A water conservancy business dynamic response method and system based on a large language model

Abstract

The application provides a water conservancy business dynamic response method and system based on a large language model, and belongs to the technical field of water conservancy, and comprises the following steps: collecting water conservancy business execution rules, establishing a water conservancy business process description standard according to the business execution rules, using a large language model to manage water conservancy components, and connecting the process standard and the component platform with the large language model; the large language model learns these information and masters the construction rules of the water conservancy business. When the large language model receives a business requirement, a representation chain of the water conservancy business is generated based on the learned rules, and the water conservancy business application is generated in a representation chain-based mode. The application solves the problem that the existing water conservancy application is difficult to adapt to the dynamically changing water conservancy business requirements after delivery due to fixed architecture, thereby improving the flexibility and practicality of the application.

Description

Technical Field

[0001] This invention belongs to the field of water conservancy technology, specifically relating to a dynamic response method and system for water conservancy operations based on a large language model. Background Technology

[0002] As a vital component of the national economy, the water conservancy sector encompasses key operations such as flood control, irrigation district management, and water resource allocation. With the increasing prominence of global climate change and water scarcity, the complexity and dynamism of water conservancy operations are becoming increasingly apparent. Therefore, efficiently addressing these challenges has become a key focus of water conservancy informatization development.

[0003] While existing dynamic response methods for water conservancy utilize advanced technologies such as real-time monitoring and big data, they still suffer from insufficient flexibility. Current water conservancy operations are typically based on fixed rules and procedures, with pre-defined logic initially meeting specific operational needs. However, the water conservancy industry exhibits significant seasonality and regional variations. Different time periods and regions impose different requirements on water resource management and allocation. For example, flood warnings and responses require dynamic adjustments to strategies based on real-time meteorological and hydrological data, while drought management demands timely assessment of water demand and adjustments to water supply plans. The needs of various water conservancy scenarios constantly change with environmental and temporal variations. The dynamic and complex nature of the water conservancy industry makes it difficult for such fixed-architecture systems to cope with real-time environmental changes, resulting in the system's inability to adjust strategies promptly in the face of different hydrological conditions. This fixed-logic development model leads to slow system response and a lack of flexibility when dealing with complex scenarios, failing to meet the demands of modern water resource management for rapid response and dynamic adjustment.

[0004] Therefore, in order to solve the problem of insufficient flexibility of existing water conservancy business systems in responding to dynamic changes, there is an urgent need for a dynamic response method for water conservancy business based on a large language model. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a dynamic response method for water conservancy operations based on a large language model. The method aims to construct a large language model capable of autonomously learning water conservancy business rules, thereby dynamically generating business statements. Based on the business statement chain, specific business applications are built, providing a technical solution to the problem of insufficient flexibility in existing water conservancy business systems when responding to dynamic changes.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A dynamic response method for water conservancy operations based on a large language model includes the following steps:

[0008] Collect water conservancy business execution rules and establish water conservancy business process description standards based on these rules.

[0009] The system utilizes a large language model to integrate water conservancy components, provides a unified access address for component registration, integrates all component services, and builds a component platform.

[0010] By using a large language model to learn the water conservancy business description standard and to understand the relationship between components and business in the component platform, a learned large language model is obtained.

[0011] After learning, the large language model generates a business expression chain based on the user's water conservancy business needs;

[0012] Based on the business description chain, dynamically respond to user needs and generate business applications.

[0013] Preferably, the collection of water conservancy business execution rules specifically includes: collection of business description rules, business expansion rules, component input and output rules, and the relationship between components and businesses; logical rules are divided according to business types, and subject domains are constructed for each business area, and corresponding water conservancy business process description standards are formulated based on the subject domain types.

[0014] Preferably, the method of using a large language model to integrate water conservancy components also includes users independently adding components according to specific business needs, and the large language model performing semantic governance on the components based on their inputs, outputs, and application functions.

[0015] Preferably, the step of using a large language model to learn the water conservancy business description standard and understand the relationship between components and businesses in the component platform specifically includes:

[0016] The rules and relationships within each subject domain are organized and labeled according to the refined business requirements. The labeled descriptions are then fine-tuned using the QLoRA method. During the business application generation process, the large language model uses few-shot and Least-To-Most strategies to serve as reference information for executing business operations based on user-submitted questions. The corresponding business rules are then constructed as prompt templates to serve as contextual information for the large language model, resulting in a learned large language model.

[0017] Preferably, the learned large language model generates a business expression chain based on the user's water conservancy business needs, specifically including:

[0018] By recognizing intent, the business process description under the corresponding subject domain is found, and the business process is dynamically constructed based on the prompts. According to the user intent category, the corresponding components are selected from the component platform to build a business component set, and the operation logic of water conservancy business is generated into a business description chain in the form of serial or parallel connection of components. The description chain includes the called components, the connection relationship between components, input and output parameters, data source, data type and system function.

[0019] Preferably, the step of dynamically responding to user needs based on the business description chain and generating business applications specifically includes:

[0020] Users modify the data nodes in the description chain according to their actual needs. The system backend adjusts the node data according to the input and output constraints based on the content of the description chain, and filters the database tables of the business system according to the system's functions and component types, ultimately realizing the construction of applications for business needs. When water conservancy business needs change, additions, deletions or modifications are made based on the existing business description chain, thereby realizing the dynamic updating of business system functions.

[0021] This invention also provides a dynamic response system for water conservancy operations based on a large language model, comprising:

[0022] The annotation module is used to collect water conservancy business execution rules and establish water conservancy business process description standards based on these rules.

[0023] The platform construction module is used to manage water conservancy components using a large language model, provide a unified access address for component registration, integrate all component services, and build a component platform.

[0024] The learning module is used to learn the water conservancy business description standard using a large language model and understand the relationship between components and business in the component platform, so as to obtain a learned large language model.

[0025] The expression chain generation module is used to generate business expression chains based on the user's water conservancy business needs after the large language model has been learned.

[0026] The application generation module is used to dynamically respond to user needs based on the business description chain and generate business applications.

[0027] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement any of the steps in the dynamic response method for water conservancy operations based on a large language model.

[0028] The present invention also provides a computer-readable storage medium storing a computer program, which, when loaded by a processor, is capable of executing any of the steps in the water conservancy business dynamic response method based on a large language model.

[0029] The dynamic response method for water conservancy operations based on a large language model provided by this invention has the following beneficial effects:

[0030] This invention introduces a large language model to autonomously learn water conservancy business processes, enabling the system to dynamically and efficiently generate corresponding business description chains when dealing with complex and ever-changing water conservancy business scenarios. Users can flexibly adjust each node in the business description chain according to actual needs, ensuring that the business process accurately matches the current situation. This mechanism significantly improves the ability of existing water conservancy applications to handle emergencies and environmental changes, while bringing users higher decision-making efficiency and operational flexibility. Attached Figure Description

[0031] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a flowchart of the dynamic response method for water conservancy operations based on a large language model according to Embodiment 1 of the present invention;

[0033] Figure 2 This is a schematic diagram of the business description chain of the present invention. Detailed Implementation

[0034] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0035] Example 1

[0036] This invention provides a dynamic response method for water conservancy operations based on a large language model, specifically as follows: Figure 1 As shown, it includes the following steps:

[0037] S1: Collect water conservancy business execution rules and establish water conservancy business process description standards based on the business execution rules.

[0038] Collect execution rules related to water conservancy operations, including business description rules, business extension rules, component input / output rules, and the relationship between components and operations. For ease of management, these logical rules are categorized by business type and corresponding subject domains are constructed, such as: flood and drought disaster management, irrigation area water allocation, ecological compensation, forecasting and early warning, water rights trading and allocation regulation, etc., and corresponding water conservancy business process description standards are formulated based on the subject domain types.

[0039] a. Business description rules, including but not limited to the following aspects: the theme, content, type, and scope of application of the business, clearly defining the basic attributes and application scenarios of the business.

[0040] b. Business expansion rules, which mainly involve the relationship between this business and other related businesses, explaining its scalability and dependencies, so as to facilitate cross-business collaboration.

[0041] c. Component input / output rules, which define in detail the purpose, scope of application, required input data types, and output results of each component.

[0042] d. The relationship between components and business functions describes how a specific business function calls and schedules different components to achieve business functions.

[0043] To enhance the understanding of the large language model, each topic domain contains detailed descriptive text and keywords. These descriptions are vectorized using an embedding model to generate corresponding topic domain vectors. Next, the business process descriptions under each topic domain are refined and annotated, and then stored in a vector database using the same vectorization process, forming a water conservancy business process description system.

[0044] S2: Utilize a large language model to integrate water conservancy components, provide a unified access address for component registration, integrate all component services, and build a component platform.

[0045] A unified management platform is provided for water conservancy business components dedicated to the large language model, used to integrate and manage all component services. The managed components include basic components and business components. Basic components mainly include process node components, database query components, and routing components. These components are primarily built into the system. Process node components mainly define the input and output constraints at the beginning and end of the business chain; more specifically, they define what data the user needs to provide at the beginning of the business chain and what data the process needs to output at the end. The database query component mainly uses the large language model to retrieve data based on the user's intent. The routing component mainly controls the direction of the business under different conditions.

[0046] Water conservancy business components can be implemented by users uploading their own developed component interfaces or source code. When uploading component interfaces or source code, it is also necessary to upload the component's description information, including the component's purpose, function, input and output formats, etc. After uploading, the system will parse the component submitted by the user, extract and translate this description information and input and output rules, convert them into JSON Schema format that can be understood by the large language model, and store them in the vector database according to the component category. After parsing and conversion, the business component will be encapsulated into a new RESTful API interface for use in the construction of subsequent business chains.

[0047] S3: Use a large language model to learn the water conservancy business description standard and understand the relationship between components and business in the component platform to obtain a learned large language model.

[0048] Utilizing the previously constructed water conservancy business process description standards (details in S1) and the standard business expression chains generated later based on user operations (details in S4), a large language model is specifically optimized and trained using fine-tuning techniques. Specifically, the process description standards and standard business expression chains undergo data preprocessing, including cleaning, word segmentation, and annotation, to ensure data quality and consistency. After preprocessing, fine-tuning algorithms such as Qlora are used for efficient parameter fine-tuning, effectively saving computational resources while maintaining model performance. During this process, the large language model gradually builds a comprehensive understanding of various water conservancy business domains by learning from the topic domain vectors, business process descriptions, and related annotations stored in the vector database. In practical applications, it makes intelligent and dynamic responses based on specific business needs.

[0049] Building upon this foundation, when addressing specific business requirements, a prompting engineering strategy will be employed to further enhance the intelligence of the large language model in business understanding and component invocation. Specifically, this includes: constructing example datasets based on few-shot and Least-To-Most strategies using running examples of relevant questions; and rapidly learning from a small number of samples for each user's question category to improve the accuracy of question execution. Furthermore, a system task prompt template will be constructed, using relevant water conservancy business rules as contextual information for the current dialogue to constrain and inspire the behavior of the large language model, resulting in a learned large language model.

[0050] S4: After learning, the large language model generates a business expression chain based on the user's water conservancy business needs.

[0051] After the user inputs their requirements, the system first vectorizes and encodes the question. Then, it compares the vectorized question with the annotated cosine similarity of the vectors with the rule data, example data, and component data in the vector library. This retrieves the relevant parts of the question and dynamically constructs them into a prompt set, example set, and component set for the current dialogue. The large language model uses the prompt set as contextual information for the current dialogue and the example set as reference examples for the task. Based on the user's requirements, the large language model provides guidance to help the user refine their business requirements. For example, if the user doesn't mention the specific model algorithm used, the large language model will guide the user to provide supplementary information. Once the user has completed their requirements, the large language model selects the most appropriate components from the component set and concatenates the component information in a business chain format. This presents the operational logic of the water conservancy business as a chain of interconnected or parallel components—a business description chain.

[0052] like Figure 2 As shown, the business description chain can consist of a main chain or a combination of multiple sub-chains. The beginning and middle stages of the chain are described by fixed process components, controlling the input and output constraints of the entire chain. For example, the beginning stage controls the necessary data that the user needs to provide when executing the business, and the end stage controls the data type and format that the business needs to output to achieve the business objective. The remaining parts are completed by piecing together the description information of database query components, routing components, and business components. The description of these components includes basic information about the components, the connection relationships between components, the inputs and outputs of the components, and the description information of the nodes. Each piece of information is specific to the model method and data selected by the business.

[0053] Based on the structure of the link, the business representation chain can be divided into sequential chain and routing chain:

[0054] a. Sequence chain: Organize independent functional components according to their logical relationships to form a clear business execution path, that is, each step in the business process is performed in a predetermined logical order.

[0055] b. Routing Chain: Based on the user's specific needs, the routing component dynamically selects the appropriate sequence chain, that is, to enter different sequence chains to execute business under different scenarios or needs.

[0056] The rules and logic in the business execution process are reflected through sequence chains and routing chains, which guide the decision criteria and priorities for process execution.

[0057] S5: Dynamically respond to user needs based on the business description chain and generate business applications.

[0058] After generating the business statement chain, it can be corrected and optimized based on expert experience. Users can add, delete, and modify components in the statement chain, such as changing the source of the selected data, modifying the model category, and refining the model parameters, to ensure the accuracy and rationality of the business process. Once the user finally confirms the business statement chain, the backend system will organize the components according to the workflow and their connection relationships, generating the business application.

[0059] When a user raises a business question to the application, the initial component of the business statement chain first parses the user's question, extracting the required parameters. After parsing the parameters, the application executes the business logic in a sequential chain. If there are no conditional checks in the business logic, the business components are executed sequentially until the endpoint. The endpoint then summarizes the execution results of all business components and outputs standardized business result data. If the business logic includes conditional checks, the application enters a routing chain. The routing component determines what business logic to perform under what circumstances, thus leading to different sequential chains. The business process can consist of multiple sequential chains and routing chains until the endpoint is reached, yielding standardized business data.

[0060] When the needs of water conservancy operations change, the system provides two response strategies:

[0061] a. Reconstruct the business description chain.

[0062] If the existing expression chain cannot meet the business needs under the change of process, the user can re-upload the business component, or put forward the business construction requirements to the big language model on the existing component. The big language model guides the user to supplement the requirements according to the user's business needs, then selects the components, generates new business expressions, and finally generates new business applications to adapt to the current changed business logic.

[0063] b. Modifications based on the existing representation chain.

[0064] If only some parts of the existing description chain need adjustment, the information of the components can be modified on the basis of the original description chain. This includes replacing the type of model algorithm, modifying the model parameters, adding business judgments, etc., so as to dynamically adjust the business process according to actual needs and update the application's functions.

[0065] Generating water conservancy business applications using a representation chain approach significantly improves system flexibility compared to traditional application architectures, ensuring that business applications can quickly respond to changes and achieve dynamic updates and optimizations of functions.

[0066] In specific engineering fields, users initially propose application building requirements to a large language model: to build a flood evolution system for a specific river section. The process of this invention is as follows:

[0067] (1) Remind users to supplement application building requirements

[0068] Once the large language model receives a user's request, it first retrieves an example set, a hint set, and a component set related to the construction of a flood evolution system from the vector library. The example set includes samples of business description chains for flood evolution systems at various cross sections, the hint set includes construction rules for the flood evolution system, and the component set includes components under the flood evolution annotation category. The retrieved example set and hint set are then populated into a hint template to guide the large language model in parsing the user's question. After parsing the user's request, it determines whether additional requirement construction information is needed and then guides the user to supplement the information. For example, for the requirement of "constructing a flood evolution system for the XX river section," the user will be asked to supplement the data range and the specific model algorithm used, etc.

[0069] (2) Component selection business chain construction

[0070] After the user completes their requirements, the large language model selects components from the component set based on those requirements. Following the business logic of flood evolution, it queries the structural relationships of key river sections. Then, according to hydrological flow relationships, it filters and assembles the data components for the river section's starting point, runoff generation components, and flood evolution components. Default parameter input and output formats are set for each component. For the cross-sectional flood output format, it is typically a time series format, such as [(time, value), (time, value)...]. After setting the information for each component, all components are connected in series and parallel to construct a sequence chain and a routing chain, generating a business expression chain. For this problem, the sequence chain could be data query - runoff generation calculation - cross-sectional flood evolution, and the routing chain could be cross-section 1 using model a for flood evolution, and cross-section 2 using model b for flood evolution.

[0071] (3) Business chain correction generates business applications

[0072] After the business description chain is generated, the user can revise it again. After revision, components are combined according to the connection relationship of the description chain to generate the business application. After the application is generated, the user can ask questions about the application, such as "obtain the flood process of section 2". The system backend will calculate the results according to the chain and convert the results into the output format set by the user.

[0073] This invention proposes a dynamic response method for water conservancy operations based on a large language model. This method aims to construct a large language model capable of autonomously learning water conservancy business processes and logic. In the face of a dynamically changing water conservancy environment, it can dynamically generate water conservancy business expressions, thereby achieving efficient and accurate matching of business applications and enhancing the ability of existing water conservancy applications to respond to emergencies and environmental changes.

[0074] Based on the same inventive concept, this invention also provides a dynamic response system for water conservancy operations based on a large language model, comprising:

[0075] The annotation module is used to collect water conservancy business execution rules and establish water conservancy business process description standards based on these rules.

[0076] The platform construction module is used to manage water conservancy components using a large language model, provide a unified access address for component registration, integrate all component services, and build a component platform.

[0077] The learning module is used to learn the water conservancy business description standard using a large language model and understand the relationship between components and business in the component platform, resulting in a learned large language model.

[0078] The representation chain generation module is used to generate business representation chains based on the user's water conservancy business needs after the large language model has been learned.

[0079] The application generation module is used to dynamically respond to user needs based on the business description chain and generate business applications.

[0080] The modules in the aforementioned dynamic response system for water conservancy operations based on a large language model can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0081] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps in the embodiment of the dynamic response method for water conservancy operations based on a large language model. Specific implementation methods can be found in the method embodiments, and will not be repeated here.

[0082] Furthermore, the present invention also provides a non-transitory computer-readable storage medium containing instructions on which a computer program is stored. For example, a memory containing instructions that can be executed by a processor of a computer device to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc. When the computer program is executed by the processor, it can implement the steps in the embodiment of the dynamic response method for water conservancy operations based on a large language model. Specific implementation methods can be found in the method embodiments, which will not be repeated here.

[0083] Those skilled in the art will understand that embodiments of the present invention can provide methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0085] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0087] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the present invention patent. No reference numerals in the claims should be construed as limiting the scope of the claims. Any simple variations or equivalent substitutions of technical solutions that can be readily obtained by those skilled in the art within the scope of the technology disclosed in the present invention are within the protection scope of the present invention.

Claims

1. A dynamic response method for water conservancy operations based on a large language model, characterized in that, include: Collect water conservancy business execution rules and establish water conservancy business process description standards based on these rules. The system utilizes a large language model to integrate water conservancy components, provides a unified access address for component registration, integrates all component services, and builds a component platform. By using a large language model to learn the water conservancy business description standard and to understand the relationship between components and business in the component platform, a learned large language model is obtained. After learning, the large language model generates a business expression chain based on the user's water conservancy business needs; Dynamically respond to user needs based on the business description chain and generate business applications; The collection of water conservancy business execution rules specifically includes: collection of business description rules, business expansion rules, component input and output rules, and the relationship between components and businesses; logical rules are divided according to business types, and subject domains are constructed for each business area, and corresponding water conservancy business process description standards are formulated based on subject domain types; The method of using a large language model to integrate water conservancy components also includes users adding components independently according to specific business needs, and the large language model performing semantic management on the components based on their inputs, outputs and application functions. The method of using a large language model to learn water conservancy business description standards and understanding the relationship between components and business in the component platform specifically includes: The rules and relationships within each subject domain are organized and labeled according to the refined business requirements. The labeled descriptions are then fine-tuned using the QLoRA method. During the business application generation process, the large language model uses few-shot and Least-To-Most strategies to serve as reference information for executing business operations based on user-submitted questions. The corresponding business rules are then constructed as prompt templates to serve as contextual information for the large language model, resulting in a learned large language model. The learned large language model generates a business expression chain based on the user's water conservancy business needs, specifically including: By recognizing intent, the business process description under the corresponding topic domain is found, and the business process is dynamically constructed based on the prompts. According to the user intent category, the corresponding components are selected from the component platform to build a business component set, and the operation logic of water conservancy business is generated into a business description chain in the form of serial or parallel connection of components. The description chain includes the components called, the connection relationship between components, input and output parameters, data source, data type and system function. The step of dynamically responding to user needs based on the business description chain and generating business applications specifically includes: Users modify the data nodes in the description chain according to their actual needs. The system backend adjusts the node data according to the input and output constraints based on the content of the description chain, and filters the database tables of the business system according to the system's functions and component types, ultimately realizing the construction of applications for business needs. When water conservancy business needs change, additions, deletions or modifications are made based on the existing business description chain, thereby realizing the dynamic updating of business system functions. The business description chain consists of a main chain or a combination of multiple sub-chains. Each part of the chain is described by fixed process components, which control the input and output constraints of the entire chain. The initial segment controls the necessary data that the user needs to provide when executing the business, and the final segment controls the data type and data format that the business needs to output in order to achieve the business goal. The remaining part is completed by piecing together the description information of database query components, routing components, and business components. The description of these components includes the basic information of the components, the connection relationship between the components, the input and output of the components, and the description information of the nodes. Each piece of information is specific to the model method and data selected by the business. Based on the structure of the link, the business description chain is divided into a sequential chain and a routing chain; The rules and logic in the business execution process are reflected through sequence chains and routing chains, which guide the decision criteria and priorities for process execution.

2. A dynamic response system for water conservancy operations based on a large language model, used to implement the method described in claim 1, characterized in that, include: The annotation module is used to collect water conservancy business execution rules and establish water conservancy business process description standards based on these rules. The platform construction module is used to manage water conservancy components using a large language model, provide a unified access address for component registration, integrate all component services, and build a component platform. The learning module is used to learn the water conservancy business description standard using a large language model and understand the relationship between components and business in the component platform, so as to obtain a learned large language model. The expression chain generation module is used to generate business expression chains based on the user's water conservancy business needs after the large language model has been learned. The application generation module is used to dynamically respond to user needs based on the business description chain and generate business applications.

3. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method of claim 1.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is loaded by the processor, it is able to perform the steps of the method of claim 1.

Images