CN111881313A - Intelligent ticketing model visual display system - Google Patents

Abstract

The invention relates to an intelligent ticketing model visual display system. 1. Adopting various elements such as nodes, guide lines and the like to construct a graphical ticket-forming logic model to form a vector diagram of the ticket-forming logic model: 2. through the characteristic classification of the application form, the classification display of the intelligent ticketing models of various application forms is formed, and the corresponding intelligent ticketing models are convenient to view; 3. and forming a hierarchical display of the ticket forming logic model, forming two layers of a process layer and an operation layer according to the characteristics of the application form ticket forming, and dividing the process layer and the operation layer into two corresponding layers during visual display to realize the hierarchical model display.

Description

Intelligent ticketing model visual display system

Technical Field

The invention relates to an intelligent ticketing model visual display system.

Background

The domestic dispatching instruction ticket and operation ticket system is almost synchronously developed with the computer information technology, and mainly undergoes the following development processes:

(1) original writing mode

Printing a form surface and manually writing;

(2) standardized format oriented and process control mode

The first stage is as follows: the manual input stage is carried out by utilizing calculation, table software is established, and the ticket simulation mode is mainly manual input, so that the difference of manually written fonts is avoided;

and a second stage: the process control is a stage with main characteristics, which mainly solves the problems of circulation statistics of operation tickets and the like in the processes of auditing, approving and executing the scheduling instruction tickets and the operation tickets;

and a third stage: in a simple intelligent entry stage, measures such as typical tickets, fixed phrases and the like are utilized to reduce entry time, so that the entry efficiency is improved, but many problems are brought, particularly, a typical ticket mode is not necessarily suitable for the operation mode of the current equipment.

(3) Graphic ticket-drawing mode

The first stage is as follows: single step emulation mode

The method inherits the flow standardization characteristic of the previous stage, utilizes typical tickets and certain fixed phrases to improve the ticket forming efficiency to a certain extent, and can select equipment on a power grid model wiring diagram at the stage to perform pulling and closing operations of the equipment or basic four-state conversion instructions. Simple instruction descriptions can be formed by clicking primitives, and are basically simple statements formed by combining single equipment and single operation types, but the problem of forming a scheduling instruction ticket with complex comprehensive instructions is not well solved.

And a second stage: multi-step simulation mode for task dismantling

The method aims to click a single equipment primitive on a graph, select an operation type and form multi-step operation when equipment reaches a certain state. The model already has considerable difficulty, and a system capable of completing the function is provided.

(4) Mode for simulating control instruction ticket and operation ticket in advance

Due to the complexity of scheduling instructions, simulation control is mainly performed on the pull-in operation of single equipment in the mode, the main realization principle is that equipment IDs and control type IDs are recorded on the instructions to identify the operation purpose to be achieved, when the expression significance of the ticket face characters is changed, the equipment IDs and the control type IDs cannot be changed along with the change, and unless the ticket face cannot be input to modify, great risk exists.

Therefore, the current instruction ticket and operation ticket systems basically stay in a standardized format and a flow control mode for specific applications, so that the intelligent scheduling instruction tickets and operation tickets are only improved in a local method and are not really intelligent, and therefore the intelligent scheduling instruction tickets and operation ticket systems are rarely used well.

(5) Expert knowledge base mode

Some system developers are aware of the complexity of the system and begin to explore by knowledge base theory, because developers lack knowledge about industry knowledge and develop the system by taking local knowledge as a target, when the system is actually applied, the dynamics, flexibility and difference in actual requirements are found to be too strong, and the knowledge base model is difficult to be comprehensively established. The complexity of domain knowledge makes knowledge base modeling tools difficult to express. Meanwhile, the method is difficult to meet the continuously changing application requirements, and a self-learning and self-accumulation technical means is lacked.

Therefore, the research of the intelligent operation ticket system is a complex problem, the problem is difficult to completely solve through a simple knowledge expert system or graphic billing, the ticket forming logic needs to be continuously self-perfected by introducing a machine learning means, and the application effect of the intelligent ticket forming system is further improved.

At present, with the continuous development of computer technology, the intelligent ticketing logic of distribution network instruction tickets/operation tickets is developed to complete the core rules of the distribution network instruction tickets/operation tickets in a mode of customizing a knowledge base, the mode can completely adapt to the change of a power grid structure and an operation mode, meanwhile, the customization mode is flexible, the customization of the rules can be realized without modifying programs, but the following problems still exist:

1. the logic rules cannot be visually displayed: although the existing core rule is formed by adopting a mode of customizing a knowledge base, the readability is much stronger than that of the core rule realized by a program code, but the rule logic still cannot be intuitively displayed;

2. there is no way to actively adapt to changes in the rules: the ticket-forming rule is not invariable, and the ticket-forming rule can be adjusted more or less at intervals, and the knowledge base can be maintained only by a manual mode to meet new requirements;

3. the establishment of the ticket-forming rule knowledge base is derived from the actual operation experience of a dispatcher and power supply station personnel, and the dispatcher and the power supply station personnel are difficult to completely express in a hundred percent manner in the operation experience description process, so that the establishment of the intelligent ticket-forming knowledge base is difficult to cover the logic rule of the actual operation in a hundred percent manner, and a technician is required to continuously track and perfect the knowledge base to achieve a better application effect.

4. The technical requirements on the knowledge base customizer are high, although the customization of the knowledge base is not completed by modifying codes, the requirements on the business knowledge, the computer knowledge and the modeling knowledge of the customizer are very high, the common dispatcher is difficult to complete the work, the difficulty of the later maintenance work is high, and the work can be completed by professional technicians.

Aiming at the problems, in order to further improve the accuracy and the practicability of intelligent ticket forming, the research of the subject automatically forms a ticket forming model on the basis of analyzing an instruction ticket/operation ticket sample or a historical ticket by adopting a machine learning mode based on a conditional random field without the need of high-level technicians to maintain and customize, can visually display the ticket forming model, visually display the ticket forming logic to a ticket issuing person, and visually check the accuracy of the ticket forming logic.

Disclosure of Invention

The invention aims to provide an intelligent visual showing system for a ticketing model, which does not need high-level technicians to maintain and customize and can visually show the ticketing model.

In order to achieve the purpose, the technical scheme of the invention is as follows: an intelligent ticketing model visual presentation system, comprising:

and (3) a service layer: performing characteristic analysis and mutual relation characteristic analysis on the operation ticket, the instruction ticket and the application form sample, forming an intelligent ticket forming model through machine learning, then analyzing the current power grid state and the characteristics of the ticket for the request of issuing the ticket, performing mode identification on the current power grid state and the characteristics of the ticket and the intelligent ticket forming model formed through machine learning, selecting the optimal ticket forming model, and performing visual display;

and (3) a data layer: providing historical/existing power grid models, historical/existing operation tickets, historical/existing instruction tickets and historical/existing application forms of the distribution network for a service layer;

maintenance tools: customizing or generating operation tickets, instruction tickets, application form samples, and customizing or generating ticketing models.

Compared with the prior art, the invention has the following beneficial effects:

1. maintenance and customization are not needed by high-level technicians, and the ticketing model can be visually displayed;

2. the ticket forming logic is visually displayed to the invoicing personnel, and the correctness of the ticket forming logic can be visually checked.

Drawings

FIG. 1 is a system software architecture of the present invention.

Fig. 2 is a schematic diagram of an interface scheme according to the present invention.

Detailed Description

The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.

The invention provides an intelligent ticketing model visual display system, which comprises:

and (3) a service layer: performing characteristic analysis and mutual relation characteristic analysis on the operation ticket, the instruction ticket and the application form sample, forming an intelligent ticket forming model through machine learning, then analyzing the current power grid state and the characteristics of the ticket for the request of issuing the ticket, performing mode identification on the current power grid state and the characteristics of the ticket and the intelligent ticket forming model formed through machine learning, selecting the optimal ticket forming model, and performing visual display;

and (3) a data layer: providing historical/existing power grid models, historical/existing operation tickets, historical/existing instruction tickets and historical/existing application forms of the distribution network for a service layer;

maintenance tools: customizing or generating operation tickets, instruction tickets, application form samples, and customizing or generating ticketing models.

The following is a specific implementation of the present invention.

1. Software architecture of the present invention

The distribution network intelligent instruction ticket research based on machine learning forms a corresponding ticketing model by adopting a machine learning method, the research of an algorithm aiming at the distribution network intelligent ticketing machine learning is a core, and by means of pattern recognition, visual display and the like, the correctness and the practicability of the distribution network intelligent ticketing are improved, and the software architecture is as shown in fig. 1:

(1) and (3) a service layer:

machine learning service: the automatic analysis and automatic learning functions of the samples are realized, and the service of an intelligent ticket-forming model is automatically formed;

feature analysis service: performing characteristic analysis service on the operation ticket, the instruction ticket and the application form;

the relational analysis service: carrying out relation characteristic analysis on instruction items of instruction tickets, safety measures of application forms, operation items of operation tickets and instruction items of instruction tickets

Pattern recognition: for the billing request, analyzing the characteristics of the current power grid state and safety measures, and selecting the best billing model after pattern recognition with the billing model formed by machine learning;

visualization presentation service: and the service is used for visually showing the ticketing model formed by machine learning.

(2) And (3) a data layer:

and (3) power grid model: power grid model of distribution network, power grid model of sharing existing distribution network instruction/operation order system

History instruction/operation ticket: historical instruction/operation ticket of existing distribution network instruction/operation ticket system should be shared

Historical application form: historical application form for sharing existing distribution network instruction/operation ticket system

(3) Maintenance tools:

sample customization tool: a tool for customizing or automatically generating an intelligent ticketing sample for machine learning;

the ticket-forming model customizing tool comprises: tool for customizing or automatically generating ticketing model

2. Hardware architecture

Because the distribution network dispatching instruction/operation ticket service function is completely realized in one area at present, the servers and the workstations required by the research of the project can share the existing server workstations in one area.

3. Interface scheme

The research of the subject is mainly a machine learning method for forming an intelligent ticket forming model of the distribution network instruction/operation ticket, so that the method is mainly used for performing an interface of related data with the existing distribution network instruction/operation ticket system, obtaining a corresponding power grid model, a historical application form and a historical instruction/operation ticket from the interface, and simultaneously sharing the intelligent ticket forming model formed by machine learning, as shown in fig. 2.

The main innovation points of the invention are as follows:

1. the intelligent ticketing model is visually displayed by adopting a vector diagram mode for the first time.

2. The layered display of the intelligent ticket forming is realized for the first time, and the intelligent ticket forming model is displayed more clearly.

The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.

Claims (1)

1. An intelligent ticketing model visual presentation system, comprising:

and (3) a service layer: performing characteristic analysis and mutual relation characteristic analysis on the operation ticket, the instruction ticket and the application form sample, forming an intelligent ticket forming model through machine learning, then analyzing the current power grid state and the characteristics of the ticket for the request of issuing the ticket, performing mode identification on the current power grid state and the characteristics of the ticket and the intelligent ticket forming model formed through machine learning, selecting the optimal ticket forming model, and performing visual display;

and (3) a data layer: providing historical/existing power grid models, historical/existing operation tickets, historical/existing instruction tickets and historical/existing application forms of the distribution network for a service layer;

maintenance tools: customizing or generating operation tickets, instruction tickets, application form samples, and customizing or generating ticketing models.

Images