Intelligent auxiliary compilation method and system based on large electric power safety knowledge model

Abstract

The embodiment of the invention provides an intelligent auxiliary compilation method and system based on an electric power safety knowledge large model, and belongs to the technical field of electric power auxiliary compilation. The intelligent auxiliary compilation method comprises the following steps: acquiring electric power safety knowledge data; constructing an electric power safety knowledge base according to the electric power safety knowledge data; selecting a basic large model, and finely adjusting the basic large model by adopting the electric power safety knowledge base; acquiring real-time content input by a current user; inputting the real-time content into the basic large model to obtain a compiled first draft; checking the compiled first draft, and obtaining a compiled final draft; and inputting the programmed final draft into the electric power safety knowledge base. According to the method, data updating can be effectively and continuously carried out on the electric power safety knowledge base, so that the first draft compiling accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of power auxiliary compilation technology, specifically to an intelligent auxiliary compilation method and system based on a large power safety knowledge model. Background Technology

[0002] The country attaches great importance to the development of artificial intelligence technology and is currently in a transitional period from weak AI to strong AI. Traditional small-scale AI model applications in the power industry are mainly carried out in a "small workshop" model, resulting in low R&D efficiency and insufficient adaptability to cross-business scenarios. Accelerating the development of large-scale model technology in the power industry has become an important lever for promoting the digital transformation of the power grid.

[0003] Traditional processes for compiling site survey records, work plans, and "three measures and one plan" documents are time-consuming, labor-intensive, inefficient, and wasteful of human resources. At the same time, the document compilation process involves a wide range of fields, complex stages and procedures, and a variable construction environment. The work measures are interconnected, which poses challenges and requires a high level of knowledge and experience from the compilers, making it difficult to produce high-quality documents.

[0004] In the process of realizing this invention, the inventors of this application discovered that the above-mentioned solutions in the prior art have the defects of high complexity and low efficiency in power text compilation. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent auxiliary compilation method and system based on a large power safety knowledge model, which has the function of high efficiency in power text compilation.

[0006] To achieve the above objectives, embodiments of the present invention provide an intelligent assisted compilation method based on a large-scale power safety knowledge model, comprising: Acquire electrical safety knowledge data; A power safety knowledge base is constructed based on the aforementioned power safety knowledge data; The basic large model was selected and fine-tuned using the aforementioned power safety knowledge base; Get the real-time content of the current user input; The real-time content is input into the basic large model to obtain a draft; The initial draft was reviewed, and the final draft was obtained. The compiled draft is then entered into the power safety knowledge base.

[0007] Optionally, the power safety knowledge data includes historically compiled documents, standardized operating instructions, and power safety work procedures.

[0008] Optionally, constructing a power safety knowledge base based on the power safety knowledge data includes: Preprocessing of power safety knowledge data, wherein the preprocessing includes data cleaning; The preprocessed power safety knowledge data is then divided into data segments; A power safety knowledge base is constructed based on the segmented power safety knowledge data.

[0009] Optionally, the basic large model can be fine-tuned using the power safety knowledge base, including by using a LoRA adapter.

[0010] Optionally, the underlying large model includes an NLP module.

[0011] Optionally, the real-time content is input into the basic large model to obtain the initial draft, including: Keyword extraction is performed on the real-time content to obtain multiple keywords; The task is decomposed based on the multiple keywords mentioned above, and the text content of each task is obtained separately. The text content of each task is integrated to generate a draft.

[0012] Optionally, reviewing the initial draft and obtaining the final draft includes: Determine whether the accuracy of the draft is greater than or equal to a threshold. If the accuracy of the draft is greater than or equal to the threshold, the draft is deemed valid. The initial draft is optimized to obtain the final draft; If the accuracy of the draft is determined to be less than a threshold, the draft is deemed invalid. Retrieve user input.

[0013] On the other hand, the present invention also provides an intelligent assisted compilation system based on a large-scale knowledge model of power safety, comprising: The input acquisition module is used to collect user input in real time. The controller, connected to the input acquisition module, is used to execute any of the intelligent assisted programming methods described above.

[0014] In another aspect, the present invention also provides a computer-readable storage medium storing instructions for being read by a machine to cause the machine to execute any of the above-described intelligent assisted programming methods.

[0015] Through the above technical solution, the intelligent assisted compilation method and system based on a large power safety knowledge model provided by this invention constructs a power safety knowledge base based on power safety knowledge data. Simultaneously, a basic large model is selected, and the basic large model is fine-tuned and trained using this power safety knowledge base. Combined with real-time content input by the current user, a draft can be obtained. This draft is then reviewed by personnel to obtain a finalized draft, thus achieving the purpose of assisted compilation and effectively improving the efficiency and ease of power document compilation. Finally, the finalized draft is input into the power safety knowledge base, effectively and continuously updating the data in the power safety knowledge base to improve the accuracy of the draft.

[0016] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of an intelligent assisted compilation method based on a large power safety knowledge model according to an embodiment of the present invention; Figure 2 This is a flowchart of constructing a power safety knowledge base in an intelligent assisted compilation method based on a large power safety knowledge model according to an embodiment of the present invention; Figure 3 This is a flowchart of obtaining the initial draft in an intelligent assisted compilation method based on a large power safety knowledge model according to an embodiment of the present invention; Figure 4 This is a flowchart of obtaining the final draft in an intelligent assisted compilation method based on a large power safety knowledge model according to an embodiment of the present invention. Detailed Implementation

[0018] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0019] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0020] Figure 1 This is a flowchart of an intelligent assisted compilation method based on a large-scale power safety knowledge model according to an embodiment of the present invention. Figure 1 In this context, the intelligent assisted programming method may include: In step S1, power safety knowledge data is acquired. This power safety knowledge data may include historically compiled documents, standardized operating instructions, and power safety work procedures.

[0021] In step S2, an electrical safety knowledge base is constructed based on electrical safety knowledge data.

[0022] In step S3, a basic large-scale model is selected and fine-tuned using a power safety knowledge base. This basic large-scale model may include an NLP module, i.e., a natural language processing module. Specifically, fine-tuning the basic large-scale model may include using a LoRA adapter, a method for efficient parameter fine-tuning.

[0023] In step S4, the real-time content of the current user input is obtained. The current user input may include text and voice, etc.

[0024] In step S5, the real-time content is input into the basic large model to obtain an initial draft. The fine-tuned basic large model performs semantic recognition analysis on the real-time content, which can then be used to generate individual task drafts to obtain the final draft.

[0025] In step S6, the initial draft is reviewed and the final draft is obtained. The initial draft may contain some errors and requires further review to obtain the final draft.

[0026] In step S7, the final draft is input into the power safety knowledge base. The final draft can then be used as new data to update the power safety knowledge base.

[0027] In steps S1 to S7, power safety knowledge data is first acquired, and then a power safety knowledge base is constructed based on this data. Simultaneously, a basic large-scale model is selected, and this power safety knowledge base is used to fine-tune and train the model. Combined with real-time user input, a draft is generated. This draft is then reviewed by personnel to obtain a final draft. Finally, the power safety knowledge base is updated based on the final draft to continuously fine-tune the basic large-scale model.

[0028] Traditional methods for compiling on-site survey records and work plans are time-consuming, labor-intensive, inefficient, and wasteful of human resources. Furthermore, the process involves a wide range of areas, complex stages and procedures, variable construction environments, and interconnected operational measures, demanding high levels of knowledge and experience from the compilers, which is detrimental to producing high-quality documents. In this embodiment of the invention, a method of fine-tuning the basic model of the power safety knowledge base is used to assist in the compilation of power-related documents, improving efficiency and convenience. Moreover, as the power safety knowledge base is continuously updated, the accuracy of the initial drafts increases.

[0029] In this embodiment of the invention, after obtaining power safety knowledge data, a power safety knowledge base can be constructed based on the power safety knowledge data. The specific construction steps are as follows: Figure 2 As shown. Specifically, in Figure 2 Furthermore, this intelligent assisted programming method may also include: In step S20, the power safety knowledge data is preprocessed, including data cleaning.

[0030] In step S21, the preprocessed power safety knowledge data is segmented. This segmentation can be based on localized information such as the power grid structure, equipment models, and site environment of different regions.

[0031] In step S22, an electrical safety knowledge base is constructed based on the segmented electrical safety knowledge data.

[0032] In this embodiment of the invention, after obtaining the real-time content input by the user, the real-time content needs to be input into the fine-tuned basic large model to obtain the initial draft. The specific steps are as follows: Figure 3 As shown. Specifically, in Figure 3 Furthermore, this intelligent assisted programming method may also include: In step S50, keywords are extracted from the real-time content to obtain multiple keywords. These keywords may include key parameters such as "voltage level," "power outage status," and "operation type."

[0033] In step S51, the task is decomposed based on multiple keywords, and the text content of each task is obtained. Specifically, the job task can be decomposed into specific, executable logical subtasks.

[0034] In step S52, the text content of each task is integrated to generate a draft.

[0035] In steps S50 to S52, the fine-tuned large model, for the formation of the initial draft, first performs semantic understanding and context construction on the real-time content input by the user, which may include keyword identification and context understanding. Then, the task is decomposed and logically reasoned, breaking down the macro-objective into a sequence of executable micro-operations, and identifying the temporal dependencies and logical constraints between tasks. Simultaneously, each decomposed sub-task is matched against a power safety knowledge base to obtain relevant safety risks and specific content. Finally, the text content of each task is integrated to obtain the initial draft.

[0036] In this embodiment of the invention, after obtaining the initial draft, it is necessary to review the initial draft to obtain the final draft. The specific steps can be as follows: Figure 4 As shown. Specifically, in Figure 4 Furthermore, this intelligent assisted programming method may also include: In step S60, it is determined whether the accuracy of the draft is greater than or equal to a threshold. The accuracy of the draft can be determined by the completion rate and the accuracy rate. The completion rate can be determined based on the percentage of blank content, while the accuracy rate can be obtained through manual scoring. The final accuracy is obtained by averaging the two.

[0037] In step S61, if the accuracy of the initial draft is greater than or equal to a threshold, the initial draft is deemed valid. Specifically, if the accuracy of the initial draft is high, it indicates that the initial draft generated with the assistance of the large model is valid and can be used.

[0038] In step S62, the initial draft is optimized to obtain the final draft. This optimization can be achieved through expert review and modification to arrive at the final draft.

[0039] In step S63, if the accuracy of the initial draft is less than a threshold, the initial draft is deemed invalid. Specifically, if the accuracy of the initial draft is low, it means that the initial draft generated with the assistance of the large model is essentially unusable and needs to be regenerated.

[0040] In step S64, the user's input is retrieved again.

[0041] In steps S60 to S64, after obtaining the initial draft, the accuracy of the initial draft can be obtained. Determining this accuracy can include methods such as manual scoring. Simultaneously, a certain threshold is set, and the accuracy of the initial draft is compared to the threshold. If the accuracy is high, simple manual modifications can be made to generate the final draft; otherwise, the user's input needs to be re-acquired and recognized, and a new initial draft needs to be generated.

[0042] On the other hand, the present invention also provides an intelligent auxiliary compilation system based on a large power safety knowledge model. Specifically, the intelligent auxiliary compilation system may include an input acquisition module and a controller.

[0043] The input acquisition module is used to collect user input in real time. The controller is connected to the input acquisition module and is used to execute any of the intelligent auxiliary programming methods mentioned above.

[0044] In another aspect, the present invention also provides a computer-readable storage medium storing instructions for being read by a machine to cause the machine to execute any of the above-mentioned intelligent assisted programming methods.

[0045] Through the above technical solution, the intelligent assisted compilation method and system based on a large power safety knowledge model provided by this invention constructs a power safety knowledge base based on power safety knowledge data. Simultaneously, a basic large model is selected, and the basic large model is fine-tuned and trained using this power safety knowledge base. Combined with real-time content input by the current user, a draft can be obtained. This draft is then reviewed by personnel to obtain a finalized draft, thus achieving the purpose of assisted compilation and effectively improving the efficiency and ease of power document compilation. Finally, the finalized draft is input into the power safety knowledge base, effectively and continuously updating the data in the power safety knowledge base to improve the accuracy of the draft.

[0046] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0047] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0048] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0049] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0050] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0051] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0052] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0053] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0054] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. An intelligent assisted compilation method based on a large-scale knowledge model of power safety, characterized in that, include: Acquire electrical safety knowledge data; A power safety knowledge base is constructed based on the aforementioned power safety knowledge data; The basic large model was selected and fine-tuned using the aforementioned power safety knowledge base; Get the real-time content of the current user input; The real-time content is input into the basic large model to obtain a draft; The initial draft was reviewed, and the final draft was obtained. The compiled draft is then entered into the power safety knowledge base.

2. The intelligent assisted coding method according to claim 1, characterized in that, The power safety knowledge data includes historically compiled documents, standardized operating instructions, and power safety work procedures.

3. The intelligent assisted coding method according to claim 1, characterized in that, The construction of a power safety knowledge base based on the aforementioned power safety knowledge data includes: Preprocessing of power safety knowledge data, wherein the preprocessing includes data cleaning; The preprocessed power safety knowledge data is then divided into data segments; A power safety knowledge base is constructed based on the segmented power safety knowledge data.

4. The intelligent assisted coding method according to claim 1, characterized in that, The basic large model is selected and fine-tuned using the power safety knowledge base, including the use of LoRA adapters.

5. The intelligent assisted compilation method according to claim 1, characterized in that, The basic large model includes an NLP module.

6. The intelligent assisted coding method according to claim 1, characterized in that, The real-time content is input into the basic large model to obtain the initial draft, including: Keyword extraction is performed on the real-time content to obtain multiple keywords; The task is decomposed based on the multiple keywords mentioned above, and the text content of each task is obtained separately. The text content of each task is integrated to generate a draft.

7. The intelligent assisted coding method according to claim 6, characterized in that, The initial draft was reviewed, and the final draft was obtained, including: Determine whether the accuracy of the draft is greater than or equal to a threshold. If the accuracy of the draft is greater than or equal to the threshold, the draft is deemed valid. The initial draft is optimized to obtain the final draft; If the accuracy of the draft is determined to be less than a threshold, the draft is deemed invalid. Retrieve user input.

8. An intelligent assisted compilation system based on a large-scale knowledge model of power safety, characterized in that, include: The input acquisition module is used to collect user input in real time. The controller, connected to the input acquisition module, is used to execute the intelligent assisted programming method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that are read by a machine to cause the machine to perform the intelligent assisted compilation method as described in any one of claims 1-7.

Images