nodes(p))=[]

with[head(relationships(p))]as h_r

match pp=(ns{dev_id:'%s'})-[*]-(dev)

where apoc.coll.duplicates(nodes(pp))=[]

and all(x in nodes(pp)where x.feederId='%s')

and all(x in relationships(pp)where not x in h_r and x.feederid='%s')

and all(x in nodes(pp) where not x.dev_id='%s')

return dev.dev_id AS DeviceID,labels(dev)AS DeviceLabel,dev.feederIdAS FeederID,dev.classId AS ClassID,dev.buro AS Buro,dev.subBuro AS SubBur,dev.trans_id AS TransformerID,dev.fromDev AS fromDevID",

MATCH(dev_start{dev_id:'%s'}),(dev_end{dev_id:'%s'}),

path=shortestpath((dev_start)-[*]-(dev_end))

WHERE apoc.coll.duplicates(nodes(path))=[]

RETURN length(path) AS PathLength

MATCH(dev_start{dev_id:'%s'}),(dev_end{dev_id:'%s'}),

path=shortestpath((dev_start)-[*]-(dev_end))

WHERE apoc.coll.duplicates(nodes(path))=[]

WITH path

MATCH(dev)

WHERE dev IN nodes(path)

AND(dev:%s OR dev:%s OR dev:%s)

AND NOT dev.dev_id='%s'

RETURN dev.dev_id AS DeviceID

(5) Analyze the return results of the transformer recall test, exclude the test recall results in the transformer white list and do not analyze them, and enter the recall test backtracking algorithm after the comprehensive analysis of the remaining transformer return results is no power, and the recall test backtracking algorithm is implemented by Java Write in language, look back at the delayed return signal of the last round of recall test, if the analysis result is still no power, then enter the next stage of search, find the transformer in the next line section and call for test.

(6) When the signal analysis result returned by the recall test or the delayed signal analysis result returned by the recall test backtracking algorithm is that there is power, stop searching for the upstream substation, and then modify or merge the blackout area through the graph structure constructed by neo4j, and finally return All outage transformers and other line equipment currently searched. See Figure 2 for the final result.

From the above calculation process, it can be seen that a power outage fault research method based on knowledge graph and power outage big data analysis can quickly, accurately and automatically find all relevant power outage equipment in the current outage event, save human resources to check lines, speed up fault repair, and improve Grid service quality.

It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principle of the present invention, and not to limit the present invention. Therefore, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention shall fall within the protection scope of the present invention. Furthermore, it is intended that the appended claims of the present invention embrace all changes and modifications that come within the scope and metesques of the appended claims, or equivalents of such scope and metes and bounds.

Claims (6)

1. A power outage fault research and judgment method based on knowledge map and power outage big data analysis, characterized in that it comprises the following steps, Step S1, embedding the big data information of the power grid equipment in the transmission line that needs to be studied and judged into the neo4 knowledge map, and performing thinning processing on the original map to form single-line data of the line, and obtain the constructed neo4 knowledge map; Step S2, numbering and marking the power grid equipment in the constructed neo4j knowledge map, and dividing them into several different sections with switches as nodes, and numbering them; Step S3, obtaining and filtering the transformer outage signal data of the power grid equipment in each line, Step S4, use cypher language to construct neo4j map search algorithm, combined with the transformer call test algorithm, search for the power outage area in the direction of the upstream substation in the map through the power outage signal, if the call test signal is charged, stop the search, merge the same power outage data and return Power outage area equipment; if the calling test signal is not charged, then enter step S5; Step S5. After completing the search of a section, retrospectively check the delay signal of the previous section. If there is no delay signal returned, search for the blackout area of the next section in the direction of the upstream substation in the map and execute step S4. If there is a delayed signal return, stop the next stage of search and return to the previous section; Step S6, stop searching when finding a powered area, modify or merge the blackout area through the neo4j map structure, and return to the blackout area device. 2. The filtering step in the step S3 includes: taking the processed line one-line diagram as a node, discretizing the line, filtering the power failure signal data, and using the line clustering algorithm to analyze the short time interval on the same line The outage signals with low reliability will be deleted without processing. 3. In the step S1, before embedding the big data information of the power grid equipment in the transmission line that needs to be studied and judged into the neo4 knowledge map, it also includes: extracting and processing the GIS data of the power grid equipment to make it into a physical node, Triple data of entity attributes and relationships between entities, and embedded in neo4j graph database. 4. The step S6 also includes: if the backtracking delay signal has electricity, then according to the section number identified in the backtracking algorithm, all subsequent sections from this section are marked as having electricity, and the previous sections are still Devices identified as outages and returned to those segments. 5. In the step S4, the algorithm is written in Java language. After the algorithm receives the transformer outage warning signal, it first takes the warning transformer as a starting point, searches for the direction of the substation of the line through the cypher neo4j graph search language, and sends the The line between the transformer and the substation is divided into sections with switches as nodes. 6. In step S4, if the number of transformers in the section exceeds 10, 10 transformers are randomly selected for testing in order to prevent channel blockage of the testing algorithm.

Images