Automatic identification system and method for compliance of two electric power tickets

Abstract

The invention relates to the technical field of power safety production management, in particular to a power two-ticket compliance automatic identification system and method. The power two-ticket compliance automatic identification system comprises a rule base construction module which is used for pre-storing power industry two-ticket compliance rules; the image preprocessing module is used for receiving and preprocessing two-ticket images uploaded by a user; the character extraction module is used for positioning and identifying character contents of key information fields of the two tickets; the compliance analysis module is used for comparing the key information with compliance rules pre-stored in the rule base construction module and outputting a compliance judgment result; and the report generation module is used for generating a structured compliance review report. The ticket face information is automatically extracted through the optical character recognition technology, automatic logic comparison is conducted through a rule engine, the traditional checking mode depending on reviews is thoroughly changed, the reviewing efficiency is improved, and the objectivity and consistency of reviewing results are fundamentally guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power safety production management, in particular to an electric power two-ticket compliance automatic identification system and method. BACKGROUND

[0002] The "two-ticket three-system" in the electric power industry is a core management system to ensure the safety of electric power production, and the "two-ticket" refers to the work ticket and the operation ticket, which are the key credentials to standardize the on-site operation process and ensure the safety of the operation. The work ticket is used to clarify the work content, personnel, time and safety measures of maintenance, testing, installation and other operations; the operation ticket is used to standardize the switching operation process of electrical equipment to prevent misoperation and cause accidents to people, equipment or power grid. At present, the compliance review of the "two-ticket" in electric power enterprises mainly relies on manual methods. The review personnel need to check the completeness, logicality and compliance of the information on the ticket according to the "Electric Power Safety Work Procedures" and the internal management rules of the enterprise, including whether the personnel qualifications are matched, whether the time logic is reasonable, whether the tickets are associated, and whether the operation steps conform to the standard sequence, etc.

[0003] However, the manual review method has obvious defects: low review efficiency: facing a large number of tickets, manual review of each ticket is time-consuming and labor-intensive, which is difficult to meet the actual demand of fast operation rhythm and large number of tickets in electric power field. The review result is highly subjective: the quality of the review is highly dependent on the professional experience and responsibility of the review personnel, and it is easy to cause judgment deviation or omission due to different personnel levels. It is difficult to ensure the consistency of the review: different review personnel have different understandings and executions of the rules, which leads to non-uniform review standards and affects the overall safety management level. It is difficult to cope with complex logic verification: for complex compliance logic such as multi-ticket association and operation sequence, manual review is prone to errors and it is difficult to systematically and standardize the comprehensive inspection.

[0004] Therefore, the existing technology lacks a technical means that can automatically and systematically identify the high-efficiency, objective and accurate compliance of the "two-ticket", and an intelligent review system combining image recognition and rule engine is urgently needed to improve the management efficiency and reliability of electric power safety production. SUMMARY

[0005] Therefore, the technical problem to be solved by the present application is to overcome the problems of low review efficiency, poor review consistency and inability to cope with complex logic verification in the prior art.

[0006] To solve the above technical problems, the present application provides an electric power two-ticket compliance automatic identification system, comprising: a rule library construction module for pre-storing electric power industry two-ticket compliance rules; an image preprocessing module for receiving and preprocessing the two-ticket images uploaded by the user; a text extraction module connected with the image preprocessing module, configured to locate and identify the text content of the key information fields of the two tickets from the preprocessed image; a compliance analysis module connected with the rule library construction module and the text extraction module, configured to compare the key information extracted by the text extraction module with the pre-stored compliance rules of the rule library construction module, and output a compliance judgment result; a report generation module connected with the compliance analysis module, configured to generate a structured compliance review report according to the compliance judgment result.

[0007] Preferably, the power industry two-ticket compliance rules include: a personnel qualification rule for judging whether the workers on the work ticket have corresponding qualifications; a time logic rule for judging whether the planned time logic on the work ticket is reasonable; a ticket association rule for judging whether the association relationship among the work ticket, the operation ticket and the recovery ticket is complete; an operation sequence rule for judging whether the operation step sequence in the operation ticket conforms to the predetermined standard sequence.

[0008] Preferably, the text extraction module adopts an optical character recognition technology based on deep learning and is optimized and trained for the table format and handwritten font of the power industry two tickets.

[0009] Preferably, the text extraction module adopts an optical character recognition technology based on deep learning and is optimized and trained for the table format and handwritten font of the power industry two tickets. a two-ticket template based on pre-learning for locating the key information area; identifying and extracting the text content in the key information area, and the key information includes work content, workers, planned work time and operation steps.

[0010] Preferably, the structured compliance review report includes: review summary information including the total number of tickets, the number of compliance items and the number of violation items; a violation detail list detailing the violation tickets, violation contents and triggered rules; and a compliance item summary list.

[0011] The application also provides a power two-ticket compliance automatic identification method, which includes: constructing a rule library to pre-store power industry two-ticket compliance rules; receiving two-ticket images uploaded by a user and preprocessing the images; locating the key information area and identifying the text of the preprocessed image to extract key information content; The extracted key information is automatically compared with pre-stored compliance rules, and compliance logic judgments are executed. Based on the assessment results, a structured compliance review report is generated and output.

[0012] Preferably, the rule base construction, which pre-stores compliance rules for the power industry's two-ticket system, includes: The name of the person in charge of the work extracted from the work order is compared with the pre-stored personnel permission knowledge base to verify whether they have the corresponding qualifications. Compare the planned start time and planned completion time extracted from the work order to verify whether the time sequence logic is reasonable; Verify whether there are valid work orders, action orders, and recovery orders for the same work content at the same time; The sequence of operation steps extracted from the operation ticket is compared sequentially with the standard electrical operation sequence pre-stored in the rule base.

[0013] Preferably, the step of locating key information regions and recognizing text in the preprocessed image to extract key information content includes: Text recognition is performed using a deep learning-based optical character recognition engine; Based on the pre-learned fixed template structure and field layout of two tickets, key information areas in the work content, staff, planned work time and operation steps are located by image recognition technology. Identify and extract printed or handwritten text within the located area, and convert it into structured text data.

[0014] Preferably, the step of automatically comparing the extracted key information content with pre-stored compliance rules and performing compliance logic judgment includes: When the comparison results show that the information content does not conform to the preset rules, it is judged as a violation; For each violation, record the ticket information, the specific description of the violation, and the rule number that was triggered; Generate and store detailed judgment logs containing the reasons for violations.

[0015] Preferably, the generation and output of a structured compliance review report based on the judgment result includes: Summarize all compliance assessment results to generate a structured review report; the report includes a review summary, a detailed list of violations, and a summary of compliance items. The generated review report can be provided to users for download through the system interface, or automatically pushed to an external safety production management system for further processing through a preset system interface.

[0016] The technical solution of the present invention has the following advantages compared with the prior art: This invention discloses an automatic identification system and method for compliance of power-related invoices and related documents. It automatically extracts information from the invoices using optical character recognition technology and performs automated logical comparison using a rule engine, completely changing the traditional model that relies on manual inspection. The system can complete the review of batches of invoices and related documents within minutes or even seconds, greatly freeing up manpower and increasing review efficiency by tens of times, meeting the high requirements of modern power production for rapid response to safety management. By solidifying compliance judgment standards into executable computer rules, it eliminates judgment biases and omissions caused by factors such as personal experience, professional level, fatigue, and subjective emotions in manual review. For the same invoice, the system can provide a unique and definitive review conclusion, effectively solving the problems of inconsistent standards among different reviewers and fluctuations in standards among the same reviewer, greatly enhancing the standardization and authority of the safety management system. Attached Figure Description

[0017] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein: Figure 1 This is a structural diagram of an automatic identification system for compliance of two types of electricity tickets provided by the present invention; Figure 2 This is a flowchart of an automatic identification method for compliance of two electricity tickets provided by the present invention. Detailed Implementation

[0018] The core of this invention is to provide an automatic identification system for the compliance of two power invoices. It automatically extracts invoice information through optical character recognition technology and uses a rule engine for automated logical comparison, which completely changes the traditional mode of relying on reviewers for verification, improves review efficiency, and fundamentally guarantees the objectivity and consistency of review results.

[0019] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please refer to Figure 1. Figure 1 The logical relationship diagram of the automatic identification system for compliance of two-ticket electricity use provided by the present invention is as follows: The specific operation steps are as follows: The rule base construction module is used to pre-store compliance rules for the power industry's two-ticket system. Among them, the compliance rules for the two-ticket system in the power industry include: Personnel qualification rules are used to determine whether the staff on the work permit have the corresponding qualifications; Time logic rules are used to determine whether the planned time logic on the work order is reasonable; Ticket association rules are used to determine whether the association relationship between work tickets, operation tickets, and recovery tickets is complete. Operation sequence rules are used to determine whether the order of operation steps in an operation ticket conforms to a predetermined standard sequence.

[0021] The image preprocessing module is used to receive and preprocess the two images uploaded by the user; The text extraction module, connected to the image preprocessing module, is used to locate and identify the text content of the key information fields of the two tickets from the preprocessed image; Specifically, the text extraction module uses deep learning-based optical character recognition technology and has been optimized and trained for the table format and handwritten font of the two invoices in the power industry. Based on pre-learned two-ticket templates, key information areas are located; Identify and extract the text content within the key information area, whereby the key information includes: work content, staff, planned work time, and operating steps.

[0022] The compliance analysis module is connected to the rule base construction module and the text extraction module. It is used to compare the key information extracted by the text extraction module with the compliance rules pre-stored in the rule base construction module and output the compliance judgment result. The report generation module, connected to the compliance analysis module, is used to generate a structured compliance review report based on the compliance judgment results.

[0023] Specifically, review summary information, including the total number of votes, the number of compliant items, and the number of non-compliant items; a detailed list of non-compliant items, listing the non-compliant tickets, the content of the non-compliant items, and the rules that triggered them; and a summary list of compliant items.

[0024] This embodiment provides an automatic identification system for the compliance of power invoices and related documents. It automatically extracts information from the invoices using optical character recognition technology and performs automated logical comparison using a rule engine, completely changing the traditional model that relies on manual inspection. The system can complete the review of batches of invoices and related documents within minutes or even seconds, greatly freeing up manpower and increasing review efficiency by tens of times, meeting the high requirements of modern power production for rapid response to safety management. By solidifying compliance judgment standards into executable computer rules, it eliminates judgment biases and omissions caused by factors such as personal experience, professional level, fatigue, and subjective emotions in manual review. For the same invoice, the system can provide a unique and definitive review conclusion, effectively solving the problems of inconsistent standards among different reviewers and fluctuations in standards set by the same reviewer, greatly enhancing the standardization and authority of the safety management system.

[0025] like Figure 2 As shown, Figure 2 The present invention provides an automatic identification method for the compliance of two electricity tickets, as detailed below: Step S201: Build a rule base and pre-store the compliance rules for the two types of invoices in the power industry; Specifically, the name of the person in charge of the work extracted from the work order is compared with the pre-stored personnel permission knowledge base to verify whether they have the corresponding qualifications. Compare the planned start time and planned completion time extracted from the work order to verify whether the time sequence logic is reasonable; Verify whether there are valid work orders, action orders, and recovery orders for the same work content at the same time; The sequence of operation steps extracted from the operation ticket is compared sequentially with the standard electrical operation sequence pre-stored in the rule base.

[0026] In one embodiment, the system backend receives and parses power plant personnel permission data imported in Excel spreadsheet format. This spreadsheet contains at least the fields of "Name," "Employee ID," "Company / Department," and the crucial "Three Types of Personnel Qualifications." The system stores this structured data in an internal database, forming a personnel permission knowledge base. Based on the "Electric Power Safety Work Regulations" and the power plant's internal rules, the system has pre-defined the following executable logical rules in the rule base: Rule R1 (Personnel Qualification Rule): The "Work Manager" on the "Work Order" must exist in the personnel permission knowledge base, and its "Three Types of Personnel Qualifications" field must include the "Work Manager" qualification.

[0027] Rule R2 (Time Logic Rule): The "Planned Start Time" on the "Work Order" must be earlier than the "Planned Completion Time".

[0028] Rule R3 (ticket association rule): An "operation ticket" (measure ticket) must correspond to at least one "work ticket" and one "recovery ticket", and the "work content" field of the operation ticket must be related to the "work content" of the work ticket.

[0029] Rule R4 (Operation Sequence Rule): The operation steps in the "Operation Ticket" must be executed in a predetermined electrical logic sequence. For example, the rule base pre-stores a standard sequence for "switching": [1. Disconnect the circuit breaker -> 2. Confirm that the circuit breaker is disconnected -> 3. Open the load-side disconnect switch -> 4. Open the power-side disconnect switch]. The system will verify the actual operation steps according to this sequence.

[0030] Step S202: Receive the two vote images uploaded by the user and preprocess the images; Specifically, a deep learning-based optical character recognition engine is used for text recognition; Based on the pre-learned fixed template structure and field layout of two tickets, key information areas in the work content, staff, planned work time and operation steps are located by image recognition technology. Identify and extract printed or handwritten text within the located area, and convert it into structured text data.

[0031] In one embodiment, after the operators complete the "work order" and the corresponding "measure operation order" and "recovery operation order", they can upload clear images of these three orders in batches through the system's web interface.

[0032] Step S203: Locate the key information region and recognize the text in the preprocessed image to extract the key information content; In one embodiment, the system calls the image preprocessing module to perform grayscale conversion, noise reduction, and tilt correction on the uploaded image to improve recognition accuracy; Subsequently, the text recognition and extraction module was activated. This module employs a deep learning-based optical character recognition engine, specifically optimized for training on the two forms and handwritten characters. Locating key areas: Based on the pre-learned two-ticket template, automatically locate fields such as "work content", "three types of people", and "planned work time" on the "work ticket".

[0033] Extracting text information: Identifying and extracting text from these areas. For example: Job Description: Inspect and troubleshoot malfunction of the No. 1 hydraulic oil pump of the governor of Unit 3 at Tongtou Power Station. Supervisor: Guo Hong. Planned Start Time: 2025-08-18 09:40. Planned Completion Time: 2025-08-20 18:00.

[0034] Related Operation Ticket: Check and take measures to address the abnormal operation of the No. 1 oil pump of the governor of Unit 3 at Tongtou Power Station.

[0035] Step S204: Automatically compare the extracted key information content with the pre-stored compliance rules and perform compliance logic judgment; Specifically, when the comparison results show that the information content does not conform to the preset rules, it is judged as a violation; For each violation, record the ticket information, the specific description of the violation, and the rule number that was triggered; Generate and store detailed judgment logs containing the reasons for violations.

[0036] In one embodiment, the system's compliance analysis engine begins to work, automatically comparing the extracted information with the rule base: Execution rule R1: The engine queries the personnel permission knowledge base to determine whether "Guo Hong" possesses the qualifications of "work supervisor". The query result is "yes", indicating compliance.

[0037] Execution rule R2: The engine compares the "planned start time" and the "planned completion time" and determines that the time logic is reasonable and the item is compliant.

[0038] Rule R3 was executed: The engine check found that the uploaded tickets included "Work Ticket", "Measure Operation Ticket" and "Recovery Operation Ticket", and the ticket association was complete, which is compliant.

[0039] Rule R4 was executed: The engine analyzed the operation steps of the "Measures Operation Ticket" and found that the sequence was [Hang a "Do Not Close, People Working" sign at power switch 5D-2; Turn off power switch 5D-2]. This does not match the standard sequence [Operate the equipment first, then hang the sign] stored in the rule base. Therefore, the engine determined that the operation ticket had a logical sequence error and recorded the reason for the violation as: "Violation of the 'Operations Ticket Execution Sequence Rule': The safety sign should be hung after the equipment operation."

[0040] Step S205: Based on the judgment results, generate and output a structured compliance review report.

[0041] Specifically, all compliance assessment results are summarized to generate a structured review report; the report includes a review summary, a detailed list of violations, and a summary of compliance items. The generated review report can be provided to users for download through the system interface, or automatically pushed to an external safety production management system for further processing through a preset system interface.

[0042] In one embodiment, the report generation module summarizes the judgment results of all rules and automatically generates a structured compliance review report. This report is presented as an HTML webpage or a PDF document, and its content clearly includes: Review Summary: Total number of votes, number of compliant items, number of non-compliant items.

[0043] Violation Details List: This list details the ticket for each violation, the violation content, the triggering rule number, and the specific reason for the violation. For example: "Operation Ticket (Ticket No.: XX-XXX), incorrect sequence of steps, violating Rule R4: Safety measures (hanging identification tags) should be performed after the equipment is de-energized." Compliance Summary: Lists all items that passed the inspection.

[0044] The system supports exporting reports with one click or pushing them to the safety production management system via an interface to complete the entire automated review process.

[0045] This embodiment provides an automatic identification method for the compliance of power-related invoices and related documents. It automatically extracts information from the invoices using optical character recognition technology and performs automated logical comparison using a rule engine, completely changing the traditional model that relies on manual inspection by reviewers. The system can complete the review of batches of invoices and related documents within minutes or even seconds, greatly freeing up manpower and increasing review efficiency by tens of times, meeting the high requirements of modern power production for rapid response to safety management. By solidifying compliance judgment standards into executable computer rules, it eliminates judgment biases and omissions caused by factors such as personal experience, professional level, fatigue, and subjective emotions in manual review. For the same invoice, the system can provide a unique and definitive review conclusion, effectively solving the problems of inconsistent standards among different reviewers and fluctuations in standards by the same reviewer, greatly enhancing the standardization and authority of the safety management system.

[0046] Based on the above embodiments, this embodiment provides a detailed description of an automatic identification system for compliance of electricity two-tickets, as follows: This embodiment takes a specific case of "inspection and handling of abnormal operation of No. 1 oil pump of governor of Unit 3 of Tongtou Power Station" as an example to fully demonstrate the operation process of the system of the present invention.

[0047] Step 1: System Initialization and Rule Base Construction The system administrator performs initial configuration in the system backend.

[0048] Personnel permission import: as attached Figure 1 As shown, import an Excel spreadsheet containing "Name", "Employee ID", "Department", and "Three Types of Qualifications" into the system. In this example, the qualification field for the employee "Guo Hong" includes "Work Supervisor".

[0049] Compliance rule configuration: In accordance with the "Electric Power Safety Work Regulations" and the power plant's detailed rules, the following core rules are pre-defined in the rule base: Rule R1 (Personnel Qualification Rule): The "Work Manager" on the "Work Order" must exist in the personnel permission database and possess the qualifications of a "Work Manager".

[0050] Rule R2 (Time Logic Rule): The "Planned Start Time" on the "Work Order" must be earlier than the "Planned Completion Time".

[0051] Rule R3 (Ticket Association Rule): A maintenance work must have a valid "Work Ticket", "Measurement Operation Ticket" and "Resumption Operation Ticket" at the same time.

[0052] Rule R4 (Operation Sequence Rule - Measures): The standard sequence of steps on the measures operation ticket should be: [1. Disconnect the equipment power supply -> 2. Test for voltage -> 3. Hang the safety sign].

[0053] Step 2: Ticket Image Upload and Preprocessing: After the operators have completed all the necessary permits for this task, they can upload three clear images of the permits in batches through the system's web interface: work permit, operational permit, and recovery permit. After receiving the image, the system automatically calls the image preprocessing module to perform grayscale conversion, noise reduction, and tilt correction on the uploaded image to improve the accuracy of subsequent text recognition.

[0054] Step 3: Locating and Extracting Key Information The system's optical character recognition engine starts up, and based on the pre-learned "two-ticket" template structure, it automatically locates key field areas and recognizes and extracts text information.

[0055] Extracted from the work order: Job Responsibilities: Inspect and troubleshoot malfunctions of the No. 1 hydraulic pump of the governor of Unit 3 at the Tongtou Power Station. Work Supervisor: Guo Hong Planned start date: 2025-08-18 09:40 Planned completion time: 2025-08-20 18:00 Extract from the operational ticket: Related work content: Inspecting and handling abnormal operation of the No. 1 hydraulic pump of the governor of Unit 3 at Tongtou Power Station and taking appropriate measures. Operating procedure sequence: [1. Hang a "Do Not Close, People Working" sign at power switch 5D-2; 2. Turn off power switch 5D-2] Step 4: Compliance Logic Judgment and Execution The compliance analysis engine automatically compares the extracted information with the rule base: Rule R1 was executed: The engine queried the personnel permission database and confirmed that "Guo Hong" possessed the qualifications of "Work Supervisor". Result: Compliant.

[0056] Rule R2 was executed: the engine compared times and confirmed that "2025-08-18 09:40" was earlier than "2025-08-20 18:00". Result: Compliant.

[0057] Rule R3 was executed: Engine checks revealed that the "Work Order," "Measure Operation Order," and "Recovery Operation Order" have been uploaded for this task, and the order associations are complete. Judgment: Compliant.

[0058] Execution Rule R4: The engine compared the steps of the operation ticket [hang the identification sign first, then turn on the power] with the standard sequence [operate the equipment first, then hang the identification sign]. It was found that the actual operation sequence was significantly different from the standard sequence. Judgment: Violation. The engine recorded the reason for the violation as: "Violation of operation ticket execution sequence rule R4. Safety measures (hanging the identification sign) should be performed after the equipment power-off operation (turning on the power switch) has been confirmed." Step 5: Review Report Generation and Output The report generation module summarizes all judgment results and automatically generates a structured compliance review report (presented in HTML webpage format). Operators can view the report online and download the PDF version for archiving with one click, or the system can automatically push the report to the power plant's Safety Management System (SAMS) to trigger subsequent rectification processes.

[0059] This embodiment clearly demonstrates the entire process of the system of the present invention, from uploading tickets to generating reports. This example verifies that the present invention can efficiently and accurately automatically identify a major safety hazard—a logical error in the operational sequence that is easily overlooked during manual review—fully demonstrating the significant value of this system in improving the safety, standardization, and management efficiency of power operations.

[0060] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. An automatic identification system for compliance of electricity ticketing and permitting, characterized in that, include: The rule base construction module is used to pre-store compliance rules for the power industry's two-ticket system. The image preprocessing module is used to receive and preprocess the two images uploaded by the user; The text extraction module, connected to the image preprocessing module, is used to locate and identify the text content of the key information fields of the two tickets from the preprocessed image; The compliance analysis module is connected to the rule base construction module and the text extraction module. It is used to compare the key information extracted by the text extraction module with the compliance rules pre-stored in the rule base construction module and output the compliance judgment result. The report generation module, connected to the compliance analysis module, is used to generate a structured compliance review report based on the compliance judgment results.

2. The automatic identification system for compliance of electricity two-ticket system according to claim 1, characterized in that, The aforementioned compliance rules for the two-ticket system in the power industry include: Personnel qualification rules are used to determine whether the staff on the work permit have the corresponding qualifications; Time logic rules are used to determine whether the planned time logic on the work order is reasonable; Ticket association rules are used to determine whether the association relationship between work tickets, operation tickets, and recovery tickets is complete. Operation sequence rules are used to determine whether the order of operation steps in an operation ticket conforms to a predetermined standard sequence.

3. The automatic identification system for compliance of electricity two-ticket system according to claim 1, characterized in that, The text extraction module employs deep learning-based optical character recognition technology and has been optimized and trained for the table format and handwritten font of power industry invoices.

4. The automatic identification system for compliance of electricity two-ticket system according to claim 1, characterized in that, The text content used to locate and identify the key information fields of the two tickets from the preprocessed image includes: Based on pre-learned two-ticket templates, key information areas are located; Identify and extract the text content within the key information area, whereby the key information includes: work content, staff, planned work time, and operating steps.

5. The automatic identification system for compliance of electricity two-ticket system according to claim 1, characterized in that, The structured compliance review report includes: Review summary information, including the total number of votes, the number of compliant items, and the number of non-compliant items; a detailed list of non-compliant items, listing the non-compliant tickets, the content of the non-compliant items, and the rules that triggered them; and a summary list of compliant items.

6. A method for automatically identifying the compliance of electricity-related invoices, characterized in that, include: Build a rule base and pre-store the compliance rules for the two-ticket system in the power industry; Receive two images of votes uploaded by the user and preprocess the images; The key information regions of the preprocessed image are located and text is recognized to extract key information content; The extracted key information is automatically compared with pre-stored compliance rules, and compliance logic judgments are executed. Based on the assessment results, a structured compliance review report is generated and output.

7. The automatic identification method for compliance of electricity two-ticket system according to claim 6, characterized in that, The rule base being constructed includes pre-stored compliance rules for the power industry's two-ticket system, including: The name of the person in charge of the work extracted from the work order is compared with the pre-stored personnel permission knowledge base to verify whether they have the corresponding qualifications. Compare the planned start time and planned completion time extracted from the work order to verify whether the time sequence logic is reasonable; Verify whether there are valid work orders, action orders, and recovery orders for the same work content at the same time; The sequence of operation steps extracted from the operation ticket is compared sequentially with the standard electrical operation sequence pre-stored in the rule base.

8. The automatic identification method for compliance of electricity two-ticket system according to claim 6, characterized in that, The step of locating key information regions and recognizing text in the preprocessed image to extract key information content includes: Text recognition is performed using a deep learning-based optical character recognition engine; Based on the pre-learned fixed template structure and field layout of two tickets, key information areas in the work content, staff, planned work time and operation steps are located by image recognition technology. Identify and extract printed or handwritten text within the located area, and convert it into structured text data.

9. The automatic identification method for compliance of electricity two-ticket system according to claim 6, characterized in that, The step of automatically comparing the extracted key information with pre-stored compliance rules and performing compliance logic judgments includes: When the comparison results show that the information content does not conform to the preset rules, it is judged as a violation; For each violation, record the ticket information, the specific description of the violation, and the rule number that was triggered; Generate and store detailed judgment logs containing the reasons for violations.

10. The automatic identification method for compliance of electricity two-ticket system according to claim 6, characterized in that, The generation and output of a structured compliance review report based on the judgment results includes: Summarize all compliance assessment results to generate a structured review report; the report includes a review summary, a detailed list of violations, and a summary of compliance items. The generated review report can be provided to users for download through the system interface, or automatically pushed to an external safety production management system for further processing through a preset system interface.

Images