A dispatching service treatment process automatic recommendation method based on a knowledge graph

Abstract

The application relates to a dispatching business treatment process automatic recommendation method based on a knowledge graph, and the technical characteristics are as follows: the method comprises the following steps: constructing a dispatching business knowledge graph based on a business process-business object-data source; combing the correlation between a dispatching business process and a business object, enumerating to form a customized form template; setting a trigger keyword to select a business scene, automatically adapting a process form template according to the business scene; and obtaining related information of the form template through the dispatching business knowledge graph according to an input keyword. The application is rationally designed, can realize the information transmission and feedback formization between a dispatching object and a dispatcher, forms a professional knowledge graph of the dispatching business after combing, realizes machine recommendation of a dispatching work standard template, automatically collects related key elements, greatly reduces the dispatching information transmission and communication time, reduces the work load of the dispatcher, reduces the human input, alleviates the human resource status, saves the power enterprise expenditure cost, and greatly reduces the cost.

Description

Technical Field

[0001] This invention belongs to the field of power grid dispatching technology, and in particular, it is an automatic recommendation method for dispatching business processing procedures based on knowledge graphs. Background Technology

[0002] For a long time, there has been a large amount of information transmission and feedback between power dispatching agencies and numerous dispatched entities. However, due to limitations in objective technology, low efficiency in information exchange is inevitable. With the workload of dispatching increasing exponentially, traditional dispatching work models are no longer suitable for the current needs of power dispatching management. Despite the surge in dispatching workload, the focus on the safe and stable operation of the power grid is disproportionate. This is because the previously simple and repetitive tasks have been concentrated, severely diverting the attention of dispatchers. Therefore, it is urgent to analyze and organize frequently used historical dispatching information and form data to construct a dispatching business knowledge graph. This will fully leverage the advantages of graph technology, relying on the correlation between business processes and business objects to automatically recommend business handling procedures and automatically collect information, thereby reducing the workload of dispatching personnel. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide an automatic recommendation method for scheduling business processing based on knowledge graphs, which solves the problems of heavy scheduling tasks, numerous and disorganized applications, and scattered information in the power dispatching system.

[0004] The present invention solves the existing technical problems by adopting the following technical solution:

[0005] An automatic recommendation method for scheduling business processing flow based on knowledge graph includes the following steps:

[0006] Step 1: Construct a scheduling business knowledge graph based on business process, business object, and data source;

[0007] Step 2: Analyze the relationships between scheduling business processes and business objects, and enumerate to form customized form templates;

[0008] Step 3: Set trigger keywords to select business scenarios, and automatically adapt the process form template according to the business scenarios;

[0009] Step 4: Obtain relevant information about the form template through the scheduling business knowledge graph based on the entered keywords.

[0010] Furthermore, the specific implementation method of step 1 includes the following steps:

[0011] (1) Starting from the business logic, analyze the business process to initially determine the scope of the data involved;

[0012] (2) By studying and analyzing the vocabulary used by dispatchers in their daily operations, the business objects in commonly used professional phrases are materialized.

[0013] (3) Label the business object entities with data, complete entity classification, attribute listing, relationship extraction and summary, and put the obtained entity graph into the scheduling business knowledge graph;

[0014] (4) Specify the source data retrieval method for each entity data stored in the business object.

[0015] Furthermore, the specific implementation method of step 2 includes the following steps:

[0016] (1) Collect key information elements that are mainly reported during the processing of various business operations, and enumerate them to form customized templates;

[0017] (2) Match business object entities according to the customized template, and form an association mapping relationship between business process, business object entity and entity attribute based on business, and store it in the scheduling business knowledge graph.

[0018] Furthermore, the specific implementation method of step 3 includes the following steps:

[0019] (1) Set trigger keywords for each type of business process for scenarios where business processes are initiated proactively;

[0020] (2) For each type of business process, define the information collection template, mark the key information columns in the template, and configure whether the corresponding fields are required or whether the attributes are automatically obtained.

[0021] Furthermore, the triggering keyword can be a single phrase, multiple phrases, or multiple related phrases.

[0022] Furthermore, the specific implementation method of step 4 includes the following steps:

[0023] (1) Enter the trigger keyword corresponding to the business process, initiate the corresponding business process, and accumulate the number of times the business process is initiated using the trigger keyword, which is recorded and stored as an attribute of the business process.

[0024] (2) Enter the business entity as a keyword. Based on the number of times the business processes involved in the business object entity are actively triggered, the top five business processes are recommended as extended scenarios for confirmation.

[0025] (3) Obtain customized form templates based on the triggered or confirmed business scenario;

[0026] (4) Based on the configuration items of the key elements corresponding to the template, obtain the data source access method in the scheduling business knowledge graph, and automatically retrieve, collect and fill in the information.

[0027] The advantages and positive effects of this invention are:

[0028] 1. This invention employs knowledge graph technology. First, it formalizes the information transmission and feedback between the scheduling object and the dispatcher, and then organizes it into a knowledge graph of the scheduling business. This enables the machine to recommend standard templates for scheduling work. Then, it automatically collects relevant key elements. Through the implementation of these two functions, the time for scheduling information transmission and communication is greatly reduced, the workload of dispatchers is reduced, the human resource input is reduced, the current human resource situation is alleviated, and the expenditure of power companies is saved, resulting in a significant reduction in costs.

[0029] 2. This invention differs from the process declarations used in previous scheduling operations. Previous process declarations had various forms and templates, all of which were called under specific system function modules. However, due to the large number of scheduling operations and the dispersed nature of the systems, it was difficult to enter the specific business process in a timely manner. This invention sets different trigger words for different business processes and can provide business association recommendations based on the business object, thus directly eliminating the time spent searching for the business process entry point and greatly simplifying the operation process. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the processing flow of the present invention. Detailed Implementation

[0031] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0032] The design concept of this invention is to construct a scheduling business knowledge graph library and structured templates for business process forms by sorting out the data sources of commonly used business process forms, and to establish mapping relationships between business processes, object entities, trigger words, and form templates. This can be applied to scenarios such as automatic integration, reporting, statistics, and analysis of power grid anomaly information and scheduling information, providing power grid operation decision analysis support for scheduling agencies.

[0033] Based on the above design concept, this invention provides an automatic recommendation method for scheduling service processing procedures based on knowledge graphs, such as... Figure 1 As shown, it includes the following steps:

[0034] Step 1: Construct a scheduling business knowledge graph based on business processes, business objects, and data sources.

[0035] The specific implementation process of this step is as follows:

[0036] (1) Starting from the business logic, the scope of data involved is initially determined by analyzing the business process. Taking the planned power outage maintenance application approval process as an example, the main data involved includes maintenance application, equipment ledger and operating status.

[0037] (2) By studying and analyzing the vocabulary used by dispatchers in the daily work of approving planned power outage maintenance applications, the business objects in commonly used professional phrases can be materialized to form entities such as maintenance application forms, substations, and circuit breakers.

[0038] (3) Label the business object entities with data, complete entity classification, attribute listing, relationship extraction and summary, and put the obtained entity graph into the scheduling business knowledge graph.

[0039] The maintenance request form entity is classified as a business process object, with main attributes including: request number, maintenance category, maintenance object, request start time, request completion time, grid risk, operation content, approved power outage start time, and approved work end time. The substation entity is classified as a grid container, with main attributes including: substation number, name, voltage level, and operating status. The circuit breaker entity is classified as grid equipment, with main attributes including: equipment number, name, voltage level, operating status, and switch status. The maintenance object in the maintenance request form includes grid containers, which contain grid equipment. Therefore, there are reference relationships between the maintenance request form and the substation, and between the substation and the circuit breaker. This information is stored in the dispatching business knowledge graph.

[0040] (4) Specify the source data retrieval method for each stored business object entity data. For example, the maintenance application comes from the planned power outage maintenance application function of the intelligent dispatch management system and is called by data interface. The substation information and circuit breaker information come from the commercial library backup of the energy management system and can be extracted by ETL.

[0041] Step 2: Analyze the scheduling business process and the relationship between business objects, and enumerate to form a customized form template.

[0042] The specific implementation process of this step is as follows:

[0043] (1) Collect key information elements that are mainly reported during the dispatching of various business processes and enumerate them to form customized templates. For example, in the process of approving a planned power outage maintenance application, the data obtained includes the application number, maintenance category, plant / station name, maintenance content, application start time, application completion time, approved power outage start time, and approved work end time.

[0044] (2) Match business object entities according to the customized template, and form an association mapping relationship of business process - business object entity - entity attribute based on business. In this example, the following association relationships are formed: "Planned power outage maintenance application approval - maintenance application form - application number / maintenance category / maintenance object / application start time / application completion time", "Planned power outage maintenance application approval - substation - plant number / voltage level / name", "Planned power outage maintenance application approval - circuit breaker - equipment number / voltage level / name", and stored one by one in the dispatch business knowledge graph.

[0045] Step 3: Set trigger keywords to select business scenarios, and automatically adapt the process form template according to the business scenarios.

[0046] The specific implementation process of this step is as follows:

[0047] (1) Set trigger keywords for the business process of approving planned power outage maintenance applications, which can be used to initiate business processes proactively. Trigger keywords can contain multiple phrases or related phrases, such as "maintenance application approval", "planned power outage approval", and "planned maintenance approval".

[0048] (2) For each type of business process, mark the information collection template and mark the key information columns in the template. Configure the corresponding fields to be required and automatically retrieved. The application number, maintenance category, plant name, maintenance content, application start time, application completion time, approved power outage start time, and approved work end time in the approval of the planned power outage maintenance application are all core fields of this business and need to be set as required fields. Among them, the application number, maintenance category, plant name, maintenance content, application start time, and application completion time can be configured to be automatically retrieved by the data source.

[0049] Step 4: Obtain relevant information about the form template through the scheduling business knowledge graph based on the entered keywords.

[0050] The specific implementation process of this step is as follows:

[0051] (1) Enter the trigger keyword corresponding to the business process, initiate the corresponding business process, and accumulate the number of times the business process is initiated using the trigger keyword. Record and store the data as an attribute of the business process. For example, enter the keyword "220810001 Repair Application Approval", trigger the planned power outage repair application approval based on "Repair Application Approval", and automatically retrieve the template.

[0052] (2) Enter the business entity as a keyword. Based on the number of times the business process involved in the business object entity is actively triggered, the top five business processes are recommended as extended scenarios for confirmation. For example, if you enter the name of the substation, the business involving the substation as the main body may include various types of business such as "approval of planned power outage maintenance application", "fault collaborative handling", "operation ticket generation", "substation alarm handling" and "substation primary wiring diagram maintenance". Based on the frequency of use of each type of business, that is, the number of times the business is actively triggered, the priority is ranked and the five types of business with the highest number of uses are recommended as alternatives.

[0053] (3) Obtain customized form templates based on the triggered or confirmed business scenario.

[0054] (4) Based on the configuration items of the key elements corresponding to the template, obtain the data source access method in the scheduling business knowledge graph, and automatically retrieve, collect and fill in the information. For example, in this case, “220810001” can be used as the maintenance application number to obtain information such as maintenance category, plant name, maintenance content, application start time, and application completion time from the maintenance application form data interface of the intelligent scheduling management system. After supplementing the required fields such as the approved start time of power outage and the approved end time of work, the process can be quickly initiated.

[0055] By following the steps above, we can achieve the function of automatically recommending scheduling business processing procedures based on knowledge graphs.

[0056] Any aspects not covered in this invention are applicable to existing technologies.

[0057] It should be emphasized that the embodiments described in this invention are illustrative rather than limiting. Therefore, this invention includes, but is not limited to, the embodiments described in the specific implementation. Any other implementations derived by those skilled in the art based on the technical solutions of this invention are also within the scope of protection of this invention.

Claims

1. A method for automatically recommending scheduling service processing procedures based on knowledge graphs, characterized in that: Includes the following steps: Step 1: Construct a scheduling business knowledge graph based on business process, business object, and data source; Step 2: Analyze the relationships between scheduling business processes and business objects, and enumerate to form customized form templates; Step 3: Set trigger keywords to select business scenarios, and automatically adapt the process form template according to the business scenarios; Step 4: Obtain relevant information about the form template through the scheduling business knowledge graph based on the entered keywords; The specific implementation method of step 1 includes the following steps: (1) Starting from the business logic, analyze the business process to initially determine the scope of the data involved; (2) By studying and analyzing the vocabulary used by dispatchers in their daily operations, the business objects in commonly used professional phrases are materialized. (3) Label the business object entities with data, complete entity classification, attribute listing, relationship extraction and summary, and put the obtained entity graph into the scheduling business knowledge graph; (4) Specify the source data retrieval method for each stored business object entity data; The specific implementation method of step 2 includes the following steps: (1) Collect key information elements reported during the processing of various business operations and enumerate them to form customized templates; (2) Match business object entities according to the customized template, and form an association mapping relationship between business process, business object entity and entity attribute based on business, and store it in the scheduling business knowledge graph; The specific implementation method of step 3 includes the following steps: (1) Set trigger keywords for each type of business process for scenarios where business processes are initiated proactively; (2) For each type of business process, define the information collection template, mark the key information columns in the template, and configure whether the corresponding fields are required or whether the attributes are automatically obtained; The specific implementation method of step 4 includes the following steps: (1) Enter the trigger keyword corresponding to the business process, initiate the corresponding business process, and accumulate the number of times the business process is initiated using the trigger keyword, which is recorded and stored as an attribute of the business process. (2) Enter the business entity as a keyword. Based on the number of times the business processes involved in the business object entity are actively triggered, the top five business processes are recommended as extended scenarios for confirmation. (3) Obtain customized form templates based on the triggered or confirmed business scenario; (4) Based on the configuration items of the key elements corresponding to the template, obtain the data source access method in the scheduling business knowledge graph, and automatically retrieve, collect and fill in the information.

2. The method for automatically recommending scheduling service processing procedures based on knowledge graphs according to claim 1, characterized in that: The triggering keyword can be a single phrase, multiple phrases, or multiple related phrases.

Images