Power grid project processing method and device, computer equipment and storage medium
By combining the power grid knowledge base and the target language model, the phase identification, anomaly analysis and knowledge generation of power grid projects is solved, and the problems of fragmentation of knowledge and insufficient reasoning capabilities in traditional power grid project management are improved, and the accuracy and coordination of project management are improved. Regularity.
Patent Information
- Application Number
- CN202411270097.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-06
AI Technical Summary
In traditional power grid project management methods, the knowledge of the knowledge graph is too fragmented and fragmented, and has weak reasoning ability, which cannot meet the needs of precise decision-making, and it is difficult to effectively utilize the professional knowledge and historical experience of the power grid industry.
Retrieval and enhanced generation processing based on the power grid knowledge base and target language model are adopted, and project knowledge information and system information are generated through project stage identification, abnormal analysis and project analysis to improve the accuracy of project management.
Through the combination of the target language model and the grid knowledge base, the problem of insufficient inference ability of the grid knowledge base in complex text tasks is compensated, the accuracy of project knowledge information and system information is improved, the compliance of project application text is ensured, and the compliance of project application text is optimized. Grid project management.
Smart Images

Figure CN119090451A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of smart grid technology, and in particular to a method, device, computer equipment, storage medium and computer program product for processing a power grid project. Background Art
[0002] With the rapid development of power grid technology and the advancement of smart grid construction, power grid projects are increasing in number and complexity. The strict regulatory environment of the power industry also requires project management systems to respond to policy changes in a timely manner to ensure compliance. At the same time, the power grid industry has accumulated a lot of professional knowledge and historical experience. How to effectively use these resources has become the key to improving the level of power grid project management.
[0003] In traditional technologies, project management experience is often provided through knowledge graphs. However, the knowledge in the knowledge graph is too fragmented and scattered, and the reasoning ability of the knowledge graph is weak, and it is unable to derive deeper knowledge when dealing with complex tasks. Therefore, traditional power grid project management methods can no longer meet the needs of accurate decision-making. Summary of the invention
[0004] Based on this, it is necessary to provide a power grid project processing method, device, computer equipment, computer readable storage medium and computer program product that can improve the management accuracy of power grid projects in response to the above technical problems.
[0005] In a first aspect, the present application provides a method for processing a power grid project. The method comprises:
[0006] Performing project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text;
[0007] Based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain project knowledge information and project system information corresponding to the project application text;
[0008] According to the project system information, an abnormal analysis process is performed on the project application text to obtain an abnormal analysis result of the project application text;
[0009] Through the target large language model, project analysis processing is performed on the project application text, the project knowledge information and the project system information to obtain project analysis information of the power grid project;
[0010] According to the abnormal analysis result and the project analysis information, target management information of the power grid project is obtained.
[0011] In one embodiment, based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain project knowledge information and project system information corresponding to the project application text, including:
[0012] In the power grid knowledge base, similarity search is performed on the project phase identification information and the project application text to obtain candidate knowledge information and candidate system information that match the project phase identification information and the project application text;
[0013] The candidate knowledge information and the candidate system information are respectively integrated with the project application text to obtain integrated knowledge information and integrated system information;
[0014] The fused knowledge information and the fused system information are respectively input into the target large language model to obtain the project knowledge information and the project system information associated with the project application text.
[0015] In one embodiment, the project application text, the project knowledge information and the project system information are subjected to project analysis processing by the target large language model to obtain project analysis information of the power grid project, including:
[0016] Obtaining prompt words of the project application text according to the project application text and the project knowledge information;
[0017] The prompt words, the project application text, the project knowledge information and the project system information are input into the target large language model to obtain problem analysis information of the project application text; the target large language model is obtained by fine-tuning the pre-trained large language model according to the historical project knowledge information and the historical project application text;
[0018] Generate improvement suggestion information for the problem analysis information based on the project system information;
[0019] According to the problem analysis information and the improvement suggestion information, project analysis information of the power grid project is obtained.
[0020] In one embodiment, the prompt word, the project application text and the project knowledge information are input into the target large language model to obtain problem analysis information of the project application text, including:
[0021] Obtaining current project management data of the project application text through a data interaction interface related to the power grid project;
[0022] The current project management data, the prompt words, the project declaration text and the project knowledge information are input into the target large language model to obtain problem analysis information of the project declaration text.
[0023] In one embodiment, according to the project system information, the project application text is subjected to abnormal analysis processing to obtain abnormal analysis results of the project application text, including:
[0024] According to the project system information, text matching processing is performed on the project application text to obtain the text of illegal content in the project application text;
[0025] Generate a violation content identifier and violation description information corresponding to the violation content text;
[0026] The abnormality analysis result is obtained according to the illegal content identifier and the illegal description information.
[0027] In one embodiment, a project phase identification process is performed on a project application text of a power grid project to obtain project phase identification information of the project application text, including:
[0028] Performing keyword extraction processing on the project application text of the power grid project to obtain keyword information of the project application text;
[0029] According to a preset stage feature library, feature matching processing is performed on the keyword information to obtain a target stage feature that matches the project application text; the stage feature library stores a plurality of stage features and stage identification information corresponding to the stage features;
[0030] The stage identification information corresponding to the target stage feature is set as the project stage identification information of the project declaration text.
[0031] In a second aspect, the present application also provides a power grid project processing device. The device comprises:
[0032] A phase identification module is used to perform project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text;
[0033] A retrieval enhancement module is used to perform retrieval enhancement generation processing on the project phase identification information and the project application text based on the power grid knowledge base and the target large language model to obtain project knowledge information and project system information corresponding to the project application text;
[0034] An abnormality analysis module is used to perform abnormality analysis on the project application text according to the project system information to obtain abnormality analysis results of the project application text;
[0035] A project analysis module, used to perform project analysis processing on the project application text, the project knowledge information and the project system information through the target large language model to obtain project analysis information;
[0036] A management determination module is used to obtain target management information of the power grid project based on the abnormal analysis result and the project analysis information.
[0037] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:
[0038] Performing project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text;
[0039] Based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain project knowledge information and project system information corresponding to the project application text;
[0040] According to the project system information, an abnormal analysis process is performed on the project application text to obtain an abnormal analysis result of the project application text;
[0041] Through the target large language model, project analysis processing is performed on the project application text, the project knowledge information and the project system information to obtain project analysis information of the power grid project;
[0042] According to the abnormal analysis result and the project analysis information, target management information of the power grid project is obtained.
[0043] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:
[0044] Performing project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text;
[0045] Based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain project knowledge information and project system information corresponding to the project application text;
[0046] According to the project system information, an abnormal analysis process is performed on the project application text to obtain an abnormal analysis result of the project application text;
[0047] Through the target large language model, project analysis processing is performed on the project application text, the project knowledge information and the project system information to obtain project analysis information of the power grid project;
[0048] According to the abnormal analysis result and the project analysis information, target management information of the power grid project is obtained.
[0049] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:
[0050] Performing project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text;
[0051] Based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain project knowledge information and project system information corresponding to the project application text;
[0052] According to the project system information, an abnormal analysis process is performed on the project application text to obtain an abnormal analysis result of the project application text;
[0053] Through the target large language model, project analysis processing is performed on the project application text, the project knowledge information and the project system information to obtain project analysis information of the power grid project;
[0054] According to the abnormal analysis result and the project analysis information, target management information of the power grid project is obtained.
[0055] The above-mentioned power grid project processing method, device, computer equipment, storage medium and computer program product perform project stage identification processing on the project application text of the power grid project to obtain the project stage identification information of the project application text; based on the power grid knowledge base and the target large language model, perform retrieval enhancement generation processing on the project stage identification information and the project application text to obtain the project knowledge information and project system information corresponding to the project application text; perform abnormal analysis processing on the project application text according to the project system information to obtain the abnormal analysis results of the project application text; perform project analysis processing on the project application text, project knowledge information and project system information through the target large language model to obtain project analysis information; obtain the target management information of the power grid project according to the abnormal analysis results and the project analysis information. By adopting this method, through combining the target large language model with the power grid knowledge base, the text processing advantage of the target large language model can be used to make up for the weak reasoning ability of the power grid knowledge base for complex text tasks, thereby improving the accuracy of the processed project knowledge information and project system information. The project system information is also used to analyze whether there are abnormal situations of illegal content in the project application text, and then the abnormal analysis results and project analysis information are coordinated to comprehensively determine the target management information of the power grid project, further improving the accuracy of the processed target management information, so as to optimize the management of the power grid project based on the target management information, thereby improving the management accuracy of the power grid project. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 A process environment diagram of a power grid project processing method in one embodiment;
[0057] Figure 2 A flowchart of the steps of performing retrieval, enhancement and generation processing on project phase identification information and project application text in one embodiment;
[0058] Figure 3 A flowchart of project analysis and processing steps for project application text, project knowledge information and project system information in one embodiment;
[0059] Figure 4 A schematic diagram of a flow chart of a method for processing a power grid project in another embodiment;
[0060] Figure 5 is a structural block diagram of a power grid project processing device in one embodiment;
[0061] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0062] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0063] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0064] In one embodiment, Figure 1 As shown, a method for processing power grid projects is provided. This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:
[0065] Step S101, performing project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text.
[0066] Among them, power grid projects refer to innovative projects in the power grid industry.
[0067] Among them, the project application document refers to the formal document used to apply for resource support from relevant departments, organizations or institutions for real-time power grid projects. In practical applications, the project application document can be a project application for a power grid innovation project.
[0068] Among them, the project stage identification information refers to the identification information used to reflect the current declaration stage of the power grid project.
[0069] Specifically, the terminal displays a preset project management page in the interface, and obtains the project application text of the power grid project uploaded by the user in the project management page. The terminal identifies the current project application stage of the project application text based on the key information in the project application text, and obtains the project stage identification information of the project application text.
[0070] In actual applications, a title is displayed at the top of the project management page, such as "Power Grid Innovation Project Intelligent Management Assistant". The left side of the project management page displays the upload area for the project application. The middle of the project management page displays the content display area of the project application, and supports text highlighting. The right side of the project management page displays the display area for the target management information of the project application.
[0071] Step S102, based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain the project knowledge information and project system information corresponding to the project application text.
[0072] Among them, the power grid knowledge base refers to a knowledge base that stores a large amount of knowledge, rules and regulations and other information related to power grid projects.
[0073] Among them, the target large language model refers to a large language model used to analyze and process text data related to power grid projects. Large language model (LLM) is a natural language processing model based on deep learning. Large language model is different from ordinary natural language processing (NLP) model. Large language model is usually larger in scale, has more parameters, and has better text processing capabilities (such as text generation and semantic understanding). It also has the characteristics of self-supervised learning (mainly trained through self-supervised tasks such as predicting the next word), versatility (can perform a variety of NLP tasks, such as text generation, question answering, translation, etc.), few-sample learning (can quickly adapt to new tasks with a small number of samples) and contextual understanding (can understand the context of long texts).
[0074] Among them, project knowledge information refers to the knowledge related to the project application text. For example, project knowledge information can be procedural knowledge (such as operating steps, workflow), technical knowledge, factual knowledge (such as statistical data, factual events), and conceptual knowledge (such as definitions, categories) related to the project application text.
[0075] The project system information refers to information describing the rules and regulations of the power grid project. For example, the project system information may be a rules and regulations document of the power grid project.
[0076] Specifically, the terminal can collect information such as knowledge, rules and regulations related to the power grid project in advance, and then convert these information into vector form for storage, so that the terminal constructs a power grid knowledge base. The terminal can also use the knowledge, rules and regulations related to the power grid project to fine-tune the pre-trained large language model to obtain a large language model that can more accurately analyze and process power grid-related information, and then the terminal fine-tunes to obtain the target large language model. The terminal combines the power grid knowledge base and the target large language model to perform retrieval enhancement and generation processing on the project stage identification information and project application text, so as to make up for the limitation of the weak reasoning ability of the power grid knowledge base when processing complex text tasks with the help of the target large language model's learning knowledge and excellent text processing capabilities in power grid projects. Finally, the terminal processes the project knowledge information and project system information corresponding to the project application text.
[0077] It should be noted that Retrieval-Augmented Generation (RAG) is an artificial intelligence technology that combines information retrieval and text generation. RAG technology is different from conventional knowledge base retrieval technology. RAG technology mainly covers two stages: "Retrieval" and "Generation". Retrieval refers to the process of extracting relevant information from a large-scale knowledge base, while generation uses the retrieved information to create responses or answer questions. The core advantage of RAG technology is that it can dynamically access and utilize a large amount of structured or unstructured information, thereby improving the accuracy, relevance and timeliness of the generated content, making RAG technology show great potential in smart grid applications. It can effectively combine massive data with powerful language comprehension capabilities to provide more accurate and informative project knowledge information and project system information to grid project stakeholders.
[0078] Step S103, performing abnormal analysis on the project application text according to the project system information to obtain abnormal analysis results of the project application text.
[0079] Among them, the abnormal analysis results are used to characterize whether there are any illegal contents in the project application text and the specific description of the illegal contents.
[0080] Specifically, based on the project system information, the terminal identifies whether there is potential illegal content in the project application text, and then the terminal identifies the abnormal analysis result of the project application text. If the abnormal analysis result indicates that the illegal content is empty, it means that there is no illegal content in the project application text. If the abnormal analysis result indicates that the illegal content is not empty, it means that there is illegal content in the project application text.
[0081] Step S104, using the target large language model, performing project analysis on the project application text, project knowledge information and project system information to obtain project analysis information.
[0082] Specifically, the terminal can also obtain real-time current project management data of the project application text through a data interaction interface related to the power grid project; then the terminal inputs the current project management data, project application text, project knowledge information and project system information into the target large language model, so that the target large language model can combine the real-time current project management data, project knowledge information and project system information to comprehensively perform content analysis on the project application text, and output project analysis information.
[0083] Step S105, obtaining target management information of the power grid project according to the abnormal analysis result and the project analysis information.
[0084] The target management information is used to indicate the optimization of the management mode of the power grid project. For example, the target management information may include technical route optimization information and budget optimization information for the power grid project.
[0085] Specifically, the terminal obtains abnormal analysis results, project analysis information, and other analysis information, and classifies and integrates each analysis information to generate a structured management analysis report, that is, the terminal obtains the target management information of the power grid project. The terminal displays the target management information on the project management page for users to view. In the process of displaying the target management information, the terminal can also highlight the problematic areas in the project application text.
[0086] For example, the above-mentioned power grid project processing method can be used to optimize the management of power grid projects. Suppose a project team is preparing a project application for a power grid innovation project. If the terminal recognizes that the power grid innovation project is currently in the project establishment stage based on the project application text uploaded by the project team, it uses the power grid knowledge base and the target large language model to retrieve relevant project establishment requirements and successful cases; then analyze the content of the project application text, and compare it with industry standards to provide specific improvement suggestions, such as improving the technical route, optimizing the budget, etc., and finally the terminal outputs the target management information of the power grid innovation project. An example of the output of the target management information is as follows: "It is recommended to add a detailed description of the technical innovation points in Chapter X of the project application text, and refer to the successful case of the XX project. The budget part should be detailed to the specific equipment procurement list to meet the requirements of Article X of the "XXX Project System"."
[0087] To give another example, the above-mentioned power grid project processing method can also be used for the management of cross-grid projects. Suppose a user needs to query relevant experience from multiple historical projects. The terminal receives the user's query request, retrieves relevant information from the knowledge base, and generates a comprehensive analysis report based on the lessons learned from multiple historical projects, and uses the comprehensive analysis report as the target management information for multiple historical projects. An output example of the comprehensive analysis report is as follows: "Based on your query, we found that there were 5 related projects involving similar technologies in the past three years. Among them, Project A and Project B successfully overcame the technical difficulties you faced. It is recommended that you focus on the following experiences: 1) ... 2) ... 3) ... A detailed technical solution comparison analysis has been generated, please see the attachment."
[0088] In the above-mentioned power grid project processing method, the project application text of the power grid project is subjected to project phase identification processing to obtain the project phase identification information of the project application text; based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are subjected to retrieval enhancement generation processing to obtain the project knowledge information and project system information corresponding to the project application text; according to the project system information, the project application text is subjected to abnormal analysis processing to obtain the abnormal analysis result of the project application text; through the target large language model, the project application text, the project knowledge information and the project system information are subjected to project analysis processing to obtain the project analysis information; according to the abnormal analysis result and the project analysis information, the target management information of the power grid project is obtained. By adopting this method, through combining the target large language model with the power grid knowledge base, the text processing advantage of the target large language model can be used to make up for the weak reasoning ability of the power grid knowledge base for complex text tasks, thereby improving the accuracy of the processed project knowledge information and project system information. The project system information is also used to analyze whether there are abnormal situations of illegal content in the project application text, and then the abnormal analysis results and project analysis information are coordinated to comprehensively determine the target management information of the power grid project, further improving the accuracy of the processed target management information, so as to optimize the management of the power grid project based on the target management information, thereby improving the management accuracy of the power grid project.
[0089] In one embodiment, Figure 2 As shown, the above step S102, based on the power grid knowledge base and the target large language model, performs retrieval enhancement generation processing on the project phase identification information and the project application text to obtain the project knowledge information and project system information corresponding to the project application text, which specifically includes the following contents:
[0090] Step S201 : performing similarity search on the project phase identification information and the project application text in the power grid knowledge base to obtain candidate knowledge information and candidate system information matching the project phase identification information and the project application text.
[0091] Specifically, the terminal can convert the project phase identification information into a project phase identification vector, and convert the project declaration text information into a project declaration text vector, and then construct a search formula based on the project phase identification vector and the project declaration text vector; and then the terminal uses the search formula to perform similarity search in the power grid knowledge base to retrieve candidate knowledge vectors and candidate system vectors that match the search formula, for example, the candidate knowledge vectors and candidate system vectors that have the highest similarity to the search formula can be retrieved. The terminal can also restore the candidate knowledge vector in vector form to the candidate knowledge information in the original data form, and restore the candidate system vector in vector form to the candidate system information in the original data form.
[0092] Step S202, the candidate knowledge information and the candidate system information are respectively integrated with the project application text to obtain integrated knowledge information and integrated system information respectively.
[0093] Specifically, the terminal integrates the candidate knowledge information with the project application text to obtain integrated knowledge information; similarly, the terminal integrates the candidate system information with the project application text to obtain integrated system information.
[0094] Step S203, inputting the fused knowledge information and the fused system information into the target large language model respectively, and obtaining the project knowledge information and the project system information associated with the project application text respectively.
[0095] Specifically, the terminal inputs the fused knowledge information into the target large language model to generate project knowledge information that is more strongly related to the project application text and more accurate through the target large language model; similarly, the terminal inputs the fused system information into the target large language model to generate project system information that is more strongly related to the project application text and more accurate through the target large language model.
[0096] In this embodiment, by performing similarity search on the project stage identification information and the project application text in the power grid knowledge base, candidate knowledge information and candidate system information that match the project stage identification information and the project application text can be preliminarily retrieved, and then the candidate knowledge information and candidate system information are respectively fused with the project application text, so that the fused knowledge information and fused system information can highlight the content that is more relevant to the power grid project, and then the fused knowledge information and fused system information are respectively input into the target large language model, so as to output project knowledge information and project system information that are more closely related to the project application text and more accurate through the target large language model, thereby effectively improving the information quality of the acquired project knowledge information and project system information, and facilitating improving the accuracy of the subsequent analysis and processing based on the project knowledge information and project system information, and based on the professionalism of the power grid knowledge base, effectively ensuring that the generated target management information has a high degree of relevance to the power grid field.
[0097] In one embodiment, Figure 3 As shown, in the above step S104, the project application text, project knowledge information and project system information are subjected to project analysis processing by the target large language model to obtain project analysis information, which specifically includes the following contents:
[0098] Step S301, obtaining prompt words of the project application text according to the project application text and project knowledge information.
[0099] Specifically, the terminal constructs prompt words for the project application text based on the project application text and the context information of the project application text, as well as the project knowledge information and the context information of the project knowledge information, so as to assist the target large language model to more accurately understand the textual intent of the project application text and the project knowledge information through the prompt words.
[0100] Step S302, input prompt words, project application text, project knowledge information and project system information into the target large language model to obtain problem analysis information of the project application text.
[0101] Among them, the target large language model is obtained by fine-tuning the pre-trained large language model according to the historical project knowledge information and the historical project application text.
[0102] Specifically, before the above step S101, the terminal can first obtain a pre-trained large language model, and then use the historical project knowledge information and the historical project application text to fine-tune the model parameters of the pre-trained large language model, so that the trained target large language model can more accurately process the text data related to the power grid project. After obtaining the prompt words of the project application text in the above step S301, the terminal can input the prompt words, the project application text, the project knowledge information and the project system information into the target large language model. The target large language model can perform semantic understanding of the project application text based on the prompt words, and perform problem analysis on the project application text in combination with the project knowledge information and the project system information, so as to analyze the potential problems in the project application text, and output the problem analysis information of the project application text.
[0103] Step S303: Generate improvement suggestion information for the problem analysis information according to the project system information.
[0104] Specifically, the target large language model can also generate improvement suggestion information corresponding to the problem analysis information based on the project system information, wherein the improvement suggestion information is used to represent the improvement plan proposed for the problem described in the problem analysis information.
[0105] Step S304: obtaining project analysis information of the power grid project based on the problem analysis information and improvement suggestion information.
[0106] Specifically, the terminal uses the problem analysis information and the improvement suggestion information as the project analysis information of the project application text.
[0107] In this embodiment, prompt words are constructed based on the project application text and project knowledge information, and the prompt words are used to assist the target large language model to fully understand the semantics of the project application text, and then the problem analysis information of the project application text is output, and then combined with the project system information, improvement suggestion information for the problem analysis information is generated, and the problems and improvement plans of the power grid project are comprehensively described through the problem analysis information and the improvement suggestion information, which effectively improves the comprehensiveness and accuracy of the project analysis information of the power grid project.
[0108] In one embodiment, the above step S302 inputs prompt words, project application text and project knowledge information into the target large language model to obtain problem analysis information of the project application text, which specifically includes the following contents: obtaining the current project management data of the project application text through a data interaction interface related to the power grid project; inputting the current project management data, prompt words, project application text and project knowledge information into the target large language model to obtain problem analysis information of the project application text.
[0109] Among them, the current project management data refers to the real-time project management data of the power grid project.
[0110] It should be noted that in order to solve the problem of inconsistent data formats and interfaces of multi-source heterogeneous data in the power grid, which leads to a long data extraction and fusion process and low efficiency in real-time decision-making, the terminal has constructed data interaction interfaces with different request types related to the power grid project. Data interaction interfaces with different request types are used to provide different types of data to the terminal.
[0111] Specifically, an executor is deployed in the terminal, and the executor can receive the operation instructions generated by the terminal, parse the instruction content and generate a data request, and then send the data request to the corresponding data interaction interface according to the request type of the data request. The data interaction interface receives the data request from the executor, constructs an API (Application Programming Interface) call parameter for the data interaction interface, calls the real-time data corresponding to the data request in the specified business system through the API call parameter, performs necessary format conversion and data cleaning processing on the real-time data, and returns the processed data as the current project management data to the executor. The executor captures the current project management data returned by the data interaction interface and returns the current project management data to the terminal. Then, the terminal can input the current project management data, prompt words, project application text and project knowledge information into the target large language model, so that the target large language model can analyze the problem of analyzing the project application text in combination with the real-time current project management data of the power grid project, and output the problem analysis information of the project application text.
[0112] In actual applications, the executor can also capture the execution results or exceptions corresponding to the operation instructions, and report the execution results and exceptions to the terminal.
[0113] In this embodiment, the current project management data of the project application text is acquired in real time by integrating the data interaction interface related to the power grid project, and then the target large language model combines the real-time current project management data of the power grid project to analyze the project application text for problems, providing the target large language model with richer high-quality multi-source data, thereby greatly improving the comprehensiveness and accuracy of the target large language model in problem analysis and improvement suggestions, and also allowing the target large language model to continuously learn the real-time data of the power grid project, continuously accumulate knowledge for the target large language model, and achieve continuous performance optimization and experience inheritance of the target large language model, significantly improving the management capabilities of the power grid project.
[0114] In one embodiment, the above step S103 performs abnormal analysis on the project application text according to the project system information to obtain the abnormal analysis result of the project application text, which specifically includes the following contents: performing text matching on the project application text according to the project system information to obtain the illegal content text in the project application text; generating the illegal content identification and illegal description information corresponding to the illegal content text; obtaining the abnormal analysis result according to the illegal content identification and illegal description information.
[0115] The illegal content identification refers to the type of illegal content indicated by identification information, and the illegal description information is used to describe the specific situation of the illegal content.
[0116] Specifically, based on the project system information, the terminal uses natural language processing technology to perform text matching processing on the project application text to detect whether there is text content matching the project system information in the project application text, and the terminal obtains the system matching text; then, the terminal uses natural language processing technology to perform semantic recognition processing on the system matching text to identify whether there is illegal content in the system matching text, and the terminal obtains the illegal content text in the project application text. If the illegal content text in the project application text is empty, the illegal content identification and illegal description information corresponding to the illegal content text are also empty; if the illegal content text in the project application text is not empty, the illegal content identification and illegal description information corresponding to the illegal content text are generated; the illegal content identification and illegal description information are used as the abnormal analysis results of the project application text.
[0117] In this embodiment, the project system information is used to analyze whether the content in the project application text complies with the provisions of relevant rules and regulations, and the abnormal analysis results of the project application text are obtained, which effectively ensures that the generated target management information complies with the relevant rules and regulations of the power grid project and reduces the risk of violations of the power grid project.
[0118] In one embodiment, the above step S101, performing project phase identification processing on the project application text of the power grid project to obtain project phase identification information of the project application text, specifically includes the following contents: performing keyword extraction processing on the project application text of the power grid project to obtain keyword information of the project application text; performing feature matching processing on the keyword information according to a preset phase feature library to obtain a target phase feature that matches the project application text; the phase feature library stores a plurality of phase features and phase identification information corresponding to the phase features; and setting the phase identification information corresponding to the target phase feature as the project phase identification information of the project application text.
[0119] Among them, the stage feature library stores the feature information of different declaration stages (ie, stage features) and the identification information corresponding to each stage feature (ie, stage identification information).
[0120] Specifically, the terminal can use natural language processing technology to perform keyword extraction processing on the project application text of the power grid project, extract keywords, key sentences and other contents in the project application text, and set them as keyword information. The terminal performs feature matching processing on the keyword information and the phase features in the pre-built phase feature library, obtains the phase features in the phase feature library that match the keyword information, and sets them as the target phase features. The terminal obtains the phase identification information corresponding to the target phase feature and sets it as the project phase identification information of the project application text.
[0121] In this embodiment, based on the keyword information of the project application text, the target stage feature that matches the keyword information in the stage feature library is matched; then the stage identification information corresponding to the target stage feature is set as the project stage identification information of the project application text, thereby realizing automatic identification of the project stage corresponding to the project application text, so that in subsequent steps, the project knowledge information that matches the project stage identification information can be accurately obtained from the power grid knowledge base based on the project stage identification information, thereby improving the accuracy of analysis and processing performed by the target large language model, and providing accurate and reliable analysis results and improvement suggestions for the management of power grid projects.
[0122] In one embodiment, Figure 4 As shown, another power grid project processing method is provided, which is described by taking the method applied to a terminal as an example, and includes the following steps:
[0123] Step S401: perform keyword extraction processing on the project application text of the power grid project to obtain keyword information of the project application text.
[0124] Step S402, performing feature matching processing on the keyword information according to a preset stage feature library to obtain a target stage feature that matches the project application text; and setting the stage identification information corresponding to the target stage feature as the project stage identification information of the project application text.
[0125] Step S403: perform similarity search on the project phase identification information and the project application text in the power grid knowledge base to obtain candidate knowledge information and candidate system information matching the project phase identification information and the project application text.
[0126] Step S404, the candidate knowledge information and the candidate system information are respectively integrated with the project application text to obtain integrated knowledge information and integrated system information respectively.
[0127] Step S405, inputting the fused knowledge information and the fused system information into the target large language model respectively, and obtaining the project knowledge information and the project system information associated with the project application text respectively.
[0128] Step S406, according to the project system information, text matching processing is performed on the project application text to obtain the illegal content text in the project application text; illegal content identification and illegal description information corresponding to the illegal content text are generated; according to the illegal content identification and illegal description information, an abnormal analysis result is obtained.
[0129] Step S407, obtaining prompt words of the project application text according to the project application text and the project knowledge information.
[0130] Step S408, obtaining the current project management data of the project application text through the data interaction interface related to the power grid project.
[0131] Step S409, input the current project management data, prompt words, project application text and project knowledge information into the target large language model to obtain problem analysis information of the project application text.
[0132] Step S410, generating improvement suggestion information for the problem analysis information according to the project system information; obtaining project analysis information of the power grid project according to the problem analysis information and the improvement suggestion information.
[0133] Step S411, obtaining target management information of the power grid project according to the abnormal analysis result and the project analysis information.
[0134] The above-mentioned power grid project processing method can achieve the following beneficial effects: by combining the target large language model and the power grid knowledge base, the text processing advantages of the target large language model can be used to make up for the weak reasoning ability of the power grid knowledge base for complex text tasks, thereby improving the accuracy of the processed project knowledge information and project system information, and also using the project system information to analyze whether there are abnormal situations of illegal content in the project application text, and then coordinate the abnormal analysis results and project analysis information to comprehensively determine the target management information of the power grid project, further improving the accuracy of the processed target management information, so as to optimize the management of the power grid project based on the target management information, thereby improving the management accuracy of the power grid project.
[0135] In order to more clearly illustrate the power grid project processing method provided by the embodiment of the present disclosure, the above power grid project processing method is specifically described below with a specific embodiment. Another power grid project processing method is provided, which can be applied to a terminal, and specifically includes the following contents:
[0136] The terminal is equipped with a user interface module, a project stage identification module, a RAG retrieval module, a large language model module, a business system interface module, an actuator module, a rules and regulations analysis module, a declaration analysis module and a result integration module. The specific workflow is as follows:
[0137] 1) The power grid innovation project application form uploaded by the user through the terminal interface;
[0138] 2) The project phase identification module identifies the current project phase of the project proposal;
[0139] 3) The RAG search module searches for relevant knowledge and rules and regulations based on the project stage and the content of the application;
[0140] 4) The executor module calls the business system interface to obtain real-time current project management data;
[0141] 5) The rules and regulations analysis module conducts abnormal and illegal analysis on the project application form;
[0142] 6) The application analysis module combines the target large language model to conduct in-depth analysis of the content of the project application;
[0143] 7) The result integration module summarizes the analysis results of each module;
[0144] 8) The user interface module displays the analysis results, highlights potential problems, and provides improvement suggestions.
[0145] In this embodiment, the large language model is combined with RAG technology and real-time data integration, so that when dealing with power grid innovation project management issues, it is possible to automatically retrieve and integrate scattered professional knowledge, call the data interface of the relevant business system in real time, and combine multi-module collaborative analysis to provide comprehensive, accurate and dynamically updated decision-making suggestions, which greatly improves decision-making efficiency, improves analysis accuracy, optimizes professional knowledge management of power grid innovation projects, greatly improves the management level and efficiency of power grid innovation projects, simplifies the project review process, reduces decision-making risks, speeds up experience dissemination and learning, improves the practicality of suggestions and user satisfaction, reduces compliance risks, promotes innovative thinking, and can quickly adapt to new business needs of power grid innovation projects.
[0146] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0147] Based on the same inventive concept, the embodiment of the present application also provides a power grid project processing device for implementing the power grid project processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more power grid project processing device embodiments provided below can refer to the limitations of the power grid project processing method above, and will not be repeated here.
[0148] In one embodiment, Figure 5 As shown, a power grid project processing device 500 is provided, comprising: a phase identification module 501, a retrieval enhancement module 502, an abnormality analysis module 503, a project analysis module 504, and a management determination module 505, wherein:
[0149] The phase identification module 501 is used to perform project phase identification processing on the project application text of the power grid project to obtain project phase identification information of the project application text.
[0150] The search enhancement module 502 is used to perform search enhancement generation processing on the project phase identification information and the project application text based on the power grid knowledge base and the target large language model to obtain the project knowledge information and project system information corresponding to the project application text.
[0151] The abnormality analysis module 503 is used to perform abnormality analysis on the project application text according to the project system information to obtain the abnormality analysis result of the project application text.
[0152] The project analysis module 504 is used to perform project analysis on the project application text, project knowledge information and project system information through the target large language model to obtain project analysis information.
[0153] The management determination module 505 is used to obtain target management information of the power grid project according to the abnormal analysis results and the project analysis information.
[0154] In one embodiment, the retrieval enhancement module 502 is also used to perform similarity retrieval on the project phase identification information and the project application text in the power grid knowledge base to obtain candidate knowledge information and candidate system information that match the project phase identification information and the project application text; fuse the candidate knowledge information and the candidate system information with the project application text to obtain fused knowledge information and fused system information respectively; input the fused knowledge information and the fused system information into the target large language model to obtain project knowledge information and project system information associated with the project application text respectively.
[0155] In one embodiment, the project analysis module 504 is also used to obtain prompt words of the project application text based on the project application text and project knowledge information; input the prompt words, project application text, project knowledge information and project system information into the target large language model to obtain problem analysis information of the project application text; the target large language model is obtained by fine-tuning the pre-trained large language model according to historical project knowledge information and historical project application texts; according to the project system information, improvement suggestion information for the problem analysis information is generated; according to the problem analysis information and the improvement suggestion information, the project analysis information of the power grid project is obtained.
[0156] In one embodiment, the power grid project processing device 500 also includes a data acquisition module, which is used to obtain the current project management data of the project declaration text through a data interaction interface related to the power grid project; the current project management data, prompt words, project declaration text and project knowledge information are input into the target large language model to obtain problem analysis information of the project declaration text.
[0157] In one embodiment, the anomaly analysis module 503 is also used to perform text matching processing on the project application text according to the project system information to obtain the illegal content text in the project application text; generate the illegal content identification and illegal description information corresponding to the illegal content text; and obtain the anomaly analysis result according to the illegal content identification and illegal description information.
[0158] In one embodiment, the phase identification module 501 is also used to perform keyword extraction processing on the project application text of the power grid project to obtain keyword information of the project application text; perform feature matching processing on the keyword information according to a preset phase feature library to obtain a target phase feature that matches the project application text; the phase feature library stores multiple phase features and phase identification information corresponding to the phase features; and the phase identification information corresponding to the target phase feature is set as the project phase identification information of the project application text.
[0159] Each module in the above-mentioned power grid project processing device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.
[0160] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for processing a power grid project is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.
[0161] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0162] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.
[0163] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.
[0164] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.
[0165] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.
[0166] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0167] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A method for processing a power grid project, characterized in that: The method comprises: Performing project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text; Based on the power grid knowledge base and the target large language model, the project phase identification information and the project application text are searched, enhanced and generated to obtain project knowledge information and project system information corresponding to the project application text; According to the project system information, an abnormal analysis process is performed on the project application text to obtain an abnormal analysis result of the project application text; Through the target large language model, project analysis processing is performed on the project application text, the project knowledge information and the project system information to obtain project analysis information of the power grid project; According to the abnormal analysis result and the project analysis information, target management information of the power grid project is obtained.
2. The method according to claim 1, characterized in that The method of performing retrieval enhancement generation processing on the project phase identification information and the project application text based on the power grid knowledge base and the target large language model to obtain project knowledge information and project system information corresponding to the project application text includes: In the power grid knowledge base, similarity search is performed on the project phase identification information and the project application text to obtain candidate knowledge information and candidate system information that match the project phase identification information and the project application text; The candidate knowledge information and the candidate system information are respectively integrated with the project application text to obtain integrated knowledge information and integrated system information; The fused knowledge information and the fused system information are respectively input into the target large language model to obtain the project knowledge information and the project system information associated with the project application text.
3. The method according to claim 1, characterized in that The project analysis process is performed on the project application text, the project knowledge information and the project system information by using the target large language model to obtain the project analysis information of the power grid project, including: Obtaining prompt words of the project application text according to the project application text and the project knowledge information; The prompt words, the project application text, the project knowledge information and the project system information are input into the target large language model to obtain problem analysis information of the project application text; the target large language model is obtained by fine-tuning the pre-trained large language model according to the historical project knowledge information and the historical project application text; Generate improvement suggestion information for the problem analysis information based on the project system information; According to the problem analysis information and the improvement suggestion information, project analysis information of the power grid project is obtained.
4. The method according to claim 3, characterized in that The step of inputting the prompt words, the project application text and the project knowledge information into the target large language model to obtain problem analysis information of the project application text includes: Obtaining current project management data of the project application text through a data interaction interface related to the power grid project; The current project management data, the prompt words, the project declaration text and the project knowledge information are input into the target large language model to obtain problem analysis information of the project declaration text.
5. The method according to claim 1, characterized in that: The abnormal analysis process is performed on the project application text according to the project system information to obtain the abnormal analysis result of the project application text, including: According to the project system information, text matching processing is performed on the project application text to obtain the text of illegal content in the project application text; Generate a violation content identifier and violation description information corresponding to the violation content text; The abnormality analysis result is obtained according to the illegal content identifier and the illegal description information.
6. The method according to claim 1, characterized in that The project phase identification processing is performed on the project application text of the power grid project to obtain the project phase identification information of the project application text, including: Performing keyword extraction processing on the project application text of the power grid project to obtain keyword information of the project application text; According to a preset stage feature library, feature matching processing is performed on the keyword information to obtain a target stage feature that matches the project application text; the stage feature library stores a plurality of stage features and stage identification information corresponding to the stage features; The stage identification information corresponding to the target stage feature is set as the project stage identification information of the project declaration text.
7. A power grid project processing device, characterized in that: The device comprises: A phase identification module is used to perform project phase identification processing on a project application text of a power grid project to obtain project phase identification information of the project application text; A retrieval enhancement module is used to perform retrieval enhancement generation processing on the project phase identification information and the project application text based on the power grid knowledge base and the target large language model to obtain project knowledge information and project system information corresponding to the project application text; An abnormality analysis module is used to perform abnormality analysis on the project application text according to the project system information to obtain abnormality analysis results of the project application text; A project analysis module, used to perform project analysis processing on the project application text, the project knowledge information and the project system information through the target large language model to obtain project analysis information; A management determination module is used to obtain target management information of the power grid project based on the abnormal analysis result and the project analysis information.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
