CN-121998592-A - Special job ticket checking method and system based on large model

CN121998592ACN 121998592 ACN121998592 ACN 121998592ACN-121998592-A

Abstract

The application provides a special job ticket checking method and system based on a large model, and relates to the technical field of job ticket checking, wherein the method comprises the steps of extracting information from a job ticket and acquiring job ticket information; the method comprises the steps of determining checking content based on the job ticket information, checking the checking content through a job knowledge base to obtain a checking result, generating a correction proposal according to the checking result, and checking the correction. The application can solve the technical problems of poor checking quality caused by limited operation ticket checking coverage and low manual processing efficiency in the prior art, realizes the technical aim of constructing an intelligent operation ticket checking system capable of automatically executing efficient checking, achieves the technical effects of improving checking coverage rate and timeliness, guaranteeing stability and accuracy of checking quality, effectively reducing artificial omission risk, and improving the safety risk management and control level of special operation in a park.

Inventors

Jiang Chaolu
ZHANG WEN
ZHOU QIAN

Assignees

智园数字科技(山东)有限公司

Dates

Publication Date: 20260508
Application Date: 20260128

Claims (10)

1. The special job ticket checking method based on the large model is characterized by comprising the following steps of: Extracting information from the operation ticket to obtain operation ticket information; determining checking content based on the job ticket information; checking the checking content through an operation knowledge base to obtain a checking result; and generating a correction proposal according to the checking result, and checking the correction.
2. The special job ticket checking method based on the large model as set forth in claim 1, wherein the extracting of the information of the job ticket to obtain the information of the job ticket comprises: receiving an electronic job ticket and a paper job ticket; extracting information of the electronic job ticket based on the job ticket elements to obtain first job content; Performing picture conversion and text extraction on the paper job ticket, and performing information extraction based on the job ticket elements to obtain second job content; And formatting the first job content and the second job content to acquire the job ticket information.
3. The large model based special job ticket checking method as set forth in claim 2, wherein the job ticket element includes job base information including job name information, job type information, job time information, job location information, and job step information, job personnel information, job approval information, job risk assessment information, job safety measure information, and job end report information.
4. The special job ticket checking method based on the big model as set forth in claim 3, wherein determining the checking contents based on the job ticket information comprises: deconstructing the job ticket information based on the job type information, the job approval information, the job risk assessment information and the job safety measure information to obtain a checking element; And generating a checking prompt word according to the checking element, and extracting the checking content based on the checking prompt word.
5. The special job ticket checking method based on big model as set forth in claim 4, wherein generating a checking prompt word according to the checking element comprises: Performing verification center of gravity adjustment on the verification prompt word based on the job type information, the job approval information, the job risk assessment information and the job safety measure information; according to the adjusted center of gravity, the verification prompt word is used as a central prompt word, and the extension of the verification prompt word is executed to obtain an auxiliary verification prompt word; and adding the auxiliary checking prompt word to the checking prompt word.
6. The special job ticket checking method based on big model as set forth in claim 4, wherein before checking the checking contents through a job knowledge base to obtain a checking result, comprising: Introducing a working knowledge base, and carrying out structural storage on the working knowledge base to obtain metadata tags; Taking the metadata tag as a query index, traversing the check prompt words in the metadata tag for query, and obtaining retrieval content; and carrying out compliance evaluation on the checking content according to the searching content to obtain an initial checking result.
7. The special job ticket checking method based on big model as set forth in claim 6, wherein checking the checking contents through a job knowledge base to obtain a checking result comprises: Performing membership evaluation of standard execution dimension and safety compliance dimension of the initial checking result to generate a checking membership; And if the check membership of the standard execution dimension is greater than or equal to the check membership of the safety compliance dimension, performing discipline adjustment on the initial check result to obtain the check result.
8. The big model based special job ticket checking method as set forth in claim 1, wherein generating a correction proposal according to the checking result comprises: Performing operation problem marking based on the checking result that the items are not passed to obtain checking problems; and generating a rectification proposal based on the checking problem, wherein the rectification proposal comprises rectification measures, rectification targets and rectification time limits.
9. The big model based special job ticket checking method as set forth in claim 1, wherein after performing the check correction, comprising: Generating a checking record based on the solved correction proposal, and carrying out statistics to generate a statistics report; Checking key marks of the job ticket information according to the statistical report; And preferentially extracting the checking content of the job ticket to be received with the checking emphasis.
10. A big model based special job ticket checking system characterized by the steps for implementing the big model based special job ticket checking method according to any one of claims 1 to 9, comprising: The job ticket information acquisition module is used for extracting information of the job ticket and acquiring job ticket information; The checking content determining module is used for determining checking content based on the job ticket information; the checking result obtaining module is used for checking the checking content through the operation knowledge base to obtain a checking result; and the checking and modifying module is used for generating modifying and modifying advice according to the checking result and checking and modifying.

Description

Special job ticket checking method and system based on large model Technical Field The application relates to the technical field of job ticket checking, in particular to a special job ticket checking method and system based on a large model. Background With the continuous improvement of the operation safety control requirements of electric power, chemical industry, coal mine and various high-risk industrial production scenes, the operation ticket is used as an important management carrier for standardizing operation behaviors, realizing safety measures and defining operation responsibilities, and is widely applied to various operation scenes such as fire operation, overhead operation, limited space operation, temporary electricity utilization operation and the like. Currently, the existing mainstream checking method mainly relies on a chemical safety expert to review and judge the integrity, compliance and risk control measures of the content of a job ticket in a manual spot check mode. However, the manually dominated audit mode has a number of inherent drawbacks. Firstly, in the coverage area, limited by manpower and time cost, the inspection can be generally only sampled according to proportion or risk level, and full coverage can not be realized, so that a large number of unchecked job tickets are in a supervision blind area, potential risks of the job tickets can not be effectively identified and controlled, and a gap exists in the whole security line. In the aspect of checking quality, the result is seriously dependent on the experience level and knowledge reserve of expert individuals, and when different checking staff and even the same staff are in different energy states, the judgment standard and the detail degree of the expert individuals can possibly fluctuate, so that the quality of a checking conclusion is unstable and the subjectivity is strong. Thirdly, in terms of operation efficiency and stability, aiming at massive and differently formatted operation tickets, manual review is required to check the relation among texts, data and logic item by item, time and effort are consumed, efficiency is low, negligence, omission or misjudgment easily occurs in a fatigue state, and continuous and stable high-quality output is difficult to ensure. Finally, the method is difficult to systematically integrate and rapidly apply the continuously updated rule standard and expert experience knowledge base, and knowledge iteration lags. In summary, in the prior art, because the checking coverage is limited, the individual expert experience is excessively depended, the manual processing efficiency is low and the state is unstable, a large number of special operation tickets are in an uncontrolled risk state, the checking quality is uneven, and the supervision response is lagged, the controllability of the whole special operation safety risk of the chemical industry park and the continuous improvement of the management efficiency are further affected. Disclosure of Invention The application aims to provide a special job ticket checking method and system based on a large model, which are used for solving the technical problems that a large number of special job tickets are in an uncontrolled risk state, uneven in checking quality and lag in supervision response due to limited checking coverage, excessive dependence on individual expert experience, low manual processing efficiency and unstable state in the prior art, and further influencing the controllability of the whole special job security risk of a chemical industry park and the continuous improvement of management efficiency. In view of the above problems, the present application provides a special job ticket checking method and system based on a large model. In a first aspect, the application provides a special job ticket checking method based on a large model, which is realized by a special job ticket checking system based on the large model, and comprises the steps of extracting information of a job ticket and acquiring the information of the job ticket; the method comprises the steps of determining checking content based on the job ticket information, checking the checking content through a job knowledge base to obtain a checking result, generating a correction proposal according to the checking result, and checking the correction. Preferably, the special job ticket checking method based on the large model further comprises the steps of receiving an electronic job ticket and a paper job ticket, extracting information of the electronic job ticket based on job ticket elements to obtain first job content, converting pictures and extracting texts of the paper job ticket, extracting information of the paper job ticket elements to obtain second job content, formatting the first job content and the second job content, and obtaining job ticket information. Preferably, the special job ticket checking method based on the large model further comprises the steps tha