A power question number query system based on multi-agent fusion

Abstract

The present invention relates to the field of power systems, and specifically to a power question number query system based on multi-agent fusion. The system includes a user interface, a data processing module, and a service customization output module. The user interface includes a data input unit and a data display unit. The data processing module includes a data storage unit and an agent module. The service customization output module includes a customization processing engine and an output middleware. The output middleware is also connected to the data processing module. The present invention can overcome the problems of low query accuracy, insufficient business relevance, and lack of customization output ability faced by large models in the prior art. The core concept of this system is to effectively decompose complex power query tasks into a series of smaller and more manageable subtasks, and each subtask is independently responsible for by a dedicated agent to improve the overall query accuracy and efficiency.

Description

Technical Field

[0001] The present invention relates to the field related to power systems, and specifically to a power question number query system based on multi-agent fusion. Background Art

[0002] In the current power system, data query is a crucial link, which involves multiple aspects such as power production, transmission, distribution, and consumption. Traditional data query methods mainly rely on manual operations and traditional database query technologies. The query method of manual operation requires operators to input query instructions one by one according to the specific needs of users and retrieve data through the database. This process is not only time-consuming and laborious, but also extremely inefficient when dealing with a large number of query requests. Operators need to spend a lot of time on data screening, sorting, and analysis, which not only increases the workload but also prolongs the query response time.

[0003] In modern power systems, with the development of big data and artificial intelligence technologies, some attempts have been made to directly use large language models for power data query. This method attempts to directly understand the query text information output by users using a large model, directly generate the query SQL statement, then use the SQL statement to automatically query the database, and directly return the results to users. However, although large models have shown powerful data processing capabilities in some fields, in the specific application of power data query, they still face a series of challenges and problems:

[0004] 1. Weak accuracy: Large models usually require a large amount of data for training to identify and predict patterns. However, in the field of power data query, due to the complexity and diversity of data, large models often have difficulty achieving ideal accuracy. Especially when facing specific or non-standard query requests, the predictions and query results of the model may deviate far from the actual needs.

[0005] 2. Low business relevance: Data query in the power industry often requires an in-depth understanding of business processes and operations. Existing large models may lack in-depth learning of the specific business logic of the power industry, resulting in low relevance between query results and actual business needs and being unable to meet the complex query needs of professional users.

[0006] 3. Unable to perform customized output: Although large models can handle a wide range of query tasks, they usually provide standardized output and lack the ability to be customized. In the power industry, users may need to obtain customized data reports or views according to specific business scenarios or analysis purposes, and it is difficult for existing large models to be flexibly adjusted to meet this need. Summary of the Invention

[0007] The purpose of the present invention is to provide a virtual driving vehicle test system with high immersion human-computer interaction to solve the problems mentioned in the above background technology.

[0008] To achieve the above purpose, the present invention provides the following technical solutions:

[0009] A power question number query system based on multi-agent fusion includes a user interface, a data processing module, and a service customization output module. The user interface includes a data input unit and a data display unit. The user inputs a query request to be performed through the data input unit and transfers the data input by the user to the data processing module. The data display unit displays the data.

[0010] The data processing module includes a data storage unit and an agent module. The data storage unit is responsible for storing and managing power data, which is a database. To facilitate querying the database, an SQL statement needs to be input, and the SQL statement is generated by the agent module. The agent module mainly includes four agents. The data input unit communicates with the first agent. After the user inputs data from the data input unit, the first agent will deeply understand the user's query intention and requirements.

[0011] The second agent parses the data input by the user from the data input unit. The second agent uses an advanced intention recognition algorithm to parse the query intention in the data.

[0012] The third agent extracts key information from the data input by the user from the data input unit. The key information includes a time range, specific parameters, or data fields.

[0013] The fourth agent will obtain the data query voice generated by the first agent, the query intention parsed by the second agent, and the key information extracted by the third agent. The fourth agent will generate corresponding SQL query statements with these data. The fourth agent has the ability to convert natural language queries into database query languages.

[0014] Through the above four agents, the data input by the user from the data input unit is converted into an SQL query statement, and the SQL query statement is input into the data storage unit. The corresponding SQL query results will be generated in the data storage unit.

[0015] The service customization output module includes a customization processing engine and an output middleware. The customization processing engine obtains the results of SQL queries from the data processing module, and the customization processing engine performs secondary processing on the results according to business requirements, and finally generates customized query results. The query results are in a format that is easy to understand and operate, specifically charts, reports, and specific data views. The output middleware obtains the customized query results generated by the customization processing engine and processes the customized query results.

[0016] Preferably, the intelligent agent module includes a memory module and an execution module. All four intelligent agents include a memory module and an execution module. The memory module is responsible for storing the memory information of previous queries, and the execution module, according to these memory information and the specific requirements of the current task, guides the large model to perform targeted task output through a prompt template.

[0017] Preferably, the intelligent agent module further includes a fine-tuning model library. Professional knowledge and data in the power industry are introduced through the fine-tuning model library, including training the model on industry terms, concepts, and business processes to ensure that the generated query results are highly relevant to power operations.

[0018] Preferably, the output middleware communicates with the data display unit. The output middleware transfers the customized query results to the data display unit, and the data display unit displays the data and presents the customized query results to the user.

[0019] Preferably, before the SQL query statement generated by the fourth intelligent agent is input into the data storage unit, the generated SQL query statement is input into the output middleware, and the output middleware inputs the SQL query statement into the data display unit. When the generated SQL query statement is correct, without any operation by the user, the unchanged SQL query statement will be sent back to the data processing module through the output middleware again and input into the data storage unit;

[0020] When the generated SQL query statement is incorrect, the user modifies it at the data display unit and then sends the changed SQL query statement back to the data processing module through the output middleware again and inputs the new SQL query statement into the data storage unit.

[0021] Preferably, when the data processing module processes the data collected by the data input unit and requires additional information, the user inputs the additional information from the data input unit, and the additional information will be directly given to the second intelligent agent. The second intelligent agent analyzes the query intention from the additional information and transfers the query intention to the third intelligent agent to facilitate the third intelligent agent to extract key information.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0023] 1. A power question number query system based on multi-agent fusion of the present invention adopts a task decomposition strategy to break down complex query tasks into multiple small tasks, and each small task corresponds to a more specific query target and data set. This strategy reduces the complexity of a single task, enabling the model to more accurately understand and process each subtask, thereby improving the accuracy of queries as a whole.

[0024] 2. A power question number query system based on multi-agent fusion of the present invention fine-tunes the model by introducing industry expert knowledge, enabling it to better understand and adapt to the business requirements of the power industry. By training the model to recognize specific terms, concepts, and business processes in the power industry, the present invention can generate query results highly relevant to power operations, meeting the specific needs of professional users.

[0025] 3. A power question number query system based on multi-agent fusion of the present invention introduces dedicated processing logic and processes to perform secondary analysis and adjustment on the initial output results of the large model. In this way, the system can generate customized query results according to the specific business needs of users, such as specific data report formats, visual displays, or analysis suggestions, thereby providing more personalized and business-oriented services. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic diagram of the system structure of the present invention;

[0027] Figure 2 is a system flow chart of the present invention;

[0028] Figure 3 is a power query task flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0030] It should be understood that in the embodiments of the present invention, the term "and / or" is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0031] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0032] Please refer to Figure 1 , the present invention provides a technical solution: a power question number query system based on multi-agent fusion, including a user interface, a data processing module, and a service customization output module. The user interface includes a data input unit and a data display unit. The user inputs a query request to be performed through the data input unit and transfers the data input by the user to the data processing module. The data display unit displays the data;

[0033] The data processing module includes a data storage unit and an agent module. The data storage unit is responsible for storing and managing power data, which is a database. In order to facilitate querying the database, an SQL statement needs to be input. The SQL statement is generated by the agent module. The agent module mainly includes four agents. The user interface communicates with the first agent. After the user inputs data from the user interface, the first agent will deeply understand the user's query intention and requirements. The first agent has advanced natural language understanding ability, which can enable the first agent to extract key information from the data collected by the data input unit and convert it into a data query language;

[0034] The second agent parses the data input by the user from the data input unit. The second agent uses an advanced intention recognition algorithm to parse the query intention in the data to ensure an accurate grasp of the deep purpose of the user's query;

[0035] The third agent extracts key information from the data input by the user from the data input unit. The key information includes a time range, specific parameters, or data fields, which are crucial for constructing an accurate query;

[0036] The fourth agent will obtain the data query language generated by the first agent, the query intention parsed by the second agent, and the key information extracted by the third agent. The fourth agent will generate the corresponding SQL query statement with these data. The fourth agent has the ability to convert a natural language query into a database query language to ensure the accuracy and execution efficiency of the query statement;

[0037] Through the above four intelligent agents, the data input by the user from the user interface is converted into an SQL query statement, and the SQL query statement is input into the data storage unit, where the corresponding SQL query result will be generated;

[0038] The business customization output module includes a customization processing engine and an output middleware. The customization processing engine obtains the SQL query result from the data processing module, and the customization processing engine performs secondary processing on the result according to business requirements, and finally generates a customized query result, which is in a format easy to understand and operate, specifically charts, reports, and specific data views. The output middleware obtains the customized query result generated by the customization processing engine and processes the customized query result.

[0039] Furthermore, the intelligent agent module includes a memory module and an execution module. All four intelligent agents include a memory module and an execution module. The memory module is responsible for storing the memory information of previous queries, and the execution module, according to this memory information and the specific requirements of the current task, guides the large model to perform targeted task output through a prompt template.

[0040] Furthermore, the intelligent agent module also includes a fine-tuning model library. By introducing professional knowledge and data in the power industry through the fine-tuning model library, including training the model on industry terms, concepts, and business processes, it is ensured that the generated query results are highly relevant to power operations.

[0041] Furthermore, the output middleware communicates with the data display unit. The output middleware transfers the customized query result to the data display unit, and the data display unit displays the data and presents the customized query result to the user.

[0042] Please refer to Figure 2 , furthermore, before the SQL query statement generated by the fourth intelligent agent is input into the data storage unit, the generated SQL query statement is input into the output middleware, and the output middleware inputs the SQL query statement into the data display unit. When the generated SQL query statement is correct, after the user does not perform any operations, the unchanged SQL query statement will be sent back to the data processing module through the output middleware again and input into the data storage unit;

[0043] When the generated SQL query statement is incorrect, the user modifies it at the data display unit and then sends the changed SQL query statement back to the data processing module through the output middleware again, and inputs the new SQL query statement into the data storage unit.

[0044] Please refer to Figure 3, Further, when the data processing module processes the data collected by the data input unit and requires additional information, the user inputs the additional information from the data input unit. The additional information will be directly given to the second intelligent agent, and the second intelligent agent will parse the query intention from the additional information and pass the query intention to the third intelligent agent to facilitate the third intelligent agent to extract key information.

[0045] In the embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual coupling, direct coupling, or communication connection can be an indirect coupling or communication connection through some interfaces, devices, or units, and can also be in the form of electrical, mechanical, or other connections.

[0046] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of the present invention.

[0047] In addition, in each embodiment of the present invention, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0048] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of the present invention. The foregoing storage medium includes: USB flash drive, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk, or optical disk, etc., which can store program codes.

[0049] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

[0050] It is worth noting that although the foregoing has described the spirit and principle of the present invention according to several specific implementation manners, it should be understood that the present invention is not limited to the disclosed specific implementation manners, and the division of each aspect does not mean that the features in these aspects cannot be combined. This division is only for the convenience of expression. The present invention aims to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A power question number query system based on multi-agent fusion, comprising a user interface, a data processing module, and a service customization output module, characterized in that: The user interface includes a data input unit and a data display unit. The user inputs a query request through the data input unit and transfers the input data to the data processing module, and the data display unit displays the data. The data processing module includes a data storage unit and an agent module. The data storage unit is responsible for storing and managing power data, which is a database. To query the database, an SQL statement needs to be input. The SQL statement is generated by the agent module. The agent module includes four agents. The data input unit communicates with the first agent. After the user inputs data from the data input unit, the first agent will deeply understand the user's query intention and requirements. The first agent has the ability of natural language understanding, which enables the first agent to extract key information from the data collected from the data input unit and convert it into a data query language. The second agent parses the data input by the user from the data input unit. The second agent uses an intention recognition algorithm to parse the query intention in the data. The third agent extracts key information from the data input by the user from the data input unit. The key information includes a time range, specific parameters, or data fields. The fourth agent obtains the data query language generated by the first agent, the query intention parsed by the second agent, and the key information extracted by the third agent. The fourth agent generates a corresponding SQL query statement based on these data. The fourth agent has the ability to convert a natural language query into a database query language. Through the above four agents, the data input by the user from the user interface is converted into an SQL query statement. The SQL query statement is input into the data storage unit, and corresponding SQL query results will be generated in the data storage unit. The service customization output module includes a customization processing engine and an output middleware. The customization processing engine obtains the SQL query results from the data processing module. The customization processing engine performs secondary processing on the results according to service requirements and finally generates customized query results. The query results are in a format that is easy to understand and operate, specifically charts, reports, and specific data views. The output middleware obtains the customized query results generated by the customization processing engine and processes the customized query results.

2. The power question number query system based on multi-agent fusion according to claim 1, wherein: The agent module includes a memory module and an execution module. Each of the four agents includes a memory module and an execution module. The memory module is responsible for storing memory information of previous queries. The execution module, according to these memory information and the specific requirements of the current task, guides the large model to perform targeted task output through a prompt template.

3. A power question number query system based on multi-agent fusion according to claim 1, characterized in that: The agent module also includes a fine-tuning model library. Professional knowledge and data in the power industry are introduced through the fine-tuning model library, including training the model on industry terms, concepts, and business processes to ensure that the generated query results are highly relevant to power operations.

4. A power question number query system based on multi-agent fusion according to claim 1, characterized in that: The output middleware communicates with the data display unit. The output middleware transfers the customized query results to the data display unit, and the data display unit displays the data and presents the customized query results to the user.

5. A power question number query system based on multi-agent fusion according to claim 1, characterized in that: Before the SQL query statement generated by the fourth intelligent agent is input into the data storage unit, the generated SQL query statement is input into the output middleware, and the output middleware inputs the SQL query statement into the data display unit. When the generated SQL query statement is correct and the user does not perform any operations, the unmodified SQL query statement will be sent back to the data processing module through the output middleware again and input into the data storage unit. When the generated SQL query statement is incorrect, the user modifies it at the data display unit and then sends the modified SQL query statement back to the data processing module through the output middleware again, and inputs the new SQL query statement into the data storage unit.

6. The power query system based on multi-agent fusion according to claim 1, wherein: When the data processing module processes the data collected by the data input unit and requires additional information, the user inputs the additional information from the data input unit, and the additional information is directly given to the second intelligent agent. The second intelligent agent analyzes the query intention from the additional information and transfers the query intention to the third intelligent agent to facilitate the third intelligent agent to extract key information.

Images