CN115965222A - Automatic scheduling service handling process recommendation method based on knowledge graph - Google Patents

Abstract

The invention relates to a scheduling service handling process automatic recommendation method based on a knowledge graph, which is technically characterized by comprising the following steps: the method comprises the following steps: constructing a scheduling service knowledge graph based on a service process, a service object and a data source; the method comprises the steps of combing the relevance of scheduling business processes and business objects, and enumerating to form a customized form template; setting a trigger keyword selection service scene, and automatically adapting to a flow form template according to the service scene; and acquiring form template related information through a scheduling service knowledge graph according to the keyed-in keywords. The method is reasonable in design, information transmission and feedback tabulation between the scheduling object and the controller are realized, the professional knowledge map of the scheduling service is formed after the information transmission and feedback tabulation is carried out, the machine recommendation scheduling work standard template is realized, relevant key elements are automatically collected, the scheduling information transmission and communication time is greatly shortened, the workload of the controller is reduced, the human input is reduced, the current situation of human resources is relieved, the expenditure expense of power enterprises is saved, and the cost is greatly reduced.

Description

Automatic scheduling service handling process recommendation method based on knowledge graph

Technical Field

The invention belongs to the technical field of power grid dispatching, and particularly relates to a method for automatically recommending a dispatching service handling process based on a knowledge graph.

Background

Although there are many cases of information transmission and feedback between the power dispatching mechanism and many dispatching targets in the past, the problem of low information interaction efficiency is inevitable due to objective technical condition limitation. As the scheduling workload has multiplied, the traditional scheduling mode of operation no longer accommodates the needs of current power scheduling management. Under the condition of drastic increase of dispatching business volume, attention on safe and stable operation of a power grid is not in proportion, and the reason is that original simple and repeated work is concentrated together, and the energy of regulators is seriously dispersed, so that a dispatching business knowledge graph needs to be constructed by analyzing and sorting information and form data commonly used in historical dispatching, the technical advantages of the graph are fully exerted, automatic recommendation and automatic information collection of business disposal processes are carried out according to the relevance between businesses and business objects, and the working pressure of dispatching personnel is reduced.

Disclosure of Invention

The invention aims to overcome the defects of the prior art, provides a method for automatically recommending a scheduling service handling process based on a knowledge graph, and solves the problems of heavy scheduling work task, disordered system numerous applications and scattered information in an electric power scheduling system.

The invention solves the technical problems in the prior art by adopting the following technical scheme:

a method for automatically recommending a scheduling service handling process based on a knowledge graph comprises the following steps:

step 1, constructing a scheduling service knowledge graph based on a service process, a service object and a data source;

step 2, combing the relevance of the scheduling business process and the business object, and enumerating to form a customized form template;

step 3, setting a trigger keyword selection service scene, and automatically adapting to a flow form template according to the service scene;

and 4, acquiring form template related information through a scheduling service knowledge graph according to the key-in keywords.

Further, the specific implementation method of step 1 includes the following steps:

starting from a business logic, preliminarily judging a related data range by analyzing a business flow;

secondly, performing research and analysis on daily use vocabularies of a dispatcher in the dispatching business to materialize business objects in the commonly used professional phrases;

thirdly, performing data annotation on the business object entity, finishing entity classification, attribute listing and relationship extraction summary, and putting the obtained entity map into a scheduling business knowledge map;

and fourth, a source data calling mode for storing entity data of the business objects is noted one by one.

Further, the specific implementation method of step 2 includes the following steps:

the method comprises the steps of collecting key information elements mainly reported in the process of scheduling each business, and enumerating to form a customized template;

and secondly, matching business object entities according to a customized template, forming an association mapping relation of a business process, the business object entities and the entity attributes by taking the business as a reference, and storing the association mapping relation into a scheduling business knowledge map.

Further, the specific implementation method of step 3 includes the following steps:

the method includes the steps that a triggering keyword is set for each type of business process and used for initiatively initiating a scene of the business process;

and secondly, calibrating an information collection template for each type of business process, marking key information columns in the template, and configuring whether corresponding fields need to be filled and whether attributes are automatically acquired.

Further, the triggering keyword is a single phrase, a plurality of phrases or a plurality of relational phrases.

Further, the specific implementation method of step 4 includes the following steps:

the method comprises the steps of typing a trigger keyword corresponding to a business process, initiating the corresponding business process, performing accumulation statistics on the times of initiating the business process by using the trigger keyword, and recording and storing the times as the attribute of the business process;

secondly, typing in business entities as keywords, and recommending the first five business processes as expansion scenes for confirmation according to the active triggering times of the business processes related to the business object entities;

thirdly, acquiring a customized form template according to the triggered or confirmed service scene;

and acquiring a data source access mode in the scheduling service knowledge graph according to the configuration items of the key elements corresponding to the template, and automatically retrieving, collecting and filling information.

The invention has the advantages and positive effects that:

1. the invention adopts knowledge graph technology, firstly, information transmission and feedback tabulation between a scheduling object and a controller are realized, a professional knowledge graph of a scheduling service is formed after the information transmission and feedback tabulation is combed, a machine recommended scheduling work standard template is realized, then, relevant key elements are automatically collected, and through the realization of the two functions, the scheduling information transmission and communication time is greatly shortened, the workload of the controller is reduced, the human input is reduced, the current situation of human resources is relieved, the expenditure expense of an electric power enterprise is saved, and the cost is greatly reduced.

2. The invention is different from the flow declaration used in the prior scheduling service, the prior flow declaration has a plurality of forms and templates, the contents are all called under the appointed system function module, but the scheduling service is numerous and the system is dispersed, and the specific service flow is difficult to enter in the first time.

Drawings

FIG. 1 is a schematic process flow diagram of the present invention.

Detailed Description

The embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The design idea of the invention is as follows: the method is applied to scenes such as abnormal information of a power grid, automatic integration of scheduling information, reporting, statistics, analysis and the like, and provides power grid operation decision analysis support for a scheduling mechanism.

Based on the design concept, the invention provides a method for automatically recommending a scheduling service handling process based on a knowledge graph, which comprises the following steps as shown in fig. 1:

step 1, a scheduling service knowledge graph based on a service process, a service object and a data source is constructed.

The specific implementation process of the step is as follows:

(1) Starting from the business logic, the data range involved is preliminarily judged by analyzing the business process. Taking the planned blackout overhaul application replication process as an example, the mainly related data comprises overhaul application, an equipment ledger and an operation state.

(2) Through researching and analyzing daily use vocabularies of a dispatcher in planned power failure overhaul application and approval service, business objects in commonly used professional phrases are materialized, and entities such as overhaul application forms, stations, circuit breakers and the like can be formed.

(3) And carrying out data annotation on the service object entity to complete entity classification, attribute listing and relationship extraction summary, and putting the obtained entity map into a scheduling service knowledge map.

The maintenance request form entity is classified as a business process object, and the main attributes comprise: the method comprises the following steps of applying a serial number, a maintenance type, a maintenance object, applying start time, applying completion time, power grid risk, operation content, approved power failure starting time, approved work ending time and the like; the transformer substation entity is classified into a power grid container, and the main attributes comprise: station number, name, voltage level, running state and the like; the circuit breaker entity, categorised as the electric wire netting equipment, main attribute includes: equipment number, name, voltage level, running state and switch state; the overhaul object in the overhaul application form comprises a power grid container, and power grid equipment is contained in the power grid container, so that the quoting relations from the overhaul application form to the transformer substation and from the transformer substation to the circuit breaker exist, and the information is stored in the scheduling service knowledge map.

(4) And (3) noting the source end data calling mode for storing the entity data of the business object one by one, if the maintenance application is from the planned power failure maintenance application function of the intelligent scheduling management system, calling by adopting a data interface, and the transformer substation information and the breaker information are from commercial library backup of the energy management system, and extracting data by adopting an ETL mode.

And 2, combing the relevance of the scheduling business process and the business object, and enumerating to form a customized form template.

The specific implementation process of the step is as follows:

(1) And collecting key information elements mainly reported in the process of scheduling each business, and enumerating to form a customized template. For example, in the planned blackout overhaul application approval process, the acquired data includes an application number, an overhaul category, a station name, overhaul content, application start time, application completion time, approved blackout start time, approved work end time, and the like.

(2) And matching the business object entity according to the customized template, and forming an associated mapping relation of the business process, the business object entity and the entity attribute by taking the business as a reference. In the embodiment, incidence relations such as 'planning power outage overhaul application reply-overhaul application form-application number/overhaul category/overhaul object/application start time/application completion time', 'planning power outage overhaul application reply-transformer substation-station number/voltage grade/name', 'planning power outage overhaul application reply-breaker-equipment number/voltage grade/name' and the like are formed and are stored in the dispatching service knowledge map one by one.

And 3, setting a trigger keyword selection service scene, and automatically adapting to the flow form template according to the service scene.

The specific implementation process of the step is as follows:

(1) And setting a trigger keyword for the planned power failure overhaul application approval business process, wherein the trigger keyword is used for actively initiating a scene of the business process. The trigger keyword may include a plurality of phrases and include relational phrases, such as "overhaul application repeat", "planned blackout repeat", "planned overhaul repeat".

(2) The method comprises the steps of calibrating an information collection template for each type of service flow, marking key information columns in the template, configuring attributes of whether corresponding fields are required to be filled or not, whether the corresponding fields are automatically obtained or not, planning application numbers, maintenance types, station names, maintenance contents, application start time, application completion time, approved power failure start time and approved work end time in power failure maintenance application reply to be service core fields, and setting the fields as required items, wherein the fields of the application numbers, the maintenance types, the station names, the maintenance contents, the application start time, the application completion time and the like can be configured to be automatically obtained by data sources.

And 4, acquiring form template related information through a scheduling service knowledge graph according to the key-in keywords.

The specific implementation process of the step is as follows:

(1) And (3) inputting a trigger keyword corresponding to the business process, initiating the corresponding business process, performing accumulation statistics on the times of initiating the business process by using the trigger keyword, and recording and storing the accumulated statistics as the attribute of the business process, wherein the input keyword is ' 220810001 overhaul application reply ', and the template is automatically retrieved according to the ' overhaul application reply ' trigger plan power failure overhaul application reply '.

(2) The method comprises the steps of typing a business entity as a key word, sequencing according to the number of times of active triggering of business processes related to a business object entity, recommending the first five business processes as an extended scene for confirmation, for example, typing a name of a transformer substation, wherein the business related to the transformer substation as a main body may comprise various types of business such as 'planned power failure overhaul application approval', 'fault cooperative handling', 'operation ticket generation', 'transformer substation alarm handling', 'transformer substation primary wiring diagram maintenance', and the like, and according to the use frequency of various types of business, namely the number of times of active triggering of the type of business, priority is discharged, and the five types of business with the highest use frequency are recommended as alternatives.

(3) And acquiring a customized form template according to the triggered or confirmed service scene.

(4) According to configuration items of key elements corresponding to the template, data source access modes in the scheduling service knowledge graph are obtained, information is automatically retrieved, collected and reported, for example, "220810001" can be used as a maintenance application serial number in this example, information such as maintenance types, station names, maintenance contents, application start time, application completion time and the like can be obtained at a maintenance application list data interface of the intelligent scheduling management system, and a flow is quickly initiated after essential fields such as the power failure starting time of approval, the finishing time of approval and the like are complemented.

Through the steps, the automatic recommendation function of the scheduling service handling process based on the knowledge graph is realized.

Nothing in this specification is said to apply to the prior art.

It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.

Claims (6)

1. A method for automatically recommending a scheduling service handling process based on a knowledge graph is characterized by comprising the following steps: the method comprises the following steps:

step 1, constructing a scheduling service knowledge graph based on a service process, a service object and a data source;

step 2, combing the relevance of the scheduling business process and the business object, and enumerating to form a customized form template;

step 3, setting a trigger keyword selection service scene, and automatically adapting to a flow form template according to the service scene;

and 4, acquiring form template related information through a scheduling service knowledge graph according to the key-in keywords.

2. The method according to claim 1, wherein the method comprises: the specific implementation method of the step 1 comprises the following steps:

starting from a business logic, preliminarily judging a related data range by analyzing a business flow;

secondly, performing research and analysis on daily used vocabularies of a dispatcher in the dispatching service, and materializing service objects in the commonly used professional phrases;

thirdly, performing data annotation on the business object entity, completing entity classification, attribute listing and relationship extraction and summarization, and putting the obtained entity map into a scheduling business knowledge map;

and fourth, a source data calling mode for storing entity data of the business objects is noted one by one.

3. The method according to claim 1, wherein the method comprises: the specific implementation method of the step 2 comprises the following steps:

the method includes the steps of collecting key information elements mainly reported in the process of scheduling each business disposal, and enumerating to form a customized template;

and matching the business object entities according to the customized template, forming an association mapping relation of the business process, the business object entities and the entity attributes on the basis of the business, and storing the association mapping relation into a scheduling business knowledge map.

4. The method according to claim 1, wherein the method comprises: the specific implementation method of the step 3 comprises the following steps:

the method includes the steps that a triggering keyword is set for each type of business process and used for initiatively initiating a scene of the business process;

and secondly, calibrating an information collection template for each type of business process, marking key information columns in the template, and configuring whether corresponding fields need to be filled and whether attributes are automatically acquired.

5. The method according to claim 4, wherein the method comprises: the trigger key words are single phrases, multiple phrases or multiple relation phrases.

6. The method according to claim 1, wherein the method comprises: the specific implementation method of the step 4 comprises the following steps:

the method comprises the steps of typing a trigger keyword corresponding to a business process, initiating the corresponding business process, performing accumulation statistics on the times of initiating the business process by using the trigger keyword, and recording and storing the times as the attribute of the business process;

secondly, typing in business entities as keywords, sequencing according to the number of times that business processes related to business object entities are actively triggered, and recommending the first five business processes as expansion scenes for confirmation;

thirdly, acquiring a customized form template according to the triggered or confirmed service scene;

and acquiring a data source access mode in the scheduling service knowledge graph according to the configuration items of the key elements corresponding to the template, and automatically retrieving, collecting and filling information.

Images