Power grid dispatching query method and system, computer equipment, readable storage medium and program product

By combining the large language model with the power grid scheduling business system and using the React control module to obtain real-time data, the problem of untimely data updates in the traditional power grid scheduling system is solved, and the intelligent and real-time grid scheduling Q&A function is realized, which improves power System accuracy and practicality.

CN119066085APending Publication Date: 2024-12-03CHINA SOUTHERN POWER GRID ARTIFICIAL INTELLIGENCE TECHNOLOGY CO LTD

Patent Information

Application Number
CN202411185286.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional power grid scheduling systems rely on manual operations and are difficult to meet the efficiency and accuracy requirements of complex power systems. In the field of power scheduling, the information or suggestions provided are poorly accurate.

Method used

By combining the large language model with the power grid scheduling business system, the decomposition processing and decision-making reasoning of query tasks are realized, and the React control module is used to call the business system interface to obtain real-time data, and generate accurate scheduling decision-making and descriptive language to achieve intelligent and real-time grid dispatching Q&A function.

Benefits of technology

It improves the accuracy and practicality of the power system, reduces the uncertainty and risks of decision-making, realizes automated scheduling processes and data interactions, and improves the efficiency and accuracy of power grid scheduling.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066085A_ABST
    Figure CN119066085A_ABST
Patent Text Reader

Abstract

The invention relates to a power grid dispatching query method and system, computer equipment, a readable storage medium and a program product. The method comprises the following steps: decomposing a received query task to obtain a plurality of query sub-tasks, and calling a service system interface to obtain task information corresponding to each query sub-task; under the condition that the query task is scheduling query, reasoning decision making is conducted on the task information through a large language model, a decision making result is obtained, decision making operation corresponding to the decision making result is executed through a function executor, and an operation result is obtained; integrating the operation results corresponding to all the query sub-tasks to obtain a target scheduling decision corresponding to the query task, and displaying the target scheduling decision in a display area; and under the condition that the query task is data query, generating a descriptive language according to the task information, and displaying the descriptive language in the display area. By adopting the method, an intelligent and real-time power grid dispatching question and answer function can be realized, and the accuracy and practicability of a power system are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a power grid dispatch query method, system, computer equipment, readable storage medium and program product. Background Art

[0002] Traditional power grid dispatching systems require dispatchers to have rich professional knowledge and experience, but with the increase in the scale and complexity of power systems, manual dispatching alone can no longer meet the requirements of efficiency and accuracy.

[0003] At present, with the rapid development of artificial intelligence technology, large language models have demonstrated powerful capabilities in natural language processing, automated question answering, etc. The question answering system based on large language models can understand and process complex natural language problems, and provide intelligent solutions by combining with professional domain knowledge.

[0004] However, when applied to the field of power dispatching, due to the lack of direct connection with the actual business system, the data that the large model relies on may not be updated in real time, resulting in the information or suggestions provided based on outdated data when dealing with power dispatching related issues, resulting in poor decision-making accuracy. Summary of the invention

[0005] Based on this, it is necessary to provide a power grid dispatch query method, system, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy and practicality of the power system in response to the above technical problems.

[0006] In a first aspect, the present application provides a power grid dispatch query method, comprising:

[0007] Decompose the received query task to obtain multiple query subtasks, and call the business system interface to obtain task information corresponding to each query subtask;

[0008] When the query task is a scheduling query, the task information is inferred and decided through the large language model to obtain a decision result, and the decision operation corresponding to the decision result is executed through the function executor to obtain the operation result;

[0009] Integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area;

[0010] When the query task is a data query, a descriptive language is generated according to the task information and displayed in the display area.

[0011] In one embodiment, the step of decomposing the received query task to obtain multiple query subtasks includes:

[0012] Extract keywords from the received query tasks to obtain task keywords;

[0013] The task keywords are processed through a large language model to structure the query task.

[0014] Based on the results of the structured representation, multiple query subtasks are generated.

[0015] In one embodiment, the step of decomposing the received query task to obtain multiple query subtasks includes:

[0016] Extract keywords from the received query tasks to obtain task keywords;

[0017] The task keywords are processed through a large language model to structure the query task.

[0018] Based on the results of the structured representation, multiple query subtasks are generated.

[0019] In one embodiment, the step of making an inference decision on the task information through a large language model to obtain a decision result includes:

[0020] Obtaining context information corresponding to the task information;

[0021] The task information, context information and query task are semantically understood through a large language model, and reasoning decisions are made based on the results of semantic understanding to obtain decision results.

[0022] In a second aspect, the present application also provides a power grid dispatch query system, including:

[0023] The large language model module is used to make inference decisions on task information and obtain decision results when the query task is a scheduling query; when the query task is a data query, it generates descriptive language based on task information;

[0024] The function executor module is used to execute the decision operation corresponding to the decision result and obtain the operation result;

[0025] Business system interface module, used to provide data interaction interface with various power grid dispatching business systems;

[0026] The React control module is used to call the large language model module to decompose the received query task to obtain multiple query subtasks, and call the business system interface module to obtain the task information corresponding to each query subtask; it is also used to integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area; it is also used to display descriptive language in the display area.

[0027] In one embodiment, the system further comprises:

[0028] The user interface module is used to provide a user interaction interface and receive query text input by the user;

[0029] The query processing module is used to generate a query task based on the query text; extract keywords from the received query task to obtain task keywords, so that the large language model module processes the task keywords to structure the query task, and generate multiple query subtasks based on the results of the structured representation.

[0030] In one embodiment, the display interface of the user interface module includes a top area, a middle area, a lower area and a right area; the top area is used to display the system title, the middle area is used to provide a user interaction interface and receive query text entered by the user, the lower area is used to instruct the user to confirm the query text, and the right area is used to display the target scheduling decision; it is also used to display descriptive language.

[0031] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method steps of any one of the first aspects are implemented.

[0032] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements any one of the method steps in the first aspect when the computer program is executed by a processor.

[0033] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements any one of the method steps in the first aspect when executed by a processor.

[0034] The above-mentioned power grid dispatching query method, device, computer equipment, computer-readable storage medium and computer program product obtain multiple query subtasks by decomposing the received query task, calling the business system interface to obtain the task information corresponding to each query subtask, and obtaining the decision result by reasoning and deciding the task information through the large language model, and executing the decision operation corresponding to the decision result through the function executor to obtain the operation result, integrating the operation results corresponding to all query subtasks to obtain the target dispatching decision corresponding to the query task, and displaying the target dispatching decision in the display area. Since the large language model is combined with the power grid dispatching business system, it can realize intelligent and real-time power grid dispatching question and answer functions, and improve the accuracy and practicality of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0036] FIG1 is an application environment diagram of a power grid dispatch query method in one embodiment;

[0037] FIG2 is a structural block diagram of a power grid dispatch query system in one embodiment;

[0038] FIG3 is a schematic diagram of a flow chart of a power grid dispatch query method in one embodiment;

[0039] FIG4 is a schematic diagram of a flow chart of a power grid dispatch query method in another embodiment;

[0040] FIG5 is a structural block diagram of a power grid dispatch query device in one embodiment;

[0041] FIG. 6 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0042] 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.

[0043] The power grid dispatch query method provided in the embodiment of the present application can be applied to the application environment shown in Figure 1. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 is used to decompose the received query task to obtain multiple query subtasks, call the business system interface to obtain the task information corresponding to each query subtask from the server 104, and in the case where the query task is a scheduling query, the task information is inferred and decided through the large language model to obtain the decision result, and the decision operation corresponding to the decision result is executed through the function executor to obtain the operation result, and the operation results corresponding to all query subtasks are integrated to obtain the target scheduling decision corresponding to the query task, and the target scheduling decision is displayed in the display area. In the case where the query task is a data query, a descriptive language is generated according to the task information, and the descriptive language is displayed in the display area. The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0044] The power system is a key infrastructure of modern society. Its safe and stable operation directly affects the normal operation of the economy and society. Grid dispatching is a crucial link in the operation of the power system, involving real-time monitoring, load forecasting, power generation planning, fault handling, etc. The traditional power grid dispatching system requires dispatchers to have rich professional knowledge and experience, but with the increase in the scale and complexity of the power system, manual dispatching alone can no longer meet the requirements of efficiency and accuracy.

[0045] At present, with the rapid development of artificial intelligence technology, large language models have shown strong capabilities in natural language processing, automated question answering, etc. The question answering system based on large language models can understand and process complex natural language problems, and provide intelligent solutions by combining with professional domain knowledge. Through large-scale pre-training, it has high accuracy and context understanding ability when processing natural language problems. The question answering ability of these models can be improved in specific professional fields (such as power grid dispatch) through further fine-tuning and supplementation of domain knowledge. In the development of autonomous function calls of large models, React refers to the prompt word framework of large models. The ReAct method combines reasoning and acting, allowing large language models to not only perform logical reasoning to solve problems, but also generate a series of action steps to execute based on these reasoning. The ReAct framework is designed to enable models to interact with the external environment and solve complex tasks by generating reasoning trajectories and action plans.

[0046] However, when applied to the field of power dispatching, due to the lack of direct connection with the actual business system, the data that the big model relies on may not be updated in real time, resulting in the information or suggestions provided based on outdated data when dealing with power dispatching related issues. This is a significant limitation in the highly dynamic power system that requires instant response. In addition, the field of power dispatching has extremely high requirements for the accuracy of data and the specificity of description, and the answers generated by the big model may not be accurate or detailed enough due to the limitations of the training data, which may introduce uncertainty and risk in the decision-making process.

[0047] Based on this, an embodiment of the present application provides a power grid dispatching query method, focusing on the decision-making difficulties caused by information dispersion and low data call efficiency in the question and answer process in the power grid dispatching domain. Compared with the existing technology, the large language model is combined with the improved React intelligent agent paradigm, so that the question and answer system can automatically select and call the data interface of the relevant business system when answering questions in the field of power grid dispatching, and provide accurate and dynamically updated answers by combining real-time data and domain knowledge, which significantly improves the accuracy and practicality of the question and answer system.

[0048] In an exemplary embodiment, as shown in FIG. 2 , a power grid dispatch query system is provided, including: a large language model module 210 , a function executor module 220 , a business system interface module 230 and a React control module 240 .

[0049] Among them, the large language model module 210 is used to make inference decisions on task information and obtain decision results when the query task is a scheduling query; when the query task is a data query, a descriptive language is generated according to the task information. The function executor module 220 is used to execute the decision operation corresponding to the decision result and obtain the operation result. The business system interface module 230 is used to provide a data interaction interface with each power grid dispatching business system. The React control module 240 is used to call the large language model module to decompose the received query task to obtain multiple query subtasks, and call the business system interface module 230 to obtain the task information corresponding to each query subtask; it is also used to integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area; it is also used to display descriptive language in the display area.

[0050] In this embodiment, the language model module is used to make inference decisions for scheduling query tasks, which can quickly analyze complex power grid data and operating conditions and generate accurate scheduling decisions. The function executor module can automatically execute operations corresponding to the decision results without human intervention, thereby realizing the automation of the scheduling process. The business system interface module provides a data interaction interface with each power grid scheduling business system, so that the system can easily obtain the required data information. The React control module 240 can call a large language model to decompose complex query tasks and generate multiple query subtasks. This decomposition process not only reduces the complexity of a single task, but also enables the system to process multiple subtasks in parallel, further improving the processing speed and efficiency, thereby realizing intelligent and real-time power grid scheduling question and answer functions and improving the accuracy and practicality of the power system.

[0051] In an exemplary embodiment, still as shown in FIG. 2 , the system further includes: a user interface module 250 and a query processing module 260 .

[0052] The user interface module 250 is used to provide a user interaction interface and receive the query text input by the user. The query processing module 260 is used to generate a query task according to the query text; extract keywords from the received query task to obtain task keywords, so that the large language model module processes the task keywords to perform a structured representation of the query task, and generate multiple query subtasks according to the result of the structured representation.

[0053] In this embodiment, through the design of the user interface module and the query processing module, the system is able to process complex query texts. Users can construct complex query tasks by entering detailed query conditions or descriptions, and the query processing module can decompose these complex query tasks into multiple simple and manageable query subtasks, thereby supporting a wider range of query needs.

[0054] In an exemplary embodiment, the display interface of the user interface module 250 includes a top area, a middle area, a lower area and a right area; the top area is used to display the system title, the middle area is used to provide a user interaction interface and receive query text entered by the user, the lower area is used to instruct the user to confirm the query text, and the right area is used to display the target scheduling decision; it is also used to display descriptive language.

[0055] In this embodiment, by placing the system title, user interaction interface, query text confirmation instructions, target scheduling decisions and descriptive language in different areas, users can more clearly identify and understand the information on the interface. Clear area division and reasonable functional layout enable users to quickly find the parts that need to be operated, reducing the time of searching for information on the interface, thereby improving the processing speed and efficiency of the system.

[0056] In an exemplary embodiment, as shown in FIG3 , a power grid dispatch query method is provided, which is described by taking the method applied to the terminal 102 in FIG1 as an example, and includes the following steps 302 to 308 . Among them:

[0057] S302: Decompose the received query task to obtain multiple query subtasks, and call the business system interface to obtain task information corresponding to each query subtask.

[0058] Optionally, the query task transmitted from the query processing module is received through the React control module, and the large language model is called to decompose the query task, so that the query task can be decomposed into multiple subtasks. The large language model module cyclically executes the processing process for the subtask until all subtasks are processed. Among them, the input data of the large language model module is the task information corresponding to the current query subtask obtained through the business system interface.

[0059] S304: When the query task is a scheduling query, the task information is inferred and decided through the large language model to obtain a decision result, and the decision operation corresponding to the decision result is executed through the function executor to obtain an operation result.

[0060] Optionally, when the user is performing a scheduling query, for example, when generating a power grid scheduling plan, the user can obtain relevant historical power data and current power grid status by calling the business system interface, and make inference decisions on the task information through a large language model to obtain a decision result, i.e., a preliminary scheduling plan. The decision operation corresponding to the decision result is executed by the function executor module to simulate the execution effect of the preliminary scheduling plan and obtain the operation result of the function executor module. The preliminary scheduling plan can be further optimized based on the operation result to obtain the final operation result.

[0061] S306: Integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area.

[0062] Optionally, the operation results corresponding to all query subtasks are integrated through the React control module to generate a target scheduling decision corresponding to the query task, and the target scheduling decision is displayed in the display area and fed back to the user through the user interface module.

[0063] S308: When the query task is a data query, a descriptive language is generated according to the task information, and the descriptive language is displayed in the display area.

[0064] Optionally, when a user performs a data query, for example, needs to query the load status of a line, the large language model module is called through the React control module to analyze the data currently required, and the corresponding real-time load data is queried through the business system interface module. The load data is used as input data of the large language model, and is analyzed through the large language model to generate corresponding descriptive language, which is then fed back to the user through the user interface module.

[0065] In the above-mentioned power grid dispatching query method, the received query task is decomposed and processed to obtain multiple query subtasks, and the business system interface is called to obtain the task information corresponding to each query subtask. Since the task information is inferred and decided through the large language model, the decision result is obtained, and the decision operation corresponding to the decision result is executed by the function executor to obtain the operation result. The operation results corresponding to all query subtasks are integrated to obtain the target dispatching decision corresponding to the query task, and the target dispatching decision is displayed in the display area. Since the large language model is combined with the power grid dispatching business system, the intelligent and real-time power grid dispatching question and answer function can be realized, thereby improving the accuracy and practicality of the power system.

[0066] In an exemplary embodiment, the steps of decomposing a received query task to obtain multiple query subtasks include: extracting keywords from the received query task to obtain task keywords; processing the task keywords through a large language model to perform a structured representation of the query task; and generating multiple query subtasks based on the results of the structured representation.

[0067] Optionally, when processing query tasks, in order to improve processing efficiency, the query processing module can perform preliminary processing on the received query tasks, extract the keywords of the query tasks, and pass the task keywords to the React control module. The React control module calls the large language model module to analyze the task keywords, and represents the query tasks in a structured form, thereby decomposing them into multiple query subtasks.

[0068] In this embodiment, unnecessary redundant information is removed by performing keyword extraction on the received query task, thereby improving the accuracy and relevance of the query; and by decomposing the query task into multiple query subtasks, these subtasks can be processed in parallel, thereby shortening the time of the entire query process and realizing intelligent and real-time power grid dispatching question and answer functions.

[0069] In an exemplary embodiment, the step of calling a business system interface to obtain task information corresponding to each query subtask includes: for each query subtask, determining the business system interface corresponding to the query subtask; sending an information request to the business system interface through a function executor, and receiving task data returned by the business system interface; and preprocessing the task data to obtain task information corresponding to the query subtask.

[0070] Optionally, the business system interface module can provide a data interaction interface with multiple power grid business systems. When receiving a query subtask from the React control module, the business system interface module selects a corresponding business system interface according to the type of the query subtask, sends an information request to the business system interface through the function executor, and receives the task data returned by the business system. It then performs pre-processing such as format conversion and data cleaning on the task data to obtain the task information corresponding to the query subtask, and returns it to the React control module.

[0071] In this embodiment, the business system interface module provides a data interaction interface with multiple power grid business systems, which can improve data accuracy. By sending information requests, receiving task data and performing preprocessing through the function executor, intelligent and real-time power grid dispatching question and answer functions can be realized, thereby improving the accuracy and practicality of the power system.

[0072] In an exemplary embodiment, the step of performing reasoning and decision making on task information through a large language model to obtain a decision result includes: obtaining context information corresponding to the task information; performing semantic understanding on the task information, context information and query task through a large language model, and performing reasoning and decision making based on the results of the semantic understanding to obtain a decision result.

[0073] Optionally, before making an inference decision, it is necessary to first obtain context information related to the task information. This context information may include historical data, background knowledge, domain knowledge, etc., which are essential for accurately understanding the task information and making reasonable decisions. The large language model has learned the complex patterns and semantic relationships of the language through its pre-trained large amount of text data. After receiving the task information, context information and query tasks, the large language model can use this information for deep semantic understanding. It can not only identify keywords and phrases in the text, but also understand the logical relationship, referential relationship and implicit meaning between them. Based on semantic understanding, the large language model can make inference decisions based on the task information and context information, combined with its own knowledge base and reasoning ability, and obtain decision results.

[0074] In this embodiment, a large language model is used to perform semantic understanding of task information, context information, and query tasks, and reasoning and decision making are performed based on the results of the semantic understanding to obtain decision results. This can process large amounts of information and perform complex reasoning more quickly, and can realize intelligent and real-time grid dispatching question-and-answer functions, thereby improving the accuracy and practicality of the power system.

[0075] In an exemplary embodiment, as shown in FIG4 , a power grid dispatch query method is provided, the method comprising the following steps:

[0076] S402: extract keywords from the received query task to obtain task keywords; process the task keywords through a large language model to perform a structured representation of the query task; and generate multiple query subtasks based on the result of the structured representation.

[0077] S404: When the query task is a scheduling query, context information corresponding to the task information is obtained; semantic understanding of the task information, context information and query task is performed through a large language model, and reasoning and decision making are performed based on the results of the semantic understanding to obtain a decision result, and a decision operation corresponding to the decision result is executed through a function executor to obtain an operation result.

[0078] S406: Integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area.

[0079] S408: When the query task is a data query, a descriptive language is generated according to the task information, and the descriptive language is displayed in the display area.

[0080] In this embodiment, the received query task is decomposed to obtain multiple query subtasks, and the business system interface is called to obtain task information corresponding to each query subtask. Since the task information is inferred and decided through the large language model, a decision result is obtained, and the decision operation corresponding to the decision result is executed by the function executor to obtain the operation result. The operation results corresponding to all query subtasks are integrated to obtain the target scheduling decision corresponding to the query task, and the target scheduling decision is displayed in the display area. Since the large language model is combined with the power grid scheduling business system, an intelligent and real-time power grid scheduling question and answer function can be realized, thereby improving the accuracy and practicality of the power system.

[0081] It should be understood that, although the steps in the flowcharts involved in the above 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 is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may 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 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.

[0082] Based on the same inventive concept, the embodiment of the present application also provides a power grid dispatch query device for implementing the power grid dispatch query 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 dispatch query device embodiments provided below can refer to the limitations of the power grid dispatch query method above, and will not be repeated here.

[0083] In an exemplary embodiment, as shown in FIG5 , a power grid dispatch query device is provided, comprising: a task decomposition module 510 , a reasoning and decision module 520 , a dispatch integration module 530 and a language generation module 540 , wherein:

[0084] The task decomposition module 510 is used to decompose the received query task to obtain multiple query subtasks, and call the business system interface to obtain task information corresponding to each query subtask.

[0085] The reasoning and decision module 520 is used to make reasoning and decisions on task information through a large language model when the query task is a scheduling query to obtain a decision result, and to execute a decision operation corresponding to the decision result through a function executor to obtain an operation result.

[0086] The scheduling integration module 530 is used to integrate the operation results corresponding to all query subtasks, obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area.

[0087] The language generation module 540 is used to generate descriptive language according to task information and display the descriptive language in the display area when the query task is a data query.

[0088] In an exemplary embodiment, the task decomposition module 510 is also used to extract keywords from the received query task to obtain task keywords; process the task keywords through a large language model to perform a structured representation of the query task; and generate multiple query subtasks based on the results of the structured representation.

[0089] In an exemplary embodiment, the task decomposition module 510 is also used to determine the business system interface corresponding to each query subtask; send an information request to the business system interface through the function executor, and receive task data returned by the business system interface; pre-process the task data to obtain task information corresponding to the query subtask.

[0090] In an exemplary embodiment, the reasoning decision module 520 is also used to obtain context information corresponding to the task information; perform semantic understanding of the task information, context information and query task through a large language model, and perform reasoning decision based on the results of the semantic understanding to obtain a decision result.

[0091] Each module in the above-mentioned power grid dispatch query 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 the operations corresponding to each of the above modules.

[0092] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG6. 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 via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via 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 an 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 may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for querying power grid scheduling is implemented. The display unit of the computer device is used to form a visually visible picture, which may 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.

[0093] Those skilled in the art will understand that the structure shown in FIG. 6 is merely a block diagram of a partial 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 shown in the figure, or combine certain components, or have a different arrangement of components.

[0094] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: decomposing a received query task to obtain multiple query subtasks, calling a business system interface to obtain task information corresponding to each query subtask; when the query task is a scheduling query, reasoning and deciding on the task information through a large language model to obtain a decision result, and executing a decision operation corresponding to the decision result through a function executor to obtain an operation result; integrating the operation results corresponding to all query subtasks to obtain a target scheduling decision corresponding to the query task, and displaying the target scheduling decision in a display area; when the query task is a data query, generating a descriptive language based on the task information, and displaying the descriptive language in a display area.

[0095] In one embodiment, when a processor executes a computer program, it decomposes a received query task to obtain multiple query subtasks, including: extracting keywords from the received query task to obtain task keywords; processing the task keywords through a large language model to perform a structured representation of the query task; and generating multiple query subtasks based on the results of the structured representation.

[0096] In one embodiment, the processor executes a computer program involving calling a business system interface to obtain task information corresponding to each query subtask, including: for each query subtask, determining the business system interface corresponding to the query subtask; sending an information request to the business system interface through a function executor, and receiving task data returned by the business system interface; preprocessing the task data to obtain task information corresponding to the query subtask.

[0097] In one embodiment, when a processor executes a computer program, reasoning and deciding on task information through a large language model to obtain a decision result includes: obtaining context information corresponding to the task information; semantically understanding the task information, context information and query task through a large language model, and reasoning and deciding based on the result of the semantic understanding to obtain a decision result.

[0098] 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 following steps are implemented: decomposing a received query task to obtain multiple query subtasks, calling a business system interface to obtain task information corresponding to each query subtask; when the query task is a scheduling query, reasoning and deciding on the task information through a large language model to obtain a decision result, and executing a decision operation corresponding to the decision result through a function executor to obtain an operation result; integrating the operation results corresponding to all query subtasks to obtain a target scheduling decision corresponding to the query task, and displaying the target scheduling decision in a display area; when the query task is a data query, generating a descriptive language based on the task information, and displaying the descriptive language in a display area.

[0099] In one embodiment, when a computer program is executed by a processor, it involves decomposing a received query task to obtain multiple query subtasks, including: extracting keywords from the received query task to obtain task keywords; processing the task keywords through a large language model to perform a structured representation of the query task; and generating multiple query subtasks based on the results of the structured representation.

[0100] In one embodiment, when a computer program is executed by a processor, the calling of a business system interface to obtain task information corresponding to each query subtask involves: for each query subtask, determining the business system interface corresponding to the query subtask; sending an information request to the business system interface through a function executor, and receiving task data returned by the business system interface; and preprocessing the task data to obtain task information corresponding to the query subtask.

[0101] In one embodiment, when a computer program is executed by a processor, reasoning and decision-making on task information is performed through a large language model to obtain a decision result, including: obtaining context information corresponding to the task information; semantically understanding the task information, context information and query task through a large language model, and reasoning and decision-making based on the results of the semantic understanding to obtain a decision result.

[0102] In one embodiment, a computer program product is provided, including a computer program, which implements the following steps when executed by a processor: decomposing a received query task to obtain multiple query subtasks, calling a business system interface to obtain task information corresponding to each query subtask; when the query task is a scheduling query, reasoning and deciding on the task information through a large language model to obtain a decision result, and executing a decision operation corresponding to the decision result through a function executor to obtain an operation result; integrating the operation results corresponding to all query subtasks to obtain a target scheduling decision corresponding to the query task, and displaying the target scheduling decision in a display area; when the query task is a data query, generating a descriptive language based on the task information, and displaying the descriptive language in a display area.

[0103] In one embodiment, when a computer program is executed by a processor, it involves decomposing a received query task to obtain multiple query subtasks, including: extracting keywords from the received query task to obtain task keywords; processing the task keywords through a large language model to perform a structured representation of the query task; and generating multiple query subtasks based on the results of the structured representation.

[0104] In one embodiment, when a computer program is executed by a processor, the calling of a business system interface to obtain task information corresponding to each query subtask involves: for each query subtask, determining the business system interface corresponding to the query subtask; sending an information request to the business system interface through a function executor, and receiving task data returned by the business system interface; and preprocessing the task data to obtain task information corresponding to the query subtask.

[0105] In one embodiment, when a computer program is executed by a processor, reasoning and decision-making on task information is performed through a large language model to obtain a decision result, including: obtaining context information corresponding to the task information; semantically understanding the task information, context information and query task through a large language model, and reasoning and decision-making based on the results of the semantic understanding to obtain a decision result.

[0106] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method 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 the present application can include at least one of non-volatile memory 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), magnetic 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, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0107] 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 application.

[0108] 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 power grid dispatch query method, characterized in that: The method comprises: Decomposing the received query task to obtain multiple query subtasks, and calling the business system interface to obtain task information corresponding to each query subtask; In the case where the query task is a scheduling query, the task information is inferred and decided by a large language model to obtain a decision result, and a decision operation corresponding to the decision result is executed by a function executor to obtain an operation result; Integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area; In the case where the query task is a data query, a descriptive language is generated according to the task information, and the descriptive language is displayed in the display area.

2. The method according to claim 1, characterized in that: The received query task is decomposed to obtain multiple query subtasks, including: Extract keywords from the received query tasks to obtain task keywords; Processing the task keywords through a large language model to perform a structured representation of the query task; Based on the results of the structured representation, multiple query subtasks are generated.

3. The method according to claim 1, characterized in that The calling of the business system interface to obtain task information corresponding to each query subtask includes: For each query subtask, determining a business system interface corresponding to the query subtask; Sending an information request to the business system interface through a function executor, and receiving task data returned by the business system interface; The task data is preprocessed to obtain task information corresponding to the query subtask.

4. The method according to claim 1, characterized in that: The reasoning and decision making of the task information by the large language model to obtain a decision result includes: Acquire context information corresponding to the task information; The task information, the context information and the query task are semantically understood through a large language model, and reasoning decisions are made based on the results of the semantic understanding to obtain a decision result.

5. A power grid dispatch query system, characterized in that: The system comprises: The large language model module is used to make inference decisions on task information and obtain decision results when the query task is a scheduling query; and to generate descriptive language based on the task information when the query task is a data query; A function executor module is used to execute the decision operation corresponding to the decision result to obtain the operation result; Business system interface module, used to provide data interaction interface with various power grid dispatching business systems; The React control module is used to call the large language model module to decompose the received query task to obtain multiple query subtasks, and call the business system interface module to obtain the task information corresponding to each query subtask; it is also used to integrate the operation results corresponding to all query subtasks to obtain the target scheduling decision corresponding to the query task, and display the target scheduling decision in the display area; it is also used to display the descriptive language in the display area.

6. The system according to claim 5, characterized in that The system further comprises: The user interface module is used to provide a user interaction interface and receive query text input by the user; The query processing module is used to generate a query task according to the query text; perform keyword extraction on the received query task to obtain task keywords, so that the large language model module processes the task keywords to perform structured representation on the query task, and generate multiple query subtasks according to the results of the structured representation.

7. The system according to claim 6, characterized in that The display interface of the user interface module includes a top area, a middle area, a lower area and a right area; the top area is used to display the system title, the middle area is used to provide a user interaction interface and receive the query text input by the user, the lower area is used to instruct the user to confirm the query text, and the right area is used to display the target scheduling decision; it is also used to display the descriptive language.

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 4 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 4 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 4 are implemented.

Citation Information

Patent Citations

  • Data query method and device, electronic equipment and storage medium

    CN117667973A

  • Database query instruction generation method and device, computer equipment and storage medium

    CN117909355A

  • Flood control scheduling rule question and answer recommendation method and system based on large language model

    CN118069797A

  • Structured query language generation method and device, electronic equipment and storage medium

    CN118093621A

  • Database Queries Enriched in Rules

    US20130117323A1

Cited By

  • Intelligent interaction method and device based on AI large model logistics data and storage medium

    CN120258251A

  • Power grid intelligent scheduling decision-making method, device and equipment based on retrieval enhancement generation

    CN121920790A