Knowledge graph platform for power grid dispatching operation and maintenance

Abstract

The invention discloses a knowledge graph platform for power grid dispatching operation and maintenance, and belongs to the technical field of power dispatching, and the knowledge graph platform comprises a data management module which uploads power grid dispatching operation and maintenance data to the knowledge graph platform and carries out the data management of the power grid dispatching operation and maintenance data; the knowledge graph construction module is used for performing knowledge extraction on the power grid dispatching operation and maintenance data based on a knowledge extraction model and constructing a power dispatching knowledge graph; the application module is used for monitoring abnormal data in the power equipment based on the equipment operation state model and determining a fault reason of the power equipment based on the abnormal data; matching corresponding processing measures in the power dispatching knowledge graph according to the fault reasons; and the model management module pre-constructs a knowledge extraction model and an equipment operation state model. By applying the knowledge graph platform provided by the invention, scattered and heterogeneous knowledge in a power dispatching operation and maintenance scene can be effectively integrated and extracted, and the range and efficiency of knowledge acquisition are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power dispatching, and specifically belongs to a knowledge graph platform for power grid dispatching and operation and maintenance. Background Art

[0002] A large amount of knowledge on power dispatching automation operation and maintenance has been accumulated in the current power system, which contains a large amount of principle knowledge, the occurrence of similar faults in the past, and the corresponding solutions. However, most of this knowledge is still in static text, with scattered knowledge distribution, inconvenient access, time-consuming and laborious knowledge search, and too much reliance on personal experience, which poses a risk of misjudgment. As a result, the knowledge of power dispatching automation operation and maintenance is scattered in various reports and business systems, and the scope and efficiency of knowledge acquisition are low, and it cannot be fully and effectively utilized. Summary of the invention

[0003] The purpose of the present invention is to meet actual needs and provide a knowledge graph platform for power grid dispatching and operation and maintenance, which can effectively integrate and extract various scattered and heterogeneous knowledge in power dispatching and operation and maintenance scenarios, thereby improving the scope and efficiency of knowledge acquisition.

[0004] The present invention provides, for example, a knowledge graph platform for power grid dispatching and operation and maintenance, the knowledge graph platform comprising: a knowledge graph construction module, a data management module, a model management module and an application module;

[0005] The data management module is used to upload the power grid dispatching and operation and maintenance data in the power data source to the knowledge graph platform, and to perform data management on the power grid dispatching and operation and maintenance data;

[0006] The knowledge graph construction module is used to extract knowledge from the power grid dispatching and operation and maintenance data based on the knowledge extraction model to construct a power grid dispatching and operation and maintenance knowledge graph;

[0007] The application module is used to monitor abnormal data in the power equipment based on the equipment operation status model, determine the cause of the failure of the power equipment based on the abnormal data; match the corresponding treatment measures in the power grid dispatching and operation knowledge graph according to the cause of the failure;

[0008] The model management module is used to pre-build a knowledge extraction model and a device operation status model.

[0009] Preferably, the knowledge graph construction module is specifically used for:

[0010] Entity data, attribute data and relationship data are extracted from the power grid dispatching and operation and maintenance data, and mapped to the ontology structure of the pre-generated knowledge graph to construct a power grid dispatching and operation and maintenance knowledge graph; wherein the relationship data is the correspondence between entity data and entity data in the power grid dispatching and operation and maintenance data.

[0011] Preferably, the knowledge graph construction module further includes a knowledge fusion submodule;

[0012] The knowledge fusion submodule is used to compare the similarity between each extracted data and pre-set standard data, and determine similar data, so that business personnel can perform knowledge fusion based on the similar data.

[0013] Preferably, the knowledge fusion submodule compares the similarity between each extracted data and pre-set standard data based on a similarity algorithm or a clustering algorithm to determine similar data.

[0014] Preferably, the knowledge graph construction module further includes a knowledge updating submodule, and the knowledge updating submodule is specifically used for:

[0015] After receiving the updated power grid dispatching operation and maintenance data sent by the power dispatching business system, extracting knowledge from the updated power grid dispatching operation and maintenance data, and updating the power grid dispatching operation and maintenance knowledge graph based on the extracted updated data; or,

[0016] The power grid dispatching and operation and maintenance data in the power dispatching business system are accessed periodically through scheduled tasks. If the power grid dispatching and operation and maintenance data is updated, the updated power grid dispatching and operation and maintenance data are synchronized to the knowledge graph platform, and knowledge extraction is performed on the updated power grid dispatching and operation and maintenance data. The power grid dispatching and operation and maintenance knowledge graph is updated based on the extracted updated data.

[0017] Preferably, the power grid dispatching and operation and maintenance data includes at least one of the following information: power equipment information, component information in the power equipment, power equipment fault information, power equipment fault causes, and power equipment fault handling measures.

[0018] Preferably, the model management module constructs the equipment operation status model through the following steps:

[0019] Acquire the operating data of the power equipment collected by the monitoring device, and construct a data set of the equipment operating status based on the operating data;

[0020] The operating data in the data set is matched with the power equipment information in the power grid dispatching and operation and maintenance knowledge graph to build an equipment operating status model.

[0021] Preferably, the operation data includes at least one of the following data: a voltage parameter of the power equipment, a current parameter of the power equipment, and a temperature parameter of the power equipment;

[0022] The electric power equipment information includes at least one of the following information: an equipment model of the electric power equipment and an operating parameter of the electric power equipment.

[0023] Preferably, the application module is specifically used for:

[0024] Determine the fault cause of the electric power equipment based on the abnormal data, wherein the fault cause includes a diagnosable fault cause and an undiagnosable fault cause;

[0025] Filtering data for the diagnosable fault causes to obtain a first set of fault causes, and using the undiagnosable fault causes as a second set of fault causes;

[0026] The union of the first fault cause set and the second fault cause set is used as a fault cause data set, and corresponding processing measures are matched in the power grid dispatching and operation and maintenance knowledge graph through the fault cause data set.

[0027] Preferably, the knowledge graph platform also includes a graph management module for unified management and maintenance of each constructed power grid dispatching and operation knowledge graph.

[0028] Compared with the prior art, the advantages and positive effects of this application are:

[0029] The knowledge graph platform for power grid dispatching and operation and maintenance provided by the embodiment of the present invention is applied. The data management module in the platform can upload the scattered power grid dispatching and operation and maintenance data to the knowledge graph platform, so as to realize the unified management and efficient use of the power grid dispatching and operation and maintenance data. At the same time, the knowledge graph construction module can use the pre-trained knowledge extraction model to extract key knowledge from the power grid dispatching and operation and maintenance data, and construct the power grid dispatching and operation knowledge graph, thereby avoiding the occurrence of data islands and improving the utilization and sharing of power chain dispatching data.

[0030] In addition, the application module can monitor the abnormal data of power equipment based on the pre-trained equipment operation status model, infer the cause of the fault based on the abnormal data, and then match the corresponding treatment measures in the knowledge graph, which helps to improve the accuracy and response speed of fault diagnosis and reduce operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0032] Figure 1 A schematic diagram of the structure of a knowledge graph platform for power grid dispatching and operation and maintenance provided by an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of a process of intelligent fault diagnosis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present invention.

[0035] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the invention can be understood according to specific circumstances.

[0036] For the first embodiment, please refer to Figure 1 , Figure 1 A knowledge graph platform for power grid dispatching and operation and maintenance is provided in an embodiment of the present invention, and the platform includes: a knowledge graph construction module, a data management module, a model management module and an application module.

[0037] The data management module is used to upload the power grid dispatching and operation and maintenance data in the power data source to the knowledge graph platform, and to perform data management on the power grid dispatching and operation and maintenance data.

[0038] Specifically, the above-mentioned power data source can be power grid dispatching and operation and maintenance data, professional technical specification agreements or power business reports, etc.; the above-mentioned power grid dispatching and operation and maintenance data can include structured data and unstructured data, unstructured data can be document data such as fault handling reports, and structured data can be data obtained from relational databases or Excel data tables.

[0039] The knowledge graph construction module is used to extract knowledge from the power grid dispatching and operation and maintenance data based on a knowledge extraction model to construct a power grid dispatching and operation and maintenance knowledge graph.

[0040] The model management module is used to pre-build a knowledge extraction model and a device operation status model.

[0041] In this embodiment, entity data, attribute data and relationship data are extracted from the power grid dispatching and operation and maintenance data and mapped to the ontology structure of a pre-generated knowledge graph to construct a power grid dispatching and operation and maintenance knowledge graph; wherein the relationship data is the correspondence between entity data and entity data in the power grid dispatching and operation and maintenance data.

[0042] Specifically, before knowledge extraction, based on the needs of the power dispatching automation operation and maintenance knowledge graph and combined with the power data that can be obtained for building the knowledge graph, different entity data, relationship data, attribute data and the correspondence between the above data are abstracted to construct the ontology structure of the knowledge graph to facilitate the subsequent instantiation of data fusion.

[0043] In addition, different knowledge extraction methods can be adopted according to the different data structures in the power grid dispatching and operation and maintenance data.

[0044] For example, for structured data in a relational database or Excel data table, a power grid dispatching and operation knowledge graph is constructed based on the entity data, relational data, and attribute data in the structured data using entity mapping, attribute mapping, and relationship mapping, respectively.

[0045] Entity mapping is the process of associating entities in the knowledge graph with data tables in relational databases or Excel data tables. Specifically, each row of data in the data table represents an instance of an entity.

[0046] Attribute mapping is to correspond the attributes of the entity to the specific fields in the data table. That is, through attribute mapping, it can be determined that each attribute is reflected in the specific field in the data table.

[0047] Relationship mapping is to establish the association between entities. Through this process, the mutual connection between entities can be clarified, thus building a complete relationship network.

[0048] For unstructured data such as fault handling reports, the model management module can pre-train the initial annotation and extraction model based on the information provided by the business personnel, and then automatically extract entity data, relationship data, and attribute data based on the extraction model. The business personnel will then review and adjust the extraction results, and map them to the ontology structure of the knowledge graph. In addition, in order to maintain the accuracy of knowledge extraction, the manually reviewed and adjusted data can be fed back to the extraction model for optimization.

[0049] After building the power grid dispatching and operation and maintenance knowledge graph, business personnel use the application module to monitor abnormal data in the power equipment based on the equipment operation status model pre-trained by the model management module, and determine the cause of the power equipment failure based on the abnormal data; and match the corresponding treatment measures in the power grid dispatching and operation and maintenance knowledge graph based on the cause of the failure.

[0050] From the above, it can be seen that the data management module in the present invention can upload the scattered power grid dispatching and operation and maintenance data to the knowledge graph platform, realize the unified management and efficient use of power grid dispatching and operation and maintenance data, thereby avoiding the phenomenon of data islands and improving the utilization and sharing of power chain dispatching data. At the same time, the knowledge graph construction module can use the pre-trained knowledge extraction model to extract key knowledge from the power grid dispatching and operation and maintenance data, and construct the power grid dispatching and operation and maintenance knowledge graph. The application module monitors the abnormal data of the power equipment based on the equipment operation status model, and infers the cause of the fault based on the abnormal data, and then matches the corresponding treatment measures in the knowledge graph. This helps to improve the accuracy and response speed of fault diagnosis and reduce operation and maintenance costs.

[0051] Second embodiment, in the above first embodiment, many of the power grid dispatching and operation and maintenance data automatically extracted by the extraction model are repeated references, and entity disambiguation is required to remove repeated entities and relationships. To this end, the knowledge graph construction module also includes a knowledge fusion submodule, which is used to compare the similarity of each extracted data with the pre-set standard data, determine similar data, so that business personnel can perform knowledge fusion based on the similar data, thereby improving the quality of the power grid dispatching and operation and maintenance knowledge graph and reducing data redundancy.

[0052] In this embodiment, similarity comparison can be performed by a similarity algorithm or a clustering algorithm. For example, the similarity algorithm may include cosine similarity or Jaccard similarity, etc. With respect to the clustering algorithm, if the extracted entity data is small, clustering can be performed by manually setting a threshold. If the extracted entity data is large, clustering can be performed by a clustering algorithm such as DBSCAN.

[0053] In the third embodiment, based on the first embodiment, the knowledge graph construction module further includes a knowledge updating submodule, that is, the knowledge graph platform provided by the present invention has a graph iterative updating function. When the dispatching business data involved in the power grid dispatching and operation knowledge graph is updated, the knowledge updating submodule can automatically update the power grid dispatching and operation knowledge graph, thereby improving the timeliness and accuracy of the power grid dispatching and operation knowledge graph.

[0054] In this embodiment, the knowledge updating submodule can update the power grid dispatching and operation knowledge graph through system integration or data integration.

[0055] With regard to the system integration method, when the knowledge update submodule receives the updated power grid dispatching and operation and maintenance data sent by the power dispatching business system, it extracts knowledge from the updated power grid dispatching and operation and maintenance data, and updates the power grid dispatching and operation and maintenance knowledge graph based on the extracted updated data.

[0056] For example, when a business personnel maintains a fault work order in the power dispatching business system, the power dispatching business system triggers the graph update interface and sends the updated power grid dispatching operation and maintenance data of the entities and relationship information required for the graph update to the knowledge graph platform. When the knowledge update sub-module in the knowledge graph platform receives the updated power grid dispatching operation and maintenance data sent by the power dispatching business system, it begins to update the power grid dispatching operation and maintenance knowledge graph.

[0057] If we go a step further, we can recommend historically similar fault cases based on fault characteristics, and employees can select and automatically fill in relevant information fields, thereby reducing the workload of employees in filling out forms. The recommendation model algorithm can also be optimized based on the selected feedback information.

[0058] Regarding data integration, the knowledge update submodule periodically accesses the power grid dispatching and operation and maintenance data in the power dispatching business system through a scheduled task. If the power grid dispatching and operation and maintenance data is updated, the updated power grid dispatching and operation and maintenance data is synchronized to the knowledge graph platform, and knowledge is extracted from the updated power grid dispatching and operation and maintenance data. The power grid dispatching and operation and maintenance knowledge graph is updated based on the extracted updated data.

[0059] From the above, we can see that by regularly accessing the power grid dispatching operation and maintenance data in the power dispatching business system through scheduled tasks, the update of the data can be detected. Once the power grid dispatching operation and maintenance data is detected to be updated, the system will synchronize the updated data to the knowledge graph platform, and extract knowledge based on the updated data to further update the power grid dispatching operation and maintenance knowledge graph. This method ensures the regular update of the knowledge graph, thereby continuously optimizing and enriching the content of the knowledge graph.

[0060] In a fourth embodiment, the model management module constructs a device operation status model through the following steps:

[0061] Step 1: Obtain the operating data of the power equipment collected by the monitoring device, and construct a data set of the equipment operating status based on the operating data. The operating data at least includes the voltage parameters, current parameters and temperature parameters of the power equipment, which will be used as input to construct the data set of the equipment operating status.

[0062] The operating data in the data set is matched with the power equipment information in the power grid dispatching and operation and maintenance knowledge graph to build an equipment operating status model.

[0063] Step 2: Use an intelligent algorithm to match the operating data in the data set with the power equipment information in the power grid dispatching and operation knowledge graph to build an equipment operating status model. The power equipment information includes at least: the equipment model of the power equipment and the operating parameters of the power equipment.

[0064] For the fifth embodiment, please refer to Figure 2 , based on the fourth embodiment, after constructing the above-mentioned equipment operation status model, the application module can use machine learning, artificial intelligence and other technologies to analyze and diagnose abnormal situations and automatically identify possible causes of failures.

[0065] Among them, the above-mentioned fault causes include diagnosable fault causes and undiagnosable fault causes.

[0066] When judging the type of the above-mentioned fault cause, data filtering is performed on the diagnosable fault cause, and the fault cause obtained after filtering is used as the first fault cause set, and the filtered fault cause is judged by the rule engine in combination with the power grid operation section data, and the undiagnosable fault cause is used as the second fault cause set.

[0067] The union of the first fault cause set and the second fault cause set is used as a fault cause data set, and corresponding processing measures are matched in the power grid dispatching and operation and maintenance knowledge graph through the fault cause data set.

[0068] In the sixth embodiment, the knowledge graph platform of the present invention also includes a graph management module, which is used to uniformly manage and maintain the constructed knowledge graphs for power grid dispatching and operation. Specifically, the details of the graph can be viewed, it can be created and deleted, it supports switching of graphs linked to graph applications, and the versions of the graphs can be uniformly viewed, which is convenient for business personnel to manage all graphs in an overall manner.

[0069] The above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the protection scope of the present invention.

Claims

1. A knowledge graph platform for power grid dispatching and operation and maintenance, characterized in that: The knowledge graph platform includes: a knowledge graph construction module, a data management module, a model management module and an application module; The data management module is used to upload the power grid dispatching and operation and maintenance data in the power data source to the knowledge graph platform, and to perform data management on the power grid dispatching and operation and maintenance data; The knowledge graph construction module is used to extract knowledge from the power grid dispatching and operation and maintenance data based on the knowledge extraction model to construct a power grid dispatching and operation and maintenance knowledge graph; The application module is used to monitor abnormal data in the power equipment based on the equipment operation status model, determine the cause of the failure of the power equipment based on the abnormal data; match the corresponding treatment measures in the power grid dispatching and operation knowledge graph according to the cause of the failure; The model management module is used to pre-build a knowledge extraction model and a device operation status model.

2. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 1, characterized in that: The knowledge graph construction module is specifically used for: Entity data, attribute data and relationship data are extracted from the power grid dispatching and operation and maintenance data, and mapped to the ontology structure of the pre-generated knowledge graph to construct a power grid dispatching and operation and maintenance knowledge graph; wherein the relationship data is the correspondence between entity data and entity data in the power grid dispatching and operation and maintenance data.

3. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 2, characterized in that: The knowledge graph construction module also includes a knowledge fusion submodule; The knowledge fusion submodule is used to compare the similarity between each extracted data and pre-set standard data, and determine similar data, so that business personnel can perform knowledge fusion based on the similar data.

4. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 3 is characterized in that: The knowledge fusion submodule compares the extracted data with pre-set standard data based on a similarity algorithm or a clustering algorithm to determine similar data.

5. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 2, characterized in that: The knowledge graph construction module also includes a knowledge updating submodule, which is specifically used to: After receiving the updated power grid dispatching operation and maintenance data sent by the power dispatching business system, extracting knowledge from the updated power grid dispatching operation and maintenance data, and updating the power grid dispatching operation and maintenance knowledge graph based on the extracted updated data; or, The power grid dispatching and operation and maintenance data in the power dispatching business system are accessed periodically through scheduled tasks. If the power grid dispatching and operation and maintenance data is updated, the updated power grid dispatching and operation and maintenance data are synchronized to the knowledge graph platform, and knowledge extraction is performed on the updated power grid dispatching and operation and maintenance data. The power grid dispatching and operation and maintenance knowledge graph is updated based on the extracted updated data.

6. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 1, characterized in that: The power grid dispatching and operation and maintenance data includes at least one of the following information: power equipment information, component information in the power equipment, power equipment failure information, power equipment failure causes, and power equipment failure handling measures.

7. The knowledge graph platform for power grid dispatching and operation according to claim 1, characterized in that: The model management module constructs the equipment operation status model through the following steps: Acquire the operating data of the power equipment collected by the monitoring device, and construct a data set of the equipment operating status based on the operating data; The operating data in the data set is matched with the power equipment information in the power grid dispatching and operation and maintenance knowledge graph to build an equipment operating status model.

8. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 7, characterized in that: The operation data includes at least one of the following data: a voltage parameter of the power equipment, a current parameter of the power equipment, and a temperature parameter of the power equipment; The electric power equipment information includes at least one of the following information: an equipment model of the electric power equipment and an operating parameter of the electric power equipment.

9. The knowledge graph platform for power grid dispatching and operation according to claim 1, characterized in that: The application module is specifically used for: Determine the fault cause of the electric power equipment based on the abnormal data, wherein the fault cause includes a diagnosable fault cause and an undiagnosable fault cause; Filtering data for the diagnosable fault causes to obtain a first set of fault causes, and using the undiagnosable fault causes as a second set of fault causes; The union of the first fault cause set and the second fault cause set is used as a fault cause data set, and corresponding processing measures are matched in the power grid dispatching and operation and maintenance knowledge graph through the fault cause data set.

10. The knowledge graph platform for power grid dispatching and operation and maintenance according to claim 1, characterized in that: The knowledge graph platform also includes a graph management module for unified management and maintenance of the constructed power grid dispatching and operation knowledge graphs.

Images