Intelligent power operation ticket generation method and system

Abstract

The invention provides a power operation ticket intelligent generation method and system, and relates to the technical field of intelligent ticket generation. Inputting the natural language operation task into a pre-training large model fused with a power industry knowledge base, performing intention analysis, and extracting a target scene, an operation object and operation action information; based on the operation object information, a power equipment ledger management system is called in an associated mode through a data interface, and equipment data is obtained; according to the intention analysis result and the equipment data, constructing a multi-dimensional matching condition, screening out an optimal historical operation ticket template from a preset standardized operation ticket knowledge base, performing consistency verification, filtering conflict operation items, and generating a to-be-verified operation ticket; and calling the anti-error rule base and associating the real-time state of the equipment, performing multi-level deep verification on the to-be-verified operation ticket, and generating a verification result and a correction suggestion. The accuracy, normalization and efficiency of power operation ticket generation are improved, and the risk of manual operation is reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent ticket generation technology, and in particular to an intelligent generation method and system for power operation tickets. Background Technology

[0002] Power operation tickets are core technical documents in power operation and maintenance that guide equipment operation and mitigate safety risks. They must clearly define operating procedures, safety measures, and equipment status requirements. They are crucial for ensuring the compliant conduct of grid switching operations and equipment maintenance, and their generation efficiency and verification accuracy directly impact the safety and efficiency of operation and maintenance work. With the advancement of power grid digitalization, the PMS 3.0 system has shifted operation tickets from "offline" to "online," and mobile apps and other tools have addressed the convenience of ticket issuance in special scenarios. However, these still fall short of meeting the advanced demands of smart grids for more efficient, accurate, and safe operation tickets.

[0003] While the current operation ticket generation mode based on PMS3.0 offers methods such as copying historical tickets and graphical ticketing, it has not broken through the core limitation of traditional reliance on manual processes. Existing technologies have significant drawbacks: First, knowledge reuse efficiency is low. Historical operation tickets and procedural clauses are scattered and unstructured, making manual querying and matching cumbersome, and rough template adaptation easily leads to non-standard ticket designs. Second, error prevention verification is lagging, relying on manual verification by operators, making it difficult to detect deep logical contradictions and inconsistencies in equipment status, easily leading to safety incidents. Third, cross-system collaboration is poor; ticketing requires manually querying equipment IDs and topology parameters across multiple systems and writing steps item by item, which is inefficient and prone to errors. Fourth, the readability of verification results is poor, only outputting "pass / fail" or error codes, making it difficult for maintenance personnel to quickly locate problems, providing weak guidance for correction, increasing the workload of frontline staff, and hindering the improvement of intelligent operation and maintenance levels. Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes an intelligent generation method and system for power operation tickets, which improves the accuracy, standardization, and efficiency of power operation ticket generation and reduces the risks associated with manual operation.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for intelligent generation of power operation tickets, comprising: The system acquires unstructured natural language operation tasks from user input, inputs them into a pre-trained large model that integrates a power industry knowledge base, performs intent parsing, and extracts information on the target scenario, operation object, and operation action. Based on the operation object information, the power equipment ledger management system is invoked through the data interface to obtain the corresponding equipment data; Based on the intent parsing results and device data, multi-dimensional matching conditions are constructed, and the optimal historical operation ticket template is selected from the preset standardized operation ticket knowledge base. Perform consistency verification on the optimal historical operation ticket template, filter conflicting operation items, and generate an operation ticket to be verified; The error prevention rule base is invoked and associated with the real-time status of the device to perform multi-level in-depth verification on the operation ticket to be verified, and verification results and correction suggestions are generated.

[0006] Secondly, the present invention provides an intelligent power operation ticket generation system, comprising: The intent parsing module is configured to acquire unstructured natural language operation tasks input by the user, input them into a pre-trained large model that integrates the power industry knowledge base, perform intent parsing, and extract information on the target scene, operation object, and operation action. The equipment data acquisition module is configured to, based on the operation object information, call the power equipment ledger management system through a data interface to obtain the corresponding equipment data; The template filtering module is configured to construct multi-dimensional matching conditions based on intent parsing results and device data, and filter out the best historical operation ticket template from the preset standardized operation ticket knowledge base. The ticket template optimization module is configured to perform consistency verification on the optimal historical operation ticket template, filter conflicting operation items, and generate an operation ticket to be verified. The deep verification module is configured to call the anti-error rule base and associate it with the real-time status of the device to perform multi-level deep verification on the operation ticket to be verified, and generate verification results and correction suggestions.

[0007] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the intelligent generation method for power operation tickets described in the first aspect.

[0008] Fourthly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the intelligent generation method for power operation tickets described in the first aspect.

[0009] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) This invention, by integrating a pre-trained large model from a power industry knowledge base, accurately parses unstructured natural language operation tasks and efficiently extracts core operation information, solving the problems of low efficiency and large information extraction deviations in traditional manual input. Simultaneously, based on the data interface and the associated equipment ledger management system, combined with real-time equipment status and error prevention rule base, it achieves fully automated processing from template selection and conflict filtering to multi-level deep verification, ensuring consistency between the operation ticket and the actual equipment status while strictly adhering to industry norms and safety standards. Compared to traditional methods, this significantly shortens the operation ticket generation cycle, reduces oversights caused by manual intervention, effectively lowers the safety risks of power operations, and provides reliable technical support for the safe and stable operation of the power system.

[0010] (2) In the traditional approach, historical operation tickets, procedural clauses, and equipment topology are scattered and unstructured, resulting in cumbersome querying and rough template matching during ticket issuance. This invention constructs a standardized knowledge base that supports natural language interaction and performs structured cleaning and semantic annotation on massive historical tickets, enabling the system to accurately parse task intent based on a large model. Through multi-dimensional similarity calculation, it dynamically matches the optimal historical template from the knowledge base, effectively improving the accuracy and standardization of ticket issuance.

[0011] (3) Traditional verification relies on manual verification by operators, making it difficult to detect deep-seated logical contradictions and inconsistencies in status. This invention uses intelligent agent technology to automatically retrieve the real-time status of the equipment after ticket issuance, associate it with the anti-error rule base, and use a semantic big data model to perform multi-level, closed-loop automatic verification of the operation logic. By using a step alignment algorithm to accurately locate errors and provide correction suggestions, it transforms post-event detection into pre-event prevention, fundamentally eliminating safety accidents caused by logical errors.

[0012] (4) Currently, invoicing requires personnel to manually query equipment IDs and topology parameters across multiple systems and write operation steps item by item, which is inefficient and prone to errors. This invention automatically parses operation tasks through a large model and automatically obtains key parameters (such as PSR-ID) by associating with the equipment ledger system. Finally, it automatically outputs standardized invoices with pre-filled logical order, security measures, and equipment permission identifiers, optimizing the manual multi-step operation into a system-driven one-click generation, effectively improving invoicing efficiency.

[0013] (5) Traditional rule validation often only outputs "pass / fail" or simple error codes, making it difficult for maintenance personnel to quickly understand the problem and make corrections. After completing the validation, this invention uses an advanced step alignment algorithm to visualize and compare the newly added, deleted, or erroneous items found by the system with the original ticket, and directly outputs a clear report containing error location and specific correction suggestions, which greatly reduces the technical threshold and time cost of result interpretation and ticket correction.

[0014] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0015] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute a limitation thereof.

[0016] Figure 1 The main flowchart of an intelligent power operation ticket generation method provided in an embodiment of the present invention is shown. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0018] Example 1 like Figure 1 As shown in the figure, this embodiment discloses a method for intelligent generation of power operation tickets, including the following steps: S1: Obtain unstructured natural language operation tasks from user input, input them into a pre-trained large model that integrates the power industry knowledge base, perform intent parsing, and extract target scene, operation object, and operation action information; S2: Based on the operation object information, the power equipment ledger management system is called through the data interface to obtain the corresponding equipment data; S3: Based on the intent parsing results and device data, construct multi-dimensional matching conditions and select the optimal historical operation ticket template from the preset standardized operation ticket knowledge base; S4: Perform consistency verification on the optimal historical operation ticket template, filter conflicting operation items, and generate an operation ticket to be verified; S5: Call the anti-error rule base and associate it with the real-time status of the device to perform multi-level deep verification on the operation ticket to be verified, and generate verification results and correction suggestions.

[0019] Next, combined Figure 1 This embodiment provides a detailed description of a method for intelligently generating power operation tickets.

[0020] In S1, based on a knowledge-driven mechanism, unstructured natural language operation tasks are transformed into standardized operation ticket templates with accurate matching capabilities, providing a compliant and efficient core basis for the automated generation of subsequent operation tickets.

[0021] First, the system receives a natural language description of the operation task input by the user. This natural language description is unstructured text information. For example, the received natural language description is "Close the low-voltage side switch of the No. 1 main transformer in substation A".

[0022] It is evident that this only indicates the user's core operational needs, but does not include key elements such as standardized equipment identification, equipment parameters, and safe operation rules required for ticket generation. Therefore, it is impossible to directly generate a standardized operation ticket that meets the requirements. Thus, this embodiment proposes to perform intent parsing based on a semantic model.

[0023] Specifically, a pre-trained large-scale model integrating a power industry operational terminology database and a safety procedure knowledge base is used to structurally decompose the received natural language operation tasks. Through keyword extraction algorithms, semantic association analysis models, and ambiguity elimination logic, core information is captured, including target scenario information, operation object information, and operation action information.

[0024] It should be understood that the pre-trained large model integrating the power industry knowledge base, the structured decomposition of the operation ticket text, and the application of keyword extraction, semantic association analysis, and ambiguity elimination are all things that can be achieved by those skilled in the art. This can be achieved by using existing natural language processing (NLP) technologies and models, and by adapting, fine-tuning, or constructing rules in conjunction with the power industry's professional terminology database and safety regulations. These will not be elaborated upon here.

[0025] Among them, the target scenario information clearly specifies the name of the substation to which the operation belongs, such as substation A; the operation object information includes attributes such as equipment type, voltage level, and unique number, such as the low-voltage side switch of main transformer No. 1; and the operation action information is the user's core operation objective, such as closing.

[0026] For natural language tasks that are vague or lack information about the target scene, the object of operation, or the operation action, the pre-trained large model can automatically identify the missing items of the core operation intent and trigger a user supplementation prompt mechanism to clarify key information such as the operation action, thus avoiding subsequent ticketing deviations due to semantic ambiguity.

[0027] It should be understood that the automatic identification process uses natural language processing techniques, such as named entity recognition and semantic slot filling, to extract key information such as substation name, equipment type, and operation action from the user's input operation request, and compares it with preset core information dimensions to determine whether there is missing information and prompt the user to supplement it.

[0028] In S2, the extracted core information is used as search keywords for data association: an automatic connection is established with the power equipment ledger management system through a preset API interface for targeted access to relevant data of the target equipment. The power equipment ledger management system, with equipment lifecycle management at its core, stores structured information such as basic attributes, technical parameters, topological relationships, operating status records, historical operation data, and unique identifiers (PSR-ID) for all power equipment. It serves as the core data support platform for ensuring equipment operation and maintenance, operation ticket generation, and maintenance plan formulation.

[0029] Specifically, the system prioritizes matching the unique identifier of the device, namely PSR-ID, based on the information of the operation object, and simultaneously obtains key data such as the topology parameters, current operating status, and historical operation records of the device. For example, the topology parameters are: the low-voltage side switch of the No. 1 main transformer connects the low-voltage winding of the No. 1 main transformer to the 380V bus, and the associated protection device P2; the operating status includes in operation and standby.

[0030] It should be understood that PSR-ID and topology parameters can be obtained through chained queries. First, based on the equipment description information in the operation request, the unique corresponding equipment is retrieved and located in the power equipment ledger management system, and its PSR-ID is obtained. Then, using this PSR-ID as a unique index, the ledger system is further queried to extract detailed information such as the topology parameters associated with the equipment.

[0031] This targeted data retrieval method effectively avoids the data redundancy and information chaos caused by blindly retrieving all ledger data in the traditional model, ensuring the accuracy and effectiveness of the data input into subsequent stages.

[0032] In S3, after data association is completed, template matching is performed: based on the intent parsing results and the device-related information obtained from data association, multi-dimensional template matching conditions are constructed, and the optimal historical ticket template is dynamically selected and recommended from the preset standardized operation ticket knowledge base.

[0033] The construction of the standardized operation ticket knowledge base is specifically based on power industry standards and internal enterprise operating procedures, combined with the equipment configuration characteristics of each substation and historical compliant operation ticket data, and is carried out through the following steps: Basic template entry: Collect standard operating procedures for various types of operations, break them down into standardized steps of inspection, confirmation, operation, and review, and form a basic template library; various types of operations include equipment commissioning, shutdown, maintenance, etc.

[0034] Data structured labeling: Supplement each type of template with related dimension information, including the applicable device type, voltage level, unique number range, corresponding task type, safety rules to be followed, and associated device PSR-ID field, forming a structured labeling system; Dynamic update mechanism: Regularly integrate the execution feedback data of historical operation tickets, including execution success rate, reasons for operation deviation, etc., eliminate templates that do not meet actual operation needs or have safety hazards, and add standardized templates corresponding to new equipment and new operation scenarios. Safety rule embedding: The special safety requirements of each substation, including the cooling system inspection before the operation of specific equipment and the load clearing requirements, are transformed into template verification rules, which are stored in association with the template to ensure compliance verification when the template is called.

[0035] Furthermore, the multi-dimensional matching conditions include four dimensions: precise task type matching, equipment level adaptation, safety rule compliance, and priority based on historical execution results. Precise task type matching refers to determining the task type based on the combination of the operation action and the operation object. For example, "closing the transformer switch" corresponds to a transformer commissioning template, avoiding misuse of cross-type templates. Equipment level adaptation involves fine-tuning based on parameters such as equipment voltage level, equipment type, and equipment unique number to ensure that the template and the operation object's attributes are completely matched. Safety rule compliance incorporates the specific safety operation requirements of the target substation, such as "the cooling system must be checked before operating the main transformer of substation A," automatically excluding templates that do not meet safety standards through a rule verification mechanism. Prioritizing historical execution results means that based on data such as the execution success rate and operation deviation records of historical templates, historical templates with a 100% success rate and no operation deviations are prioritized, improving the reliability of the ticket issuance.

[0036] For example, based on the matching condition "Close operation of the low-voltage side switch of main transformer No. 1 in substation A", the optimal template selected by the system from the standardized operation ticket knowledge base is as follows: The cooling system of main transformer No. 1 was checked and found to be operating normally. The voltage on the low-voltage side busbar of main transformer No. 1 has been confirmed to be normal. Close the low-voltage side switch of main transformer No. 1; Check that the switch is fully closed, the position indicator is correct, and the current and voltage indicators are normal.

[0037] This template integrates key information such as the specific PSR-ID of the equipment and related cooling system checks, which can lay the foundation for the rapid generation of subsequent operation tickets.

[0038] This embodiment effectively breaks down data barriers between multiple systems such as PMS, equipment ledger, and real-time measurement, enabling automatic acquisition and filling of key parameters (such as PSR-ID) and automatic feedback and synchronization of operation results. This fundamentally eliminates redundant work such as manual cross-system queries and repetitive data entry, freeing frontline staff from tedious collaborative tasks and substantially reducing their workload.

[0039] In S4, after template matching is completed, in order to ensure that the matched operation ticket template is fully consistent with the scenario, device attributes and security rules of the current operation task and to avoid operational risks due to information discrepancies, a consistency verification process is initiated.

[0040] The pre-trained large model, which integrates the power industry operation terminology database and safety procedure knowledge base, is invoked again to perform a full-dimensional scenario consistency check on the current operation task and the matched optimal historical ticket template.

[0041] Consistency verification includes three dimensions: scenario adaptability verification, device attribute consistency verification, and operation logic consistency verification.

[0042] Among them, the scenario adaptability verification compares the substation scenario and equipment configuration characteristics corresponding to the template with the target scenario information of the current task, such as whether the equipment layout and safety management requirements of substation A are consistent, so as to avoid misusing a special operation template applicable to substation B to substation A. Equipment attribute consistency verification combines the equipment PSR-ID, topology parameters, voltage level and other information obtained in the data association stage to verify whether the equipment type, associated device and parameter range of the operation object in the template are completely matched with the current target equipment. For example, it confirms that "associated protection device P2" in the template is consistent with the actual associated protection device of the low-voltage side switch of the current No. 1 main transformer. Operational logic consistency verification uses a semantic association analysis model to verify whether the template's operational flow aligns with the current task's actions and equipment operating status, thus avoiding logical inversion issues such as performing operations before checking. If inconsistencies are detected, the system will automatically trigger a template re-filtering mechanism, or perform preliminary adaptation adjustments to the original template based on the differences before re-verifying, ensuring a high degree of fit between the template and the task.

[0043] Furthermore, after the consistency check passes, the system enters the conflict filtering stage. Based on the real-time data returned by the power equipment ledger management system, the safety procedure knowledge base, and the equipment topology relationship, the system automatically identifies and filters invalid or conflicting operation items.

[0044] Specifically, the invalid operation filtering targets operation steps in the template that have become invalid due to equipment updates or procedure iterations. For example, if the template states: "Check the status of the old relay protection device," but the current equipment has been replaced with an intelligent protection device, then this step is determined to be invalid.

[0045] Conflict operation item filtering includes three conflict types: conflict with the current operating status of the device, conflict with topology association rules, and conflict with historical operation records. For example, a conflict with historical operation records might be defined as follows: if the historical operation record shows that the switch was disconnected due to a fault one hour ago and has not yet been repaired, but the template provides an operation item of "directly close the switch" which is considered a conflict.

[0046] During the filtering process, the operation items are semantically compared and logically verified with real-time operating status, topology rules, and historical data. Conflicting items are marked and automatically removed. At the same time, necessary alternative operation steps are supplemented based on safety procedures. For example, "directly close the switch" is replaced with "check that the switch fault has been eliminated, confirm that the maintenance work is completed, and close the switch".

[0047] Furthermore, after conflict filtering is completed, the optimized template is structurally integrated and standardized to automatically generate a complete, logically clear, and compliant operation ticket to be verified.

[0048] The specific integration process includes: following the standardized logical sequence of inspection, confirmation, operation, and review, supplementing the equipment PSR-ID and topology association information obtained in the data association stage, embedding special safety measures of the target substation, and marking the permission level and responsible person fields required for the operation.

[0049] For example, the final output of the operation ticket to be verified is: The cooling system of main transformer No. 1 was checked and found to be operating normally (no alarm from associated protection device P2). The voltage on the low-voltage side busbar of main transformer No. 1 was confirmed to be 380V, which meets the operating requirements. Close the low-voltage side switch of main transformer No. 1 (PSR-ID: A-BTB-001). Check that the switch is fully closed, the position indicator is correct, and the current and voltage indicators are normal. Record the operation time, operator, and reviewer information.

[0050] In this embodiment, the above process enables "one-click ticket generation" from unstructured natural language tasks to standardized operation tickets, effectively improving the efficiency and compliance of operation ticket generation and providing a reliable basis for subsequent manual verification and on-site operation.

[0051] In S5, the operation ticket to be verified is received and verified.

[0052] First, preprocessing is performed, which involves standardizing and numbering all operation steps on the received tickets to be verified, and then performing semantic word segmentation and structured annotation on each step to eliminate differences in ticket format and semantic ambiguity.

[0053] For example, steps can be labeled as "inspection", "operation", or "confirmation" to extract key elements such as equipment name, action instructions, and associated devices.

[0054] Subsequently, through a preset data interface, real-time communication is established with the power equipment online monitoring system and SCADA (Supervisory Control and Data Acquisition) system to retrieve the current real-time operating status of all equipment involved in the ticket to be verified. At the same time, a preset anti-error rule base is loaded, and a complete verification context is constructed by combining the real-time equipment status and the topology parameters in the ticket, clarifying the preconditions and constraint rules for each step of the operation.

[0055] The pre-set anti-misoperation rule base is built based on procedures and internal enterprise anti-misoperation operation specifications, and includes equipment operation logic constraints, topology association taboos, state transition rules, etc.

[0056] Furthermore, a semantic big data model integrating power industry error prevention rules is invoked to perform multi-level deep verification on the ticket.

[0057] Operation and status consistency verification compares whether each operation instruction matches the real-time status of the equipment. For example, if the ticket step is "close the low-voltage side switch of main transformer No. 1", and the real-time status shows that the switch is already in the "closed" state, it is determined that "operation conflicts with current status"; if the real-time status shows that the associated protection device P2 has an "overcurrent alarm", it is determined that "operation preconditions are not met".

[0058] The logical rationality of the steps is verified to ensure that the order of the operation steps conforms to the safe operating procedures. For example, is there a logical reversal of "operate first, check later"? Are there any missing causal relationships between the steps?

[0059] Rule compliance verification checks whether each operation complies with the prohibitions and constraints in the anti-error rule base. For example, if the ticket fails to check the 380V bus voltage before closing the circuit, it violates the rule that the bus voltage must be confirmed to be normal before closing the circuit, and is judged as a compliance error.

[0060] This embodiment constructs a proactive security defense system of "generation as verification". By deeply embedding intelligent verification into the process and associating it with the real-time status of the device and the anti-error rule base, and using a semantic big data model for closed-loop logic verification, it can discover deep-seated logical contradictions and state conflicts in advance, achieving an essential security leap from post-event remediation to pre-event prevention, and effectively reducing operational risks.

[0061] Furthermore, the output results can be presented in the following ways: A step alignment algorithm is used to correlate all the problems found in the multi-level deep verification with the steps on the original ticket, and the comparison results are presented intuitively through a visual interface.

[0062] For example, incorrect steps are marked in red, and the problem type, real-time status evidence, and corresponding error prevention rules are displayed in a pop-up window or sidebar; for steps with incorrect logical order, the correct order is marked with an arrow, so as to achieve accurate mapping between the problem and the ticket, making it convenient for maintenance personnel to quickly locate the problem.

[0063] This also includes integrating verification results to generate a standardized verification report. The report contains basic ticket information, problem statistics, precise error location, and specific correction suggestions. The report uses a combination of text and graphics, greatly improving readability and on-site guidance.

[0064] To address the problem of obscure and difficult-to-understand output results from traditional systems, this embodiment uses a step alignment algorithm and visualization mapping technology to transform complex AI verification results into intuitive and readable reports. These reports can accurately pinpoint errors and provide clear correction guidance. This significantly reduces the technical requirements for users, making human-machine collaboration smoother and more efficient, and accelerating the completion of the business loop.

[0065] Furthermore, it also includes an adaptive optimization mechanism for operation tickets and a self-learning mechanism for the knowledge base to ensure continuous iteration of the system's verification capabilities: After each operation ticket is executed, the system automatically collects information on the target scenario, operation object, and operation action through a preset data acquisition channel and feeds it back to the standardized operation ticket knowledge base and error prevention rule base. 1. Operation Tracking: By comparing the on-site operation records with the operation tickets generated by the system, the differences in actual operation are recorded, including added steps (such as adding an additional "confirm the completion of maintenance work" step during on-site execution), deleted steps (such as not executing the "check bus voltage" step on-site and noting the reason), or sequence adjustments (such as adjusting the order of the "check" and "confirm" steps on-site).

[0066] 2. On-site handling record: Collect abnormal situations encountered during operation (such as sudden alarm of protection device during operation, inconsistency between equipment status and real-time monitoring data) and corresponding handling solutions (such as suspending operation, contacting maintenance personnel to investigate alarms, and re-verifying equipment status).

[0067] 3. Feedback from Operations and Maintenance Personnel: Through the preset interactive interface of the operation terminal, the system collects operators' ratings (out of 1-5 points) on the rationality and clarity of the ticket steps, as well as specific modification suggestions (such as "suggest adding steps for verifying protection device settings" and "the step description should clearly define the voltage acceptable range"). The system periodically performs statistical analysis on the feedback data. If the execution trajectory difference rate of a certain type of operation exceeds a preset threshold (such as 30%), the system automatically triggers the template optimization process and updates the corresponding template in the standardized operation ticket knowledge base. If the on-site handling records or feedback from operations and maintenance personnel reveal omissions in the anti-error rule library (such as the rule "operation must be suspended when the protection device alarms"), the system automatically adds it to the rule library, realizing a closed-loop self-learning of "verification-execution-feedback-optimization" to continuously improve the accuracy and relevance of intelligent ticket generation and verification.

[0068] This specific embodiment utilizes natural language intent parsing and targeted data association to rapidly transform unstructured operation tasks into standardized ticket templates. This eliminates the redundant work of manually querying equipment data and breaking down tasks, significantly shortening the operation ticket generation time and reducing the burden on grassroots maintenance personnel. Simultaneously, multi-dimensional template screening and dual verification through consistency and deep validation, combined with real-time equipment status and error prevention rules, accurately avoids problems such as template adaptation errors, inverted operation logic, and state conflicts, reducing power operation risks from the source and ensuring the safety of the power grid and equipment. Furthermore, it breaks down data barriers between multiple systems, achieving a closed loop of generation and verification, and can be continuously optimized through a self-learning mechanism, balancing current efficiency improvements with long-term technological iteration, providing reliable support for the digital and intelligent upgrade of power operation and maintenance.

[0069] Example 2 This embodiment provides an intelligent power operation ticket generation system, including: The intent parsing module is configured to acquire unstructured natural language operation tasks input by the user, input them into a pre-trained large model that integrates the power industry knowledge base, perform intent parsing, and extract information on the target scene, operation object, and operation action. The equipment data acquisition module is configured to, based on the operation object information, call the power equipment ledger management system through a data interface to obtain the corresponding equipment data; The template filtering module is configured to construct multi-dimensional matching conditions based on intent parsing results and device data, and filter out the best historical operation ticket template from the preset standardized operation ticket knowledge base. The ticket template optimization module is configured to perform consistency verification on the optimal historical operation ticket template, filter conflicting operation items, and generate an operation ticket to be verified. The deep verification module is configured to call the anti-error rule base and associate it with the real-time status of the device to perform multi-level deep verification on the operation ticket to be verified, and generate verification results and correction suggestions.

[0070] Example 3 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the intelligent generation method for power operation tickets as described in Embodiment 1 above.

[0071] Example 4 This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the intelligent generation method for power operation tickets as described in Embodiment 1 above.

[0072] The steps or modules involved in Embodiments 2 to 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligently generating power operation tickets, characterized in that, include: The system acquires unstructured natural language operation tasks from user input, inputs them into a pre-trained large model that integrates a power industry knowledge base, performs intent parsing, and extracts information on the target scenario, operation object, and operation action. Based on the operation object information, the power equipment ledger management system is invoked through the data interface to obtain the corresponding equipment data; Based on the intent parsing results and device data, multi-dimensional matching conditions are constructed, and the optimal historical operation ticket template is selected from the preset standardized operation ticket knowledge base. Perform consistency verification on the optimal historical operation ticket template, filter conflicting operation items, and generate an operation ticket to be verified; The error prevention rule base is invoked and associated with the real-time status of the device to perform multi-level in-depth verification on the operation ticket to be verified, and verification results and correction suggestions are generated.

2. The intelligent generation method for power operation tickets as described in claim 1, characterized in that, The pre-trained large model that integrates the power industry knowledge base is obtained by fine-tuning and training it by injecting the power industry operation terminology database and safety procedure knowledge base into the general natural language model. When performing intent parsing, the large model employs keyword extraction, semantic association analysis, and ambiguity elimination logic to extract information about the target scene, the object of operation, and the operation action. When missing information is detected, a user supplementary prompt mechanism is triggered.

3. The intelligent generation method for power operation tickets as described in claim 1, characterized in that, The step of obtaining corresponding equipment data by invoking the power equipment ledger management system through a data interface based on the operation object information specifically includes: Using the information of the operation object as the retrieval condition, the unique identifier PSR-ID of the device is matched and obtained in the ledger management system; Based on the PSR-ID, the topology parameters, current operating status, and historical operation records of the device are further obtained through chained queries as device data.

4. The intelligent generation method for power operation tickets as described in claim 1, characterized in that, The step involves constructing multi-dimensional matching conditions based on the intent parsing results and device data, and selecting the optimal historical operation ticket template from a pre-set standardized operation ticket knowledge base. Specifically, this includes: The multi-dimensional matching conditions include four dimensions: task type matching, device level adaptation, security rule compliance, and priority of historical execution results. Based on the matching criteria, historical templates with an execution success rate exceeding the threshold and no operational deviation records are dynamically filtered and recommended from the knowledge base.

5. The intelligent generation method for power operation tickets as described in claim 1, characterized in that, The consistency verification of the optimal historical operation ticket template specifically includes: Scene compatibility verification: compare the substation scene and equipment configuration preset in the template with the target scene information of the current task to see if they are consistent. Device attribute consistency verification compares whether the device type, topology parameters, and voltage level of the operation object in the template match the actual attributes of the current target device. Operational logic consistency verification verifies whether the operation process of the template and the operation actions and equipment operating status of the current task conform to the power safety operation logic.

6. The intelligent generation method for power operation tickets as described in claim 1, characterized in that, The conflict filtering operation items specifically include: Invalid operation item filtering identifies and removes operation steps in the template that have become invalid due to equipment updates or procedure iterations. Conflict operation filtering identifies and filters operation items that conflict with the current operating status of the device, with topology association rules, or with historical operation records, based on real-time data, topology relationships, and historical records, and supplements necessary alternative operation steps based on safety procedures.

7. The intelligent generation method for power operation tickets as described in claim 1, characterized in that, The multi-level deep verification specifically includes: Operation and status consistency verification: compare the operation instructions on the ticket with the real-time status of the device obtained from the SCADA system to see if they match. Step logic rationality verification verifies whether the order of operation steps conforms to the safe operation procedure, in order to avoid logical reversal; Rule compliance verification verifies whether each operation conforms to the device operation logic constraints, topology association taboos, and state transition rules in the error prevention rule base.

8. An intelligent power operation ticket generation system, characterized in that, include: The intent parsing module is configured to acquire unstructured natural language operation tasks input by the user, input them into a pre-trained large model that integrates the power industry knowledge base, perform intent parsing, and extract information on the target scene, operation object, and operation action. The equipment data acquisition module is configured to, based on the operation object information, call the power equipment ledger management system through a data interface to obtain the corresponding equipment data; The template filtering module is configured to construct multi-dimensional matching conditions based on intent parsing results and device data, and filter out the best historical operation ticket template from the preset standardized operation ticket knowledge base. The ticket template optimization module is configured to perform consistency verification on the optimal historical operation ticket template, filter conflicting operation items, and generate an operation ticket to be verified. The deep verification module is configured to call the anti-error rule base and associate it with the real-time status of the device to perform multi-level deep verification on the operation ticket to be verified, and generate verification results and correction suggestions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the intelligent generation method for power operation tickets as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the intelligent generation method for power operation tickets as described in any one of claims 1-7.

Images