Meteorological data analysis method, device and equipment based on large model agent and medium

Abstract

The invention discloses a meteorological data analysis method, device and equipment based on a large model agent and a medium, relates to the field of meteorological data intelligent management, is applied to a preset meteorological data analysis system, and comprises the following steps: when a main agent receives a to-be-processed meteorological message, determining and calling a target agent from a plurality of extreme weather scene agents; the main agent and the plurality of extreme weather scene agents are agents constructed by using a target large language model in advance; the target agent matches a first cue word template of the target agent based on a received main agent instruction so as to determine and call a target meteorological data acquisition tool; the target meteorological data acquisition tool matches a second cue word template of the target meteorological data acquisition tool based on a received target agent instruction so as to acquire meteorological data based on the determined target interface parameters; and the target agent determines a target answer based on the data acquisition result, so that the main agent replies a message in a dialogue form. And the limitation of the existing meteorological data application can be overcome.

Description

Technical Field

[0001] The present invention relates to the field of intelligent management of meteorological data, and in particular to a meteorological data analysis method, device, equipment and medium based on a large model intelligent body. Background Art

[0002] Currently, ChatBI (Chat-based Business Intelligence) technology enables users to query and analyze data through natural language conversations, making data analysis more intuitive and easy to use. However, due to the difficulties and low efficiency in obtaining meteorological data, as well as the constraints of data confidentiality, its application in meteorological data is still in the initial exploration stage and cannot be effectively applied.

[0003] Therefore, how to overcome the limitations of existing meteorological data applications and achieve efficient use of meteorological data is an urgent problem to be solved. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a meteorological data analysis method, device, equipment and medium based on a large model intelligent agent, which can effectively overcome the limitations of existing meteorological data applications, improve the convenience and efficiency of meteorological data acquisition, and improve the accuracy and reliability of answers. The specific scheme is as follows:

[0005] In a first aspect, the present application provides a meteorological data analysis method based on a large model agent, which is applied to a preset meteorological data analysis system, wherein the large model agent is a large model agent built based on a chatbot and business intelligence functions; wherein the method comprises:

[0006] When a weather message to be processed inputted by a client is received through a main agent, a target agent is determined and called from a plurality of extreme weather scenario agents based on the weather message to be processed; the main agent and the plurality of extreme weather scenario agents are agents pre-constructed using a target large language model;

[0007] The target agent matches its own first prompt word template based on the received main agent instruction, and determines and calls the target meteorological data acquisition tool according to the obtained first template matching result;

[0008] The target meteorological data acquisition tool matches its own second prompt word template based on the received target agent instruction, and determines the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters to obtain data acquisition results;

[0009] The target agent performs analysis based on the received data acquisition result and the first template matching result to determine the corresponding target answer, and returns the target answer to the main agent so that the main agent can reply to the message using the target answer in the form of a dialogue.

[0010] Optionally, determining and calling a target agent from a plurality of extreme weather scenario agents based on the meteorological message to be processed includes:

[0011] Analyzing the to-be-processed meteorological message by the main agent to determine whether to call an extreme weather scenario agent;

[0012] When the first judgment result is yes, a target agent is determined from a plurality of extreme weather scenario agents based on the weather scenario type corresponding to the to-be-processed meteorological message, and the target agent is called.

[0013] Optionally, the target agent matches its own first prompt word template based on the received main agent instruction, and determines and calls the target meteorological data acquisition tool according to the obtained first template matching result, including:

[0014] Receiving, through the target agent, a master agent instruction corresponding to the weather message to be processed;

[0015] Matching the first prompt word template of the main agent based on the main agent instruction to obtain a corresponding first template matching result; the first prompt word template includes role prompt information, task prompt information, input parameter prompt information and question and answer prompt information;

[0016] The corresponding target meteorological data acquisition tool name is determined by analyzing the first template matching result, and the target meteorological data acquisition tool name is used to call the target meteorological data acquisition tool.

[0017] Optionally, the target meteorological data acquisition tool matches its own second prompt word template based on the received target agent instruction, and determines the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters, including:

[0018] Receiving corresponding target intelligent agent instructions through the target meteorological data acquisition tool;

[0019] Matching the second prompt word template of the target agent based on the target agent instruction to obtain a second template matching result;

[0020] Determine target interface parameters and first target answer related fields by analyzing the second template matching result;

[0021] Based on the target interface parameters, the corresponding data service interface is accessed, and the data service interface and the first target answer-related fields are used to connect to the meteorological big data platform and acquire meteorological data to obtain data acquisition results; the data acquisition results include the second target answer-related fields and answer-related data.

[0022] Optionally, the analyzing by the target agent based on the received data acquisition result and the first template matching result to determine a corresponding target answer includes:

[0023] Receiving, through the target intelligent agent, the data acquisition result corresponding to the target intelligent agent instruction returned by the target meteorological data acquisition tool;

[0024] Answer information corresponding to the meteorological message to be processed is generated by analyzing the data acquisition result and the first template matching result to obtain a target answer.

[0025] Optionally, the master agent uses the target answer to reply to the message in the form of a dialogue, including:

[0026] The main agent determines that the target answer meets a preset condition to obtain a second determination result;

[0027] If the second judgment result indicates that the preset condition is met, the to-be-processed weather message is replied to in a dialogue form based on the target answer.

[0028] Optionally, after obtaining the second judgment result, the method further includes:

[0029] If the second judgment result indicates that the preset condition is not met, the instruction information corresponding to the meteorological message to be processed is regenerated, and the obtained new main intelligent agent instruction is sent to the target intelligent agent, so as to jump again to the step of matching the first prompt word template of the target intelligent agent based on the received main intelligent agent instruction.

[0030] In a second aspect, the present application provides a meteorological data analysis device based on a large model agent, which is applied to a preset meteorological data analysis system, wherein the large model agent is a large model agent built based on a chatbot and business intelligence functions; wherein the device comprises:

[0031] An agent calling module is used to determine and call a target agent from multiple extreme weather scenario agents based on the to-be-processed weather message received from the client through the main agent; the main agent and the multiple extreme weather scenario agents are agents pre-constructed using a target large language model;

[0032] A tool calling module, used for matching the first prompt word template of the target agent based on the received main agent instruction, and determining and calling the target meteorological data acquisition tool according to the obtained first template matching result;

[0033] A data acquisition module, used to match its own second prompt word template based on the received target agent instruction through the target meteorological data acquisition tool, and determine the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters to obtain data acquisition results;

[0034] The message reply module is used to analyze the received data acquisition result and the first template matching result through the target agent to determine the corresponding target answer, and return the target answer to the main agent so that the main agent can use the target answer to reply the message in the form of a dialogue.

[0035] In a third aspect, the present application provides an electronic device, including:

[0036] Memory, used to store computer programs;

[0037] A processor is used to execute the computer program to implement the steps of the aforementioned meteorological data analysis method based on a large model intelligent agent.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the steps of the aforementioned large-model-agent-based meteorological data analysis method.

[0039] It can be seen that in the present application, in the preset meteorological data analysis system, when a to-be-processed meteorological message input by a client is received through a main intelligent agent, a target intelligent agent is determined and called from a plurality of extreme weather scene intelligent agents based on the to-be-processed meteorological message; the main intelligent agent and the plurality of extreme weather scene intelligent agents are intelligent agents pre-constructed using a target large language model; the target intelligent agent matches its own first prompt word template based on the received main intelligent agent instruction, and determines and calls a target meteorological data acquisition tool based on the obtained first template matching result; the target meteorological data acquisition tool matches its own second prompt word template based on the received target intelligent agent instruction, and determines the target interface parameters based on the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters, so as to obtain a data acquisition result; the target intelligent agent performs analysis based on the received data acquisition result and the first template matching result to determine the corresponding target answer, and returns the target answer to the main intelligent agent, so that the main intelligent agent can reply to a message in the form of a dialogue using the target answer. That is, in the present application, when the main agent in the preset meteorological data analysis system receives a meteorological message to be processed, it will trigger the corresponding extreme weather scene agent calling operation, and then the called target agent will use the instruction sent by the main agent to match its own prompt word template to call the meteorological data acquisition tool, and then the called target meteorological data acquisition tool call will use the instruction sent by the target agent to match its own prompt word template to collect meteorological data and return the result to the target agent. After the target agent generates an answer using the data acquisition result and returns it to the main agent, the main agent uses the target answer to reply to the message in the form of a dialogue. In this way, the limitations of existing meteorological data applications can be effectively overcome, and the convenience and efficiency of meteorological data acquisition are improved, and the accuracy and reliability of the answers are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0041] Figure 1 A flow chart of a meteorological data analysis method based on a large model agent provided in this application;

[0042] Figure 2 A schematic diagram of a specific process of meteorological data analysis based on a large model intelligent agent provided in this application;

[0043] Figure 3 A schematic diagram of the structure of a meteorological data analysis device based on a large model intelligent agent provided in this application;

[0044] Figure 4 A structural diagram of an electronic device provided for this application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] At present, ChatBI technology enables users to query and analyze data through natural language dialogue, making data analysis more intuitive and easy to use. However, due to the difficulties in obtaining meteorological data and the low efficiency of acquisition, as well as the constraints of data confidentiality, its application in meteorological data is still in the initial exploration stage and cannot be effectively applied. To this end, the present application provides a meteorological data analysis solution based on a large model intelligent agent, which can effectively overcome the limitations of existing meteorological data applications, improve the convenience and efficiency of meteorological data acquisition, and improve the accuracy and reliability of answers.

[0047] See also Figure 1 As shown, an embodiment of the present invention discloses a meteorological data analysis method based on a large model agent, which is applied to a preset meteorological data analysis system, wherein the large model agent is a large model agent built based on a chatbot and business intelligence functions; wherein the method comprises:

[0048] Step S11: When a weather message to be processed is received from a client through a main agent, a target agent is determined and called from a plurality of extreme weather scenario agents based on the weather message to be processed; the main agent and the plurality of extreme weather scenario agents are agents pre-constructed using a target large language model.

[0049] Combination Figure 2 As shown, in this embodiment, when the user inputs a weather message to be processed into the system, the main agent (i.e., Figure 2 That is, the weather message to be processed is received from multiple extreme weather scenario agents (i.e., the master agent in the Figure 2The method comprises: analyzing the to-be-processed meteorological message by the main agent to determine whether to call the extreme weather scenario agent; when the first judgment result is yes, determining the target agent from multiple extreme weather scenario agents based on the weather scenario type corresponding to the to-be-processed meteorological message, and calling the target agent. That is, in this embodiment, the main agent first determines whether to call the extreme weather scenario agent, and if it is determined to call, determines and calls the target agent.

[0050] It should be understood that the extreme weather scene agents include blizzard agents, severe convection agents, rainstorm agents and cold wave agents. The master agent is responsible for routing and controlling the extreme weather scene agents. Moreover, each agent is supported by the same large language model (i.e., the target large language model) according to a specific arrangement and combination. The master agent is responsible for coordinating the autonomous extreme weather scene agents so that the extreme weather scene agents can efficiently collaborate and perform complex tasks.

[0051] Regarding the extreme weather scenario agents, each of the extreme weather scenario agents can have its own prompts, LLM (Large Language Model), tools and other custom codes to collaborate with other agents. However, the same LLM can also play different roles based on the provided prompts. The prompt word templates of each extreme weather scenario agent include role prompts, task prompts, input parameter prompts and question-answer prompts. Each of the extreme weather scenario agents has multiple data acquisition tools for connecting to the meteorological data platform, which are divided into two types, data query tools and data statistical feature tools.

[0052] Taking the rainstorm agent as an example, its prompt word template can be as follows:

[0053] Role Tips: Heavy Rain Scene

[0054] Task Tips: Accurate query and statistical analysis of rainstorm scene data

[0055] Input parameter prompts: time, location, ground data elements

[0056] Question and answer prompt: Basic situation / statistical characteristics of ground data elements related to heavy rain scenes at a certain time and place.

[0057] Regarding the target large language model, the main agent and the multiple independent extreme weather scenario agents all use the target large language model, but each has different prompts, thus forming a self-reflective AI (Artificial Intelligence) agent. This method of using the same large language model in a loop with multiple different roles is constructed through the Lang Graph framework. The Lang Graph framework can also be used to create multi-agent workflows. Just like in the self-reflective AI agent, the target large language model can play multiple roles, each role acting as a different AI agent, which is the concept of multi-agent.

[0058] Additionally, it is important to understand that the connections between agents are represented by edges. Each connecting edge can have a control condition that guides the flow of information from one agent to another. Each agent has a state that can be updated with information during each flow.

[0059] Step S12: the target agent matches its own first prompt word template based on the received main agent instruction, and determines and calls the target meteorological data acquisition tool according to the obtained first template matching result.

[0060] Combination Figure 2 As shown, in this embodiment, after calling the target agent, the target agent determines and calls a meteorological data acquisition tool (i.e., the Figure 2 ). That is, the target agent matches its own first prompt word template based on the received main agent instruction, and determines and calls the target meteorological data acquisition tool according to the obtained first template matching result, including: receiving the main agent instruction corresponding to the to-be-processed meteorological message through the target agent; matching its own first prompt word template based on the main agent instruction to obtain the corresponding first template matching result; the first prompt word template includes role prompt information, task prompt information, input parameter prompt information and question-answer prompt information; determining the corresponding target meteorological data acquisition tool name by analyzing the first template matching result, and calling the target meteorological data acquisition tool using the target meteorological data acquisition tool name. That is, the called target agent matches the main agent instruction with the prompt word template, and abstracts the specific tool name of the data tool to be called. It can be understood that different extreme weather scenario agents can have different numbers of meteorological data acquisition tools.

[0061] It should be understood that, regarding the design of the meteorological data acquisition tool, the data acquisition tool for a specific extreme weather scenario connects the meteorological big data cloud platform of the corresponding region with the extreme weather scenario agent through the data service interface, and acquires and analyzes data according to the business scenario requirements. Each of the meteorological data acquisition tools has its own data tool name, data tool parameters, data tool description prompt words, and data sources and interfaces for connecting to the meteorological big data cloud platform. The meteorological data acquisition tool implements field extraction of the returned data according to the data tool function definition, analyzes statistical calculations and processing, and then outputs the numerical value to the user in the form of a dialogue. Among them, the data tool name must be an agent data tool name that is easy for the large language model agent tool to call and identify intent, the data tool parameters must be data tool parameters that are easy for the large language model agent to extract parameters based on conversational questions, and the data tool description prompt words must be description prompt words that are easy for the large model agent tool to call and identify intent, thereby reducing the misjudgment rate of the agent tool.

[0062] The design example of the meteorological data acquisition tool can be seen in Table 1 below:

[0063] Table 1

[0064]

[0065] In addition, in this embodiment, regarding the prompt word template, when designing the template, it can be oriented to the rainstorm, heavy snow, severe convection and cold wave disaster weather scenes, and the meteorological big data interface, algorithm, data source, data field and other disaster meteorological data elements can be sorted out for the meteorological data platform. Prompt word templates and annotation specifications for disaster meteorological data elements are formulated, and standardized prompt word annotation and standardized management of disaster meteorological data elements are carried out to realize intelligent retrieval and recommendation services for disaster meteorological data elements in various forms such as data, calculation, components, and modules.

[0066] Step S13, the target meteorological data acquisition tool matches its own second prompt word template based on the received target intelligent agent instruction, and determines the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters to obtain data acquisition results.

[0067] In this embodiment, the designated target meteorological data acquisition tool is called, which matches the target intelligent agent instruction with the prompt word template, abstracts the specific parameter value corresponding to the data service interface, and the fields required for the answer, and connects to the meteorological big data platform by accessing the data service interface to obtain the fields and data required to answer the user's question. That is, the target meteorological data acquisition tool matches its own second prompt word template based on the received target intelligent agent instruction, and determines the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters, including: receiving the corresponding target intelligent agent instruction through the target meteorological data acquisition tool; matching its own second prompt word template based on the target intelligent agent instruction to obtain the second template matching result; determining the target interface parameters and the first target answer related fields by analyzing the second template matching result; accessing the corresponding data service interface based on the target interface parameters, and using the data service interface and the first target answer related fields to connect to the meteorological big data platform and acquire meteorological data to obtain data acquisition results; the data acquisition results include the second target answer related fields and answer related data. It can be understood that different meteorological data acquisition tools can have different numbers and functions of data service interfaces.

[0068] Step S14: The target agent analyzes the received data acquisition result and the first template matching result to determine the corresponding target answer, and returns the target answer to the main agent so that the main agent can reply to the message in the form of a dialogue using the target answer.

[0069] In this embodiment, the target intelligent agent generates a corresponding answer based on the data acquisition result returned by the router and the prompt word template. That is, the target intelligent agent performs analysis based on the received data acquisition result and the first template matching result to determine the corresponding target answer, including: receiving the data acquisition result corresponding to the target intelligent agent instruction returned by the target meteorological data acquisition tool through the target intelligent agent; generating answer information corresponding to the meteorological message to be processed by analyzing the data acquisition result and the first template matching result to obtain the target answer.

[0070] It should be understood that, in this embodiment, if the target answer fed back by the target agent to the main agent is the final answer to the weather message to be processed, the main agent will feed back the target answer to the user. Otherwise, the main agent may send instructions to the extreme weather scene agent again. That is, the main agent uses the target answer to reply to the message in the form of a dialogue, including: judging that the target answer meets the preset conditions through the main agent to obtain a second judgment result; if the second judgment result shows that the preset conditions are met, the weather message to be processed is replied to in the form of a dialogue based on the target answer. In addition, after obtaining the second judgment result, it also includes: if the second judgment result shows that the preset conditions are not met, the instruction information corresponding to the weather message to be processed is regenerated, and the new main agent instruction is sent to the target agent to jump back to the step of matching the first prompt word template of the target agent based on the received main agent instruction.

[0071] In summary, the technical solution described in this embodiment can have the following beneficial effects:

[0072] (1) Filling the technical gap in the application of ChatBI technology in the meteorological field;

[0073] (2) It can more accurately understand the user's natural language instructions and convert them into a data query language that is easy to process. It also supports access to multiple data sources and provides a variety of data analysis and visualization tools to meet the user's multi-dimensional needs for meteorological data, making the meteorological data analysis process more intuitive and efficient.

[0074] (3) Supporting real-time data access, users can query the latest meteorological data at any time and make timely decisions, making meteorological data analysis more flexible and accurate;

[0075] (4) Allow users to query and analyze data through dialogue, improving the convenience and efficiency of obtaining meteorological data. It can not only quickly answer questions, but also automatically generate various charts and reports to help users better understand meteorological data.

[0076] (5) The security of sensitive information that may be involved in meteorological data can be protected through the private deployment of large models and data permission control, thereby ensuring data security and privacy protection;

[0077] (6) Through dialogue, intelligent data processing and analysis can be achieved, human intervention can be reduced, and the efficiency and accuracy of data analysis can be improved.

[0078] It can be seen that in the embodiment of the present application, in the preset meteorological data analysis system, when a to-be-processed meteorological message input by a client is received through a main intelligent agent, a target intelligent agent is determined and called from a plurality of extreme weather scene intelligent agents based on the to-be-processed meteorological message; the main intelligent agent and the plurality of extreme weather scene intelligent agents are intelligent agents pre-constructed using a target large language model; the target intelligent agent matches its own first prompt word template based on the received main intelligent agent instruction, and determines and calls a target meteorological data acquisition tool based on the obtained first template matching result; the target meteorological data acquisition tool matches its own second prompt word template based on the received target intelligent agent instruction, and determines the target interface parameters based on the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters, so as to obtain a data acquisition result; the target intelligent agent performs analysis based on the received data acquisition result and the first template matching result to determine the corresponding target answer, and returns the target answer to the main intelligent agent, so that the main intelligent agent can reply to a message in the form of a dialogue using the target answer. That is, in the present application, when the main agent in the preset meteorological data analysis system receives a meteorological message to be processed, it will trigger the corresponding extreme weather scene agent calling operation, and then the called target agent will use the instruction sent by the main agent to match its own prompt word template to call the meteorological data acquisition tool, and then the called target meteorological data acquisition tool call will use the instruction sent by the target agent to match its own prompt word template to collect meteorological data and return the result to the target agent. After the target agent generates an answer using the data acquisition result and returns it to the main agent, the main agent uses the target answer to reply to the message in the form of a dialogue. In this way, the limitations of existing meteorological data applications can be effectively overcome, and the convenience and efficiency of meteorological data acquisition are improved, and the accuracy and reliability of the answers are improved.

[0079] See also Figure 3 As shown, the embodiment of the present application also discloses a meteorological data analysis device based on a large model agent, which is applied to a preset meteorological data analysis system. The large model agent is a large model agent built based on a chatbot and business intelligence functions; wherein the device includes:

[0080] The agent calling module 11 is used to determine and call a target agent from multiple extreme weather scene agents based on the to-be-processed weather message received from the client through the main agent; the main agent and the multiple extreme weather scene agents are agents pre-built using the target large language model;

[0081] The tool calling module 12 is used to match the first prompt word template of the target agent based on the received main agent instruction, and determine and call the target meteorological data acquisition tool according to the obtained first template matching result;

[0082] The data acquisition module 13 is used to match the second prompt word template of the target meteorological data acquisition tool based on the received target agent instruction, and determine the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters to obtain data acquisition results;

[0083] The message reply module 14 is used to analyze the received data acquisition result and the first template matching result through the target agent to determine the corresponding target answer, and return the target answer to the main agent so that the main agent can use the target answer to reply the message in the form of a dialogue.

[0084] Among them, for more specific working processes of the above-mentioned modules, please refer to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0085] It can be seen that in this application, when the main agent in the preset meteorological data analysis system receives a meteorological message to be processed, it will trigger the corresponding extreme weather scene agent call operation, and then the called target agent will use the instruction sent by the main agent to match its own prompt word template to call the meteorological data acquisition tool, and then the called target meteorological data acquisition tool call will use the instruction sent by the target agent to match its own prompt word template to collect meteorological data and return the result to the target agent. After the target agent generates an answer using the data acquisition result and returns it to the main agent, the main agent uses the target answer to reply to the message in the form of a dialogue. In this way, the limitations of existing meteorological data applications can be effectively overcome, and the convenience and efficiency of meteorological data acquisition are improved, and the accuracy and reliability of the answers are improved.

[0086] In some specific embodiments, the agent calling module 11 can be specifically used to: analyze the meteorological message to be processed by the main agent to determine whether to call the extreme weather scene agent; when the first judgment result is yes, determine the target agent from multiple extreme weather scene agents based on the weather scene type corresponding to the meteorological message to be processed, and call the target agent.

[0087] In some specific embodiments, the tool calling module 12 can be specifically used to: receive the main agent instruction corresponding to the meteorological message to be processed through the target agent; match its own first prompt word template based on the main agent instruction to obtain the corresponding first template matching result; the first prompt word template includes role prompt information, task prompt information, input parameter prompt information and question and answer prompt information; determine the corresponding target meteorological data acquisition tool name by analyzing the first template matching result, and use the target meteorological data acquisition tool name to call the target meteorological data acquisition tool.

[0088] In some specific embodiments, the data acquisition module 13 can be specifically used to: receive corresponding target intelligent agent instructions through the target meteorological data acquisition tool; match its own second prompt word template based on the target intelligent agent instructions to obtain a second template matching result; determine the target interface parameters and the first target answer related fields by analyzing the second template matching results; access the corresponding data service interface based on the target interface parameters, and use the data service interface and the first target answer related fields to connect to the meteorological big data platform and acquire meteorological data to obtain data acquisition results; the data acquisition results include second target answer related fields and answer related data.

[0089] In some specific embodiments, the message reply module 14 can be specifically used to: receive, through the target intelligent agent, the data acquisition result corresponding to the target intelligent agent's instruction and returned by the target meteorological data acquisition tool; generate answer information corresponding to the meteorological message to be processed by analyzing the data acquisition result and the first template matching result to obtain the target answer.

[0090] In some specific embodiments, the message reply module 14 can be specifically used to: determine through the main intelligent agent whether the target answer meets the preset conditions to obtain a second judgment result; if the second judgment result indicates that the preset conditions are met, reply to the pending weather message based on the target answer and in the form of a dialogue.

[0091] In some specific embodiments, the meteorological data analysis device based on the large model intelligent agent can also be used to: if the second judgment result indicates that the preset condition is not met, regenerate the instruction information corresponding to the meteorological message to be processed, and send the obtained new main intelligent agent instruction to the target intelligent agent, so as to jump again to the step of matching the first prompt word template of the target intelligent agent based on the received main intelligent agent instruction.

[0092] Furthermore, the present application also discloses an electronic device. Figure 4This is a structural diagram of an electronic device 2 / 30 according to an exemplary embodiment. The content in the diagram cannot be regarded as any limitation on the scope of use of the present application.

[0093] Figure 4 A schematic diagram of the structure of an electronic device 2 / 30 provided in an embodiment of the present application. The electronic device 2 / 30 may specifically include: at least one processor 2 / 31, at least one memory 2 / 32, a power supply 2 / 33, a communication interface 2 / 34, an input / output interface 2 / 35, and a communication bus 2 / 36. The memory 2 / 32 is used to store a computer program, which is loaded and executed by the processor 2 / 31 to implement the relevant steps in the meteorological data analysis method based on a large model agent disclosed in any of the aforementioned embodiments. In addition, the electronic device 2 / 30 in this embodiment may specifically be an electronic computer.

[0094] In this embodiment, the power supply 2 / 33 is used to provide working voltage for each hardware device on the electronic device 2 / 30; the communication interface 2 / 34 can create a data transmission channel between the electronic device 2 / 30 and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 2 / 35 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0095] In addition, the memory 2 / 32, as a carrier for resource storage, can be a read-only memory, random access memory, disk or CD, etc. The resources stored thereon can include an operating system 2 / 321, a computer program 2 / 322, etc. The storage method can be temporary storage or permanent storage.

[0096] The operating system 2 / 321 is used to manage and control the hardware devices and computer programs 2 / 322 on the electronic device 2 / 30, which may be Windows Server, Netware, Unix, Linux, etc. In addition to the computer program that can be used to complete the meteorological data analysis method based on the large model agent executed by the electronic device 2 / 30 disclosed in any of the aforementioned embodiments, the computer program 2 / 322 may further include computer programs that can be used to complete other specific tasks.

[0097] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the aforementioned large-model agent-based meteorological data analysis method is implemented. The specific steps of the method can be referred to the corresponding contents disclosed in the aforementioned embodiments, and will not be repeated here.

[0098] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0099] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0100] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0101] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0102] The technical solution provided by the present application is introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technicians in this field, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A meteorological data analysis method based on a large model agent, characterized in that: Applied to a preset meteorological data analysis system, the large model agent is a large model agent built based on a chatbot and business intelligence functions; wherein the method comprises: When a weather message to be processed inputted by a client is received through a main agent, a target agent is determined and called from a plurality of extreme weather scenario agents based on the weather message to be processed; the main agent and the plurality of extreme weather scenario agents are agents pre-constructed using a target large language model; The target agent matches its own first prompt word template based on the received main agent instruction, and determines and calls the target meteorological data acquisition tool according to the obtained first template matching result; The target meteorological data acquisition tool matches its own second prompt word template based on the received target agent instruction, and determines the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters to obtain data acquisition results; The target agent performs analysis based on the received data acquisition result and the first template matching result to determine the corresponding target answer, and returns the target answer to the main agent so that the main agent can reply to the message using the target answer in the form of a dialogue.

2. The meteorological data analysis method based on a large model agent according to claim 1 is characterized in that: The step of determining and calling a target agent from a plurality of extreme weather scenario agents based on the meteorological message to be processed includes: Analyzing the to-be-processed meteorological message by the main agent to determine whether to call an extreme weather scenario agent; When the first judgment result is yes, a target agent is determined from a plurality of extreme weather scenario agents based on the weather scenario type corresponding to the to-be-processed meteorological message, and the target agent is called.

3. The meteorological data analysis method based on a large model agent according to claim 1 is characterized in that: The target agent matches its own first prompt word template based on the received main agent instruction, and determines and calls the target meteorological data acquisition tool according to the obtained first template matching result, including: Receiving, through the target agent, a master agent instruction corresponding to the weather message to be processed; Matching the first prompt word template of the main agent based on the main agent instruction to obtain a corresponding first template matching result; the first prompt word template includes role prompt information, task prompt information, input parameter prompt information and question and answer prompt information; The corresponding target meteorological data acquisition tool name is determined by analyzing the first template matching result, and the target meteorological data acquisition tool name is used to call the target meteorological data acquisition tool.

4. The meteorological data analysis method based on a large model agent according to claim 1 is characterized in that: The target meteorological data acquisition tool matches its own second prompt word template based on the received target agent instruction, and determines the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters, including: Receiving corresponding target intelligent agent instructions through the target meteorological data acquisition tool; Matching the second prompt word template of the target agent based on the target agent instruction to obtain a second template matching result; Determine target interface parameters and first target answer related fields by analyzing the second template matching result; Based on the target interface parameters, the corresponding data service interface is accessed, and the data service interface and the first target answer-related fields are used to connect to the meteorological big data platform and acquire meteorological data to obtain data acquisition results; the data acquisition results include the second target answer-related fields and answer-related data.

5. The meteorological data analysis method based on a large model agent according to claim 1 is characterized in that: The target agent performs analysis based on the received data acquisition result and the first template matching result to determine the corresponding target answer, including: Receiving, through the target intelligent agent, the data acquisition result corresponding to the target intelligent agent instruction returned by the target meteorological data acquisition tool; Answer information corresponding to the meteorological message to be processed is generated by analyzing the data acquisition result and the first template matching result to obtain a target answer.

6. The meteorological data analysis method based on large model agent according to claim 1 is characterized in that: The master agent uses the target answer to reply to the message in the form of a dialogue, including: The main agent determines that the target answer meets a preset condition to obtain a second determination result; If the second judgment result indicates that the preset condition is met, the to-be-processed weather message is replied to in a dialogue form based on the target answer.

7. The meteorological data analysis method based on large model agent according to claim 6 is characterized in that: After obtaining the second judgment result, the method further includes: If the second judgment result indicates that the preset condition is not met, the instruction information corresponding to the meteorological message to be processed is regenerated, and the obtained new main intelligent agent instruction is sent to the target intelligent agent, so as to jump again to the step of matching the first prompt word template of the target intelligent agent based on the received main intelligent agent instruction.

8. A meteorological data analysis device based on a large model agent, characterized in that: Applied to a preset meteorological data analysis system, the large model agent is a large model agent built based on a chatbot and business intelligence functions; wherein the device comprises: An agent calling module is used to determine and call a target agent from multiple extreme weather scenario agents based on the to-be-processed weather message received from the client through the main agent; the main agent and the multiple extreme weather scenario agents are agents pre-constructed using a target large language model; A tool calling module, used for matching the first prompt word template of the target agent based on the received main agent instruction, and determining and calling the target meteorological data acquisition tool according to the obtained first template matching result; A data acquisition module, used to match its own second prompt word template based on the received target agent instruction through the target meteorological data acquisition tool, and determine the target interface parameters according to the obtained second template matching result, so as to acquire meteorological data based on the target interface parameters to obtain data acquisition results; The message reply module is used to analyze the received data acquisition result and the first template matching result through the target agent to determine the corresponding target answer, and return the target answer to the main agent so that the main agent can use the target answer to reply the message in the form of a dialogue.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, used to execute the computer program to implement the meteorological data analysis method based on a large model intelligent agent as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, implements the meteorological data analysis method based on a large model intelligent agent as described in any one of claims 1 to 7.

Images