Distribution network fault plan knowledge graph design method and system based on graph database

Abstract

A method and system for designing a knowledge graph of a distribution network fault plan based on a graph database, the method comprising: obtaining a fault event ontology, a fault impact ontology, a fault handling measure ontology, and a fault handling process ontology as nodes in a knowledge graph of a distribution network fault plan based on the knowledge of experts in the field of power grids according to a distribution network fault handling plan; designing edge relationships in a knowledge graph of a distribution network fault plan according to the connection relationship between fault events, fault impacts, fault handling measures, and fault handling processes; obtaining distribution network fault plan entities based on a distribution network fault plan entity recognition model; importing distribution network fault plan entities into a graph database according to the nodes and edge relationships in the knowledge graph of a distribution network fault plan. The present invention establishes a knowledge graph of a distribution network fault plan by sorting and classifying distribution network fault handling plans; and provides a method for large-scale batch import of distribution network fault handling plan data based on a graph database.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network fault handling plans, and more specifically, to a distribution network fault plan knowledge graph design method and system based on a graph database. Background Art

[0002] The grid fault handling plan plays an important role in guiding the efficient and orderly emergency response of grid accidents. With the rapid development of the power system, the grid structure and operation mode are becoming more and more complex, and the difficulty of handling after a fault is increasing. The traditional dispatching decision-making mechanism that relies on manual experience or simple file retrieval is increasingly unable to cope with the rapid fault analysis and fault handling of complex large power grids.

[0003] The knowledge graph technology is used to extract, represent and manage fault handling plan information, and to assist dispatchers in fault handling, which can effectively improve the power grid's emergency handling capabilities and the level of intelligent dispatching.

[0004] Prior art 1 (CN113283704A) "Intelligent handling system and method for power grid faults based on knowledge graph", the fault handling plan parsing module of the system is used to obtain fault handling entity and relationship knowledge; the graph generation module is used to store the identified entities and entity relationships in the form of triples, and obtain the equipment entity knowledge graph, the accident plan knowledge graph and the handling process knowledge graph; the fault perception module intelligently perceives and collects information related to power grid faults in the power grid system; the fault risk assessment module is used to identify risks; the fault intelligent handling module is used to infer the handling measures suitable for the current fault based on the accident plan knowledge graph and the handling process knowledge graph after a line fault occurs in the power grid, and combined with the topological changes, flow and AC power supply frequency before and after the power grid fault, and guide the dispatching personnel to intelligently handle the power grid risk and restore power supply based on the established knowledge graph through all-round perception of comprehensive intelligent alarm information. Prior Art 2 (CN111552804A) “A method for constructing a knowledge graph of a power grid fault handling plan” constructs a knowledge ontology of the fault handling plan and forms an ontology graph based on the main contents of the fault handling plan, “post-fault methods” and “handling points”, and then automatically extracts fault plan entities by combining a named entity recognition model based on Bi-LSTM+CRF and rule-based template extraction, and finally realizes the construction of a knowledge graph. Prior Art 2 can construct unstructured text content into a structured, associative, and strong business logic fault handling knowledge representation based on the content characteristics of the power grid fault handling plan and the logical relationship of its internal information, which can overcome the shortcomings of traditional retrieval and application methods of fault plan text information, and can provide more efficient and powerful basic data support for accurate retrieval, mining reuse, and association expansion of fault plan information. However, both Prior Art 1 and 2 cannot effectively identify the fault event ontology and the fault impact ontology, and thus cannot provide accurate handling measures for distribution network faults. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, the purpose of the present invention is to provide a distribution network fault plan knowledge graph design method and system based on a graph database. The present invention establishes a distribution network fault plan knowledge graph by sorting and classifying distribution network fault handling plans; based on the graph database, a method for large-scale batch import of distribution network fault handling plan data is provided.

[0006] The present invention adopts the following technical solution.

[0007] On the one hand, the present invention proposes a distribution network fault plan knowledge graph design method based on a graph database, including:

[0008] Step 1: Collect structured data and unstructured data of distribution network fault handling plan; wherein, the structured data and unstructured data of distribution network fault handling plan both include: terms in the field of fault events, terms in the field of fault impacts, terms in the field of fault handling measures, historical records of fault events, fault impacts, and fault handling measures, and grid fault accident preconception plans;

[0009] Step 2: Based on the expert knowledge in the power grid field, the structured data and unstructured data of the distribution network fault handling plan are sorted and classified according to the actual situation of fault events, fault impacts, and fault handling measures; then, the terms in the field of fault events, the terms in the field of fault impacts, and the terms in the field of fault handling measures are extracted to obtain the fault event ontology, the fault impact ontology, and the fault handling measure ontology; the semi-structured data of the distribution network fault handling plan are formed by using the documents of the fault event ontology, the fault impact ontology, the fault handling measure ontology, and the lower-level objects of the fault event ontology, the fault impact ontology, and the fault handling measure ontology;

[0010] Step 3: Use the fault event ontology, fault impact ontology and fault handling measure ontology to establish the fault handling process ontology;

[0011] Step 4: Take the fault event ontology, fault impact ontology, fault handling measure ontology and fault handling process ontology as nodes in the distribution network fault plan knowledge graph; design the edge relationship in the distribution network fault plan knowledge graph according to the connection relationship between the fault event, fault impact, fault handling measure and fault handling process; the edge relationship is the ontology connection relationship between the fault event ontology, fault impact ontology, fault handling measure ontology and fault handling process ontology;

[0012] Step 5: Establish a distribution network fault plan entity recognition model based on the Bert+CRF model. The distribution network fault plan entity recognition model uses the semi-structured data of the distribution network fault handling plan as input data and the distribution network fault plan entity as output data; wherein the distribution network fault plan entity includes a fault event entity, a fault impact entity, a fault handling measure entity, and a fault handling process entity;

[0013] Step 6: According to the node and edge relationships in the distribution network fault plan knowledge graph, the distribution network fault plan entity is imported into the graph database, and the distribution network fault plan entity is displayed in a visual form of the distribution network fault plan knowledge graph.

[0014] In step 1, the structured data of the distribution network fault handling plan includes formatted data of distribution network dispatching centers at all levels, and the format of the formatted data includes extensible markup language and resource description framework;

[0015] The unstructured data of distribution network fault handling plans include structured documents of the dispatching fault plan manuals of distribution network dispatching centers at all levels and historical unstructured data of dispatchers handling faults.

[0016] In step 2, the fault event ontology includes terms in the field of fault events; the lower-level objects of the fault event ontology include: fault time number, fault device ontology, fault type ontology; wherein the attribute information of the fault device ontology includes: device number, device name, and device level;

[0017] The fault impact ontology includes terms in the fault impact field; the lower-level objects of the fault impact ontology include: fault impact number, fault impact user ontology, and fault impact type ontology; among them, the attribute information of the fault impact user ontology includes: user number, user name, and user level;

[0018] The fault handling measures ontology includes terms in the field of fault handling measures; the lower-level objects of the fault handling measures ontology include: fault handling measures number, local dispatching instruction ontology, and coordination handling action ontology; among them, the coordination handling action ontology includes: restore electricity instruction ontology, restore user electricity instruction ontology, and report completion status to local dispatch ontology.

[0019] The main body of the instruction for restoring electricity used and the main body of the instruction for restoring electricity used by users both include: a main body of the instruction for restoring electricity use operation; the main body of the instruction for restoring electricity use operation includes: a main body of the equipment, a main body of the switch status, and a main body of the function status.

[0020] In step 2, the attribute information includes the functions, performances and characteristics of the ontology, and one ontology corresponds to multiple different attribute information;

[0021] Attribute information includes: dispatch center number, dispatch authority level, plan number, fault number, fault type, fault level, user number, power supply unit, administrative region, electricity user number, industry, equipment number, equipment classification, power supply voltage, operating capacity, equipment model, geographical location, configuration information, audit status, overload time, fixed-cut protection status, protection switch status, voltage and current values, load capacity, activation time, and end time.

[0022] Step 3 includes:

[0023] Step 3.1: After receiving the fault information, the ground dispatcher sends the ground dispatcher instruction body to the distribution dispatcher;

[0024] Step 3.2: The dispatching department decomposes the command body issued by the local dispatching department into the power restoration command body and the user power restoration command body. The equipment body and the equipment sub-component body are respectively established for the equipment and the sub-components of the equipment involved in the power restoration operation command body. The equipment body and the equipment sub-component body are connected with the "include" relationship, that is, the equipment body contains the equipment sub-component body.

[0025] Step 3.3, after the fault is handled, the dispatcher will upload a report to the ground dispatcher on the completion status.

[0026] In step 4, the connection relationships between the fault event, fault impact, fault handling measures and fault handling process include: a "cause" relationship between the fault event and the fault impact, a "take" relationship between the fault event and the fault handling measures, a "take" relationship between the fault impact and the fault handling measures, and a "execution" relationship between the fault handling measures and the fault handling process;

[0027] The edge relationships of the distribution network fault plan knowledge graph include: logical relationships between entities, logical relationships between the attributes of entities, and logical relationships between the attributes of entities;

[0028] Among them, the logical relationships include: due to, resulting in, taking, including, executing, influencing, numerical adjustment, indicator quantity, and indicator status;

[0029] The attributes of edge relationships include: weight, start time, and end time.

[0030] Step 5 includes:

[0031] Step 5.1, manually annotating the semi-structured data of the distribution network fault handling plan to obtain the training corpus data of the distribution network fault handling plan;

[0032] Step 5.2, establish a distribution network fault plan entity recognition model based on the Bert (Bidirectional Encoder Representation from Transformers, sensor-based bidirectional encoding model) + CRF (Conditional Random Field Algorithm, conditional random field algorithm) model, and use the training corpus data of the distribution network fault handling plan to train the distribution network fault plan entity recognition model;

[0033] In step 5.3, the semi-structured data of the distribution network fault handling plan of the entity to be identified is vectorized and input into the trained distribution network fault plan entity recognition model, and the distribution network fault plan entity recognition model outputs the distribution network fault plan entity; the distribution network fault plan entity is saved in a CSV file.

[0034] Step 6 includes:

[0035] Step 6.1, design entity information for the fault event entity, fault impact entity, fault handling measure entity, and fault handling process entity; the entity information includes entity primary key, entity name, and entity attribute; wherein the primary key is in the form of ID, which is a string of character codes;

[0036] Step 6.2, based on the entity primary key, design the entity edge relationship for the fault event entity, fault impact entity, fault handling measure entity, and fault handling process entity. The entity edge relationship is from one entity primary key to another entity primary key;

[0037] Step 6.3, extracting ontology information from each node in the distribution network fault plan knowledge graph; the ontology information includes ontology primary key, ontology name, and ontology attribute;

[0038] In step 6.4, the entity primary key, entity name, and entity attributes are stored in the graph database in a one-to-one correspondence with the ontology primary key, ontology name, and ontology attributes. At the same time, the edge relationship of the entity is stored in the graph database in a one-to-one correspondence with the edge relationship in the distribution network fault plan knowledge graph, thereby realizing the import of the distribution network fault plan entity in the graph database.

[0039] On the other hand, the present invention proposes a distribution network fault plan knowledge graph design system based on a graph database, which is used to implement a distribution network fault plan knowledge graph design method based on a graph database.

[0040] The system includes: a memory, a processor, and a wireless communication module; the memory stores a computer program, and the processor calls the computer program to execute each step of the distribution network fault plan knowledge graph design method based on the graph database, and imports the distribution network fault plan entity into the graph database, and displays the distribution network fault plan entity in a visual form of the distribution network fault plan knowledge graph.

[0041] The beneficial effects of the present invention are as follows:

[0042] 1) The design method of the distribution network fault plan knowledge graph provided by the present invention is concise and efficient, which provides an accurate data basis for subsequent distribution network dispatching and operation, improves the intelligent level of fault handling in the field of distribution network dispatching, and lays the foundation for the automatic reasoning and generation of distribution network fault handling plans;

[0043] 2) When designing the knowledge graph of the distribution network fault plan, in addition to using the distribution network fault handling plan, the present invention also fully considers the distribution network fault events, the distribution network fault impact, and the relationship between the distribution network fault events, the distribution network fault impact and the distribution network fault handling plan, so that the knowledge graph of the distribution network fault plan is more complete and systematic;

[0044] 3) The distribution network fault plan knowledge graph proposed in the present invention combines the static knowledge graph and event knowledge graph of the distribution network equipment knowledge ontology, providing strong support for downstream query, reasoning application, etc. of the distribution network fault plan knowledge graph;

[0045] 4) The present invention uses a graph database to provide a storage method for the distribution network fault plan knowledge graph. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of the steps of the power grid regulation risk warning information knowledge graph design method of the present invention. DETAILED DESCRIPTION

[0047] The present application is further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present application.

[0048] Here is a storage example:

[0049] like Figure 1 On the one hand, the present invention proposes a distribution network fault plan knowledge graph design method based on a graph database, including:

[0050] Step 1, collect structured data and unstructured data of distribution network fault handling plan; wherein, the structured data and unstructured data of distribution network fault handling plan include: terms in the field of fault events, terms in the field of fault impacts, terms in the field of fault handling measures, historical records and documents of fault events, fault impacts and fault handling measures, and grid fault accident contingency plans.

[0051] In step 1, the structured data of the distribution network fault handling plan includes formatted data of distribution network dispatching centers at all levels, and the format of the formatted data includes but is not limited to: Extensible Markup Language (XML), Resource Description Framework (RDF), etc.

[0052] The unstructured data of distribution network fault handling plans include structured documents of the dispatching fault plan manuals of distribution network dispatching centers at all levels and historical unstructured data of dispatchers handling faults.

[0053] Step 2: Based on the expert knowledge in the power grid field, the structured data and unstructured data of the distribution network fault handling plan are sorted and classified according to the actual situation of fault events, fault impacts, and fault handling measures; then, the terms in the field of fault events, the terms in the field of fault impacts, and the terms in the field of fault handling measures are extracted to obtain the fault event ontology, the fault impact ontology, and the fault handling measure ontology; the semi-structured data of the distribution network fault handling plan are formed by using the documents of the fault event ontology, the fault impact ontology, the fault handling measure ontology, and the lower-level objects of the fault event ontology, the fault impact ontology, and the fault handling measure ontology;

[0054] In step 2, the fault event ontology includes terms in the field of fault events; the lower-level objects of the fault event ontology include: fault time number, fault device ontology, fault type ontology; wherein the attribute information of the fault device ontology includes: device number, device name, and device level;

[0055] The fault impact ontology includes terms in the fault impact field; the lower-level objects of the fault impact ontology include: fault impact number, fault impact user ontology, and fault impact type ontology; among them, the attribute information of the fault impact user ontology includes: user number, user name, and user level;

[0056] The fault handling measures ontology includes terms in the field of fault handling measures; the lower-level objects of the fault handling measures ontology include: fault handling measures number, local dispatching instruction ontology, and coordination handling action ontology; among them, the coordination handling action ontology includes: restore electricity instruction ontology, restore user electricity instruction ontology, and report completion status to local dispatch ontology.

[0057] The main body of the instruction for restoring electricity used and the main body of the instruction for restoring electricity used by users both include: a main body of the instruction for restoring electricity use operation; the main body of the instruction for restoring electricity use operation includes: a main body of the equipment, a main body of the switch status, and a main body of the function status.

[0058] In step 2, the attribute information includes the functions, performances and characteristics of the ontology, and one ontology corresponds to multiple different attribute information;

[0059] Attribute information includes: dispatch center number, dispatch authority level, plan number, fault number, fault type, fault level, user number, power supply unit, administrative region, electricity user number, industry, equipment number, equipment classification, power supply voltage, operating capacity, equipment model, geographical location, configuration information, audit status, overload time, fixed-cut protection status, protection switch status, voltage and current values, load capacity, activation time, and end time.

[0060] Step 3: Use the fault event ontology, fault impact ontology and fault handling measure ontology to establish the fault handling process ontology;

[0061] Step 3 includes:

[0062] Step 3.1: After receiving the fault information, the ground dispatcher sends the ground dispatcher instruction body to the distribution dispatcher;

[0063] Step 3.2: The dispatching department decomposes the command body issued by the local dispatching department into the power restoration command body and the user power restoration command body. The equipment body and the equipment sub-component body are respectively established for the equipment and the sub-components of the equipment involved in the power restoration operation command body. The equipment body and the equipment sub-component body are connected with the "include" relationship, that is, the equipment body contains the equipment sub-component body.

[0064] Step 3.3, after the fault is handled, the dispatcher will upload a report to the ground dispatcher on the completion status.

[0065] Step 4: Take the fault event ontology, fault impact ontology, fault handling measure ontology and fault handling process ontology as nodes in the distribution network fault plan knowledge graph; design the edge relationship in the distribution network fault plan knowledge graph according to the connection relationship between the fault event, fault impact, fault handling measure and fault handling process; the edge relationship is the ontology connection relationship between the fault event ontology, fault impact ontology, fault handling measure ontology and fault handling process ontology;

[0066] In step 4, the connection relationships between the fault event, fault impact, fault handling measures and fault handling process include but are not limited to: the "cause" relationship between the fault event and the fault impact, the "take" relationship between the fault event and the fault handling measures, the "take" relationship between the fault impact and the fault handling measures, and the "execution" relationship between the fault handling measures and the fault handling process;

[0067] It is worth noting that in this embodiment, the connection relationship between fault events, fault impacts, fault handling measures and fault handling processes is a non-restrictive and preferred choice. Technical personnel in the relevant field can determine the connection relationship between fault events, fault impacts, fault handling measures and fault handling processes based on historical engineering data.

[0068] The edge relationships of the distribution network fault plan knowledge graph include but are not limited to: the logical relationship between ontologies, the logical relationship between the attributes of an ontology and the attributes of an ontology, and the logical relationship between the attributes of an ontology and the attributes of an ontology;

[0069] Among them, logical relationships include but are not limited to: due to, resulting in, taking, including, executing, influencing, numerical adjustment, indicator quantity, indicator status;

[0070] The attributes of edge relationships include but are not limited to: weight, start time, and end time.

[0071] In this embodiment, not only the factual knowledge graph and the equipment knowledge graph are integrated in the top-down knowledge graph design, but also the "start time" attribute and the "end time" attribute are added to the graph edge relationship attributes, which can effectively improve the accuracy of knowledge reasoning and provide more powerful support for distribution network fault warning and distribution network fault plan generation.

[0072] Step 5: Establish a distribution network fault plan entity recognition model based on the Bert+CRF model. The distribution network fault plan entity recognition model uses the semi-structured data of the distribution network fault handling plan as input data and the distribution network fault plan entity as output data; wherein the distribution network fault plan entity includes a fault event entity, a fault impact entity, a fault handling measure entity, and a fault handling process entity;

[0073] The present invention establishes a distribution network fault plan entity recognition model based on the Bert+CRF model to perform entity recognition, which is better than the Bi-LSTM+CRF method that only uses fixed word vectors.

[0074] Step 5 includes:

[0075] Step 5.1, manually annotating the semi-structured data of the distribution network fault handling plan to obtain the training corpus data of the distribution network fault handling plan;

[0076] Step 5.2, establish a distribution network fault plan entity recognition model based on the Bert+CRF model, and use the training corpus data of the distribution network fault handling plan to train the distribution network fault plan entity recognition model;

[0077] In step 5.3, the semi-structured data of the distribution network fault handling plan of the entity to be identified is vectorized and input into the trained distribution network fault plan entity recognition model, and the distribution network fault plan entity recognition model outputs the distribution network fault plan entity; the distribution network fault plan entity is saved in a CSV file.

[0078] Step 6: According to the node and edge relationships in the distribution network fault plan knowledge graph, the distribution network fault plan entity is imported into the graph database, and the distribution network fault plan entity is displayed in a visual form of the distribution network fault plan knowledge graph.

[0079] Step 6 includes:

[0080] Step 6.1, design entity information for the fault event entity, fault impact entity, fault handling measure entity, and fault handling process entity; the entity information includes entity primary key, entity name, and entity attribute; wherein the primary key is in the form of ID, which is a string of character codes;

[0081] In this embodiment, entity information of the equipment entity is shown in Table 1, entity information of the substation entity is shown in Table 2, entity information of the fault event entity is shown in Table 3, and entity information of the fault impact entity is shown in Table 4.

[0082] Table 1 Entity information of device entity

[0083] ID Device Name Equipment number Voltage level Equip_1 Houjiatang Line 233 switch XXXXXXX 10KV

[0084] Table 2 Entity information of substation entity

[0085] ID Substation name Equipment number Voltage level station_1 Tangshan Incident XXXXXXX 35KV

[0086] Table 3 Entity information of fault event entity

[0087] ID name grade event_1 Phoenix lost power Level 1

[0088] Table 4 Entity information of fault-affected entities

[0089] ID name Impact on users event_1 Impact of power outage at all Phoenix substations Western Harbor

[0090] Step 6.2, based on the entity primary key, design the entity edge relationship for the fault event entity, fault impact entity, fault handling measure entity, and fault handling process entity. The entity edge relationship is from one entity primary key to another entity primary key;

[0091] In this embodiment, the edge relationship of the entity is shown in Tables 5 to 6:

[0092] Table 5 “Includes” relationships

[0093] ID FormId toId name Start time End time Belong_1 station_1 Equip_1 include 2010.9.1 2020.5

[0094] Table 6 “Leads to” relationships

[0095] ID FormId toId name result_1 event_1 event_1 lead to

[0096] Step 6.3, extracting ontology information from each node in the distribution network fault plan knowledge graph; the ontology information includes ontology primary key, ontology name, and ontology attribute;

[0097] In step 6.4, the entity primary key, entity name, and entity attributes are stored in the graph database in a one-to-one correspondence with the ontology primary key, ontology name, and ontology attributes. At the same time, the edge relationship of the entity is stored in the graph database in a one-to-one correspondence with the edge relationship in the distribution network fault plan knowledge graph, thereby realizing the import of the distribution network fault plan entity in the graph database.

[0098] On the other hand, the present invention proposes a distribution network fault plan knowledge graph design system based on a graph database, which is used to implement a distribution network fault plan knowledge graph design method based on a graph database.

[0099] The system includes: a memory, a processor, and a wireless communication module; the memory stores a computer program, and the processor calls the computer program to execute each step of the distribution network fault plan knowledge graph design method based on the graph database, and imports the distribution network fault plan entity into the graph database, and displays the distribution network fault plan entity in a visual form of the distribution network fault plan knowledge graph.

[0100] The applicant of the present invention has made a detailed explanation and description of the implementation examples of the present invention in conjunction with the drawings in the specification. However, those skilled in the art should understand that the above implementation examples are only preferred implementation schemes of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, but not to limit the scope of protection of the present invention. On the contrary, any improvements or modifications based on the inventive spirit of the present invention should fall within the scope of protection of the present invention.

Claims

1. A knowledge graph design method for distribution network fault emergency plan based on graph database, characterized in that: The method comprises: Step 1: Collect structured data and unstructured data of distribution network fault handling plan; wherein, the structured data and unstructured data of distribution network fault handling plan both include: terms in the field of fault events, terms in the field of fault impacts, terms in the field of fault handling measures, historical records of fault events, fault impacts, and fault handling measures, and grid fault accident preconception plans; Step 2: Based on the expert knowledge in the power grid field, the structured data and unstructured data of the distribution network fault handling plan are sorted and classified according to the actual situation of fault events, fault impacts, and fault handling measures; then, the terms in the field of fault events, the terms in the field of fault impacts, and the terms in the field of fault handling measures are extracted to obtain the fault event ontology, the fault impact ontology, and the fault handling measure ontology; the semi-structured data of the distribution network fault handling plan are formed by using the documents of the fault event ontology, the fault impact ontology, the fault handling measure ontology, and the lower-level objects of the fault event ontology, the fault impact ontology, and the fault handling measure ontology; Step 3: Use the fault event ontology, fault impact ontology and fault handling measure ontology to establish the fault handling process ontology; Step 4: Take the fault event ontology, fault impact ontology, fault handling measure ontology and fault handling process ontology as nodes in the distribution network fault plan knowledge graph; design the edge relationship in the distribution network fault plan knowledge graph according to the connection relationship between the fault event, fault impact, fault handling measure and fault handling process; the edge relationship is the ontology connection relationship between the fault event ontology, fault impact ontology, fault handling measure ontology and fault handling process ontology; Step 5: Establish a distribution network fault plan entity recognition model based on the Bert+CRF model. The distribution network fault plan entity recognition model uses the semi-structured data of the distribution network fault handling plan as input data and the distribution network fault plan entity as output data; wherein the distribution network fault plan entity includes a fault event entity, a fault impact entity, a fault handling measure entity, and a fault handling process entity; Step 6: According to the node and edge relationships in the distribution network fault plan knowledge graph, the distribution network fault plan entity is imported into the graph database, and the distribution network fault plan entity is displayed in a visual form of the distribution network fault plan knowledge graph.

2. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 1, characterized in that: In step 1, the structured data of the distribution network fault handling plan includes formatted data of distribution network dispatching centers at all levels, and the format of the formatted data includes extensible markup language and resource description framework; The unstructured data of distribution network fault handling plans include structured documents of the dispatching fault plan manuals of distribution network dispatching centers at all levels and historical unstructured data of dispatchers handling faults.

3. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 2, characterized in that: In step 2, the fault event ontology includes terms in the fault event domain; The lower-level objects of the fault event ontology include: fault time number, fault device ontology, fault type ontology; wherein the attribute information of the fault device ontology includes: device number, device name, device level; The fault impact ontology includes terms in the fault impact field; the lower-level objects of the fault impact ontology include: fault impact number, fault impact user ontology, and fault impact type ontology; among them, the attribute information of the fault impact user ontology includes: user number, user name, and user level; The fault handling measures ontology includes terms in the field of fault handling measures; the lower-level objects of the fault handling measures ontology include: fault handling measures number, local dispatching instruction ontology, and coordination handling action ontology; among them, the coordination handling action ontology includes: restore electricity instruction ontology, restore user electricity instruction ontology, and report completion status to local dispatch ontology.

4. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 3 is characterized in that: The main body of the instruction for restoring electricity used and the main body of the instruction for restoring electricity used by users both include: a main body of the instruction for restoring electricity use operation; the main body of the instruction for restoring electricity use operation includes: a main body of the equipment, a main body of the switch status, and a main body of the function status.

5. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 3 is characterized in that: In step 2, the attribute information includes the functions, performances and characteristics of the ontology, and one ontology corresponds to multiple different attribute information; Attribute information includes: dispatch center number, dispatch authority level, plan number, fault number, fault type, fault level, user number, power supply unit, administrative region, electricity user number, industry, equipment number, equipment classification, power supply voltage, operating capacity, equipment model, geographical location, configuration information, audit status, overload time, fixed-cut protection status, protection switch status, voltage and current values, load capacity, activation time, and end time.

6. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 1, characterized in that: Step 3 includes: Step 3.1: After receiving the fault information, the ground dispatcher sends the ground dispatcher instruction body to the distribution dispatcher; Step 3.2: The dispatching department decomposes the command body issued by the local dispatching department into the power restoration command body and the user power restoration command body. The equipment body and the equipment sub-component body are respectively established for the equipment and the sub-components of the equipment involved in the power restoration operation command body. The equipment body and the equipment sub-component body are connected with the "include" relationship, that is, the equipment body contains the equipment sub-component body. Step 3.3, after the fault is handled, the dispatcher will upload a report to the ground dispatcher on the completion status.

7. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 1, characterized in that: In step 4, the connection relationships between the fault event, fault impact, fault handling measures and fault handling process include: a "cause" relationship between the fault event and the fault impact, a "take" relationship between the fault event and the fault handling measures, a "take" relationship between the fault impact and the fault handling measures, and a "execution" relationship between the fault handling measures and the fault handling process; The edge relationships of the distribution network fault plan knowledge graph include: logical relationships between entities, logical relationships between the attributes of entities, and logical relationships between the attributes of entities; Among them, the logical relationships include: due to, resulting in, taking, including, executing, influencing, numerical adjustment, indicator quantity, and indicator status; The attributes of edge relationships include: weight, start time, and end time.

8. The method for designing a knowledge graph for a distribution network fault plan based on a graph database according to claim 1, characterized in that: Step 5 includes: Step 5.1, manually annotating the semi-structured data of the distribution network fault handling plan to obtain the training corpus data of the distribution network fault handling plan; Step 5.2, establish a distribution network fault plan entity recognition model based on the Bert+CRF model, and use the training corpus data of the distribution network fault handling plan to train the distribution network fault plan entity recognition model; In step 5.3, the semi-structured data of the distribution network fault handling plan of the entity to be identified is vectorized and input into the trained distribution network fault plan entity recognition model, and the distribution network fault plan entity recognition model outputs the distribution network fault plan entity; the distribution network fault plan entity is saved in a CSV file.

9. The method for designing a knowledge graph for a distribution network fault emergency plan based on a graph database according to claim 8, characterized in that: Step 6 includes: Step 6.1, design entity information for the fault event entity, fault impact entity, fault handling measure entity, and fault handling process entity; the entity information includes entity primary key, entity name, and entity attribute; wherein the primary key is in the form of ID, which is a string of character codes; Step 6.2, based on the entity primary key, design the entity edge relationship for the fault event entity, fault impact entity, fault handling measure entity, and fault handling process entity. The entity edge relationship is from one entity primary key to another entity primary key; Step 6.3, extracting ontology information from each node in the distribution network fault plan knowledge graph; the ontology information includes ontology primary key, ontology name, and ontology attribute; In step 6.4, the entity primary key, entity name, and entity attributes are stored in the graph database in a one-to-one correspondence with the ontology primary key, ontology name, and ontology attributes. At the same time, the edge relationship of the entity is stored in the graph database in a one-to-one correspondence with the edge relationship in the distribution network fault plan knowledge graph, thereby realizing the import of the distribution network fault plan entity in the graph database.

10. A distribution network fault emergency plan knowledge graph design system based on a graph database, used to implement the distribution network fault emergency plan knowledge graph design method based on a graph database as described in any one of claims 1 to 9, characterized in that: The system includes: a memory, a processor, and a wireless communication module; the memory stores a computer program, the processor calls the computer program to execute the steps described in any one of claims 1 to 9, and imports the distribution network fault plan entity into the graph database, and displays the distribution network fault plan entity in a visual form of a distribution network fault plan knowledge graph.

Images