A large model-based hydrological analysis service chain automatic generation method and system

Abstract

The application discloses a hydrological analysis service chain automatic generation method and system based on a large model, and realizes obtaining a hydrological analysis result directly from a natural language description. First, a hydrological geographic information service expression model is proposed, and a hydrological analysis model and data are published as services. Then, based on a large model and a prompt template, a user's natural language description requirement is analyzed, and the most matched service is obtained through similarity calculation. Finally, a service chain is executed through a service chain execution engine, so that a hydrological analysis result is obtained. The application automatically analyzes user requirements and constructs a service chain based on a large model, and obtains a hydrological analysis result through a service chain execution engine, so that the problem of automatic generation of a hydrological analysis service chain is effectively solved, and the automation level of hydrological analysis is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent hydrological analysis, specifically to a method and system for automatically generating hydrological analysis service chains based on large models. Background Technology

[0002] Hydrological analysis models are widely used in water resource management, soil erosion and water conservation, environmental protection, flood management, and other fields, providing decision-makers with reliable data support and scientific basis, thereby promoting sustainable development in these areas. However, most hydrological analysis models are deployed locally, which makes them difficult to share and hinders their widespread application.

[0003] With the development of Web service technology, geographic information services (GIS) have emerged. GIS, based on Web service architecture and standards, utilizes computer technology to provide functions such as querying, analyzing, and visualizing geospatial data in a network environment. It features excellent sharing, interoperability, reusability, and scalability. Therefore, based on GIS technology, publishing data and models as data services and processing services respectively can effectively improve the sharing capabilities of data and models, thereby supporting hydrological analysis tasks.

[0004] However, existing research has focused less on automated hydrological analysis, leading to a high degree of human intervention. Furthermore, some applications require the collaborative work of multiple hydrological analysis models, but currently these tasks mainly rely on manual model assembly and input data processing, increasing the user's workload. With the development of artificial intelligence technology, obtaining hydrological analysis results directly based on user needs has become an urgent requirement in the field of intelligent hydrological analysis.

[0005] Currently, existing methods for automatically generating service chains are mainly based on ontology, artificial intelligence planning, and other methods. These methods rely on formalized expressions of user requirements, requiring users to fill in their requirements according to corresponding templates, which increases the user's workload. Summary of the Invention

[0006] This invention addresses the issue of insufficient attention to automated hydrological analysis in existing research, leading to high levels of human intervention. It proposes an automated generation method for hydrological analysis service chains based on a large model, directly obtaining hydrological analysis results based on user-defined natural language requirements. Specifically, it first defines an expression model for hydrological geographic information services, including hydrological geographic information data services, hydrological geographic information processing services, and hydrological geographic information composite services. Then, based on the large model and prompt templates, it parses the user's natural language description requirements, calculates the similarity to obtain the most matching service, and executes it in the service chain execution engine to obtain the hydrological analysis results, thereby improving the level of automation in hydrological analysis.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0008] The first aspect provides a method for automatically generating hydrological analysis service chains based on large models, including:

[0009] S1: Publish hydrological analysis models as hydrological geographic information processing services, publish hydrological analysis data as hydrological geographic information data services, and publish combined models of hydrological analysis models that complete specific tasks as hydrological geographic information composite services.

[0010] S2: Based on the attribute information of hydrological geographic information processing service and hydrological geographic information composite service, the embedding representation of the service is generated through a large model and stored in a vector database. The embedding representation of hydrological geographic information processing service and hydrological geographic information composite service includes the semantic association and similarity between the services.

[0011] S3: Design a standardized requirements template for hydrological analysis, including administrative region, requirements, and year;

[0012] S4: Utilize standardized demand templates for hydrological analysis to prompt and guide large-scale models, transforming user demands described in natural language into structured demands;

[0013] S5: Input the transformed structured demand into the large model, generate the user demand embedding, and select the target hydrological and geographic information processing service or hydrological and geographic information composite service based on the similarity between the user demand embedding and the service embedding stored in the vector database.

[0014] S6: Based on the input parameter information of the target hydrological geographic information processing service or hydrological geographic information composite service, and the parsed user requirements, select hydrological geographic information data services that meet the conditions from the service registration center. The service registration center pre-stores a large number of hydrological geographic information data services.

[0015] S7: Combine the target hydrological geographic information processing service or hydrological geographic information composite service with hydrological geographic information data service into a hydrological analysis service chain;

[0016] S8: Input the hydrological analysis service chain into the service chain execution engine to obtain the hydrological analysis results.

[0017] In one implementation, the hydrological geographic information processing service includes hydrological raster data service and hydrological vector data service.

[0018] In one implementation, the attributes of a hydrological raster data service include a unique identifier for the hydrological raster data service, the name of the hydrological raster data service, a description of the hydrological raster data service, the data source type of the hydrological raster data service, and the formats supported by the hydrological raster data service. The attributes of a hydrological vector data service include a unique identifier for the hydrological vector data service, the name of the hydrological vector data service, a description of the hydrological vector data service, and the formats supported by the hydrological vector data service.

[0019] In one implementation, the attribute information of the hydrological geographic information processing service includes a unique identifier for the hydrological geographic information processing service, the name of the hydrological geographic information processing service, the description information of the hydrological geographic information processing service, the set of input parameters for the hydrological geographic information processing service, the set of output parameters for the hydrological geographic information processing service, and the type of the hydrological geographic information processing service. The attribute information of the hydrological geographic information composite service includes a unique identifier representing the hydrological geographic information composite service, the name of the hydrological geographic information composite service, the description information of the hydrological geographic information composite service, the set of input parameters for the hydrological geographic information composite service, the set of output parameters for the hydrological geographic information composite service, and the set of hydrological geographic information processing services contained in the hydrological geographic information composite service.

[0020] In one implementation, a target hydrological geographic information processing service or a hydrological geographic information composite service is selected based on the similarity between the user's required embedding and the service embeddings stored in the vector database, including:

[0021] Calculate the cosine similarity between the user demand embedding and the service embedding stored in the vector database;

[0022] Select the hydrological geographic information processing service or hydrological geographic information composite service with the highest similarity as the target hydrological geographic information processing service or hydrological geographic information composite service.

[0023] In one implementation, the input parameter information of the target hydrological geographic information processing service or hydrological geographic information composite service includes the name, data type, and parameter description of the input parameter.

[0024] Based on the same inventive concept, a second aspect of the present invention provides an automatic generation system for hydrological analysis service chains based on large models, comprising:

[0025] The service publishing module is used to publish hydrological analysis models as hydrological geographic information processing services, hydrological analysis data as hydrological geographic information data services, and combined models of hydrological analysis models that complete specific tasks as hydrological geographic information composite services.

[0026] The processing service and composite service embedding generation module is used to generate the embedding representation of hydrogeographic information processing service and hydrogeographic information composite service through a large model based on the attribute information of hydrogeographic information processing service and hydrogeographic information composite service, and store it in the vector database. The embedding representation of hydrogeographic information processing service and hydrogeographic information composite service includes the semantic correlation and similarity between hydrogeographic information services.

[0027] The standardized requirements template design module is used to design standardized requirements templates for hydrological analysis, which include administrative regions, requirements, and years.

[0028] The user demand structure transformation module is used to transform user demands described in natural language into structured demands by using standardized demand templates for hydrological analysis to prompt and guide the large model.

[0029] The service matching module is used to input the transformed structured requirements into the large model, generate the user requirement embedding, and select the target hydrological geographic information processing service or hydrological geographic information composite service based on the similarity between the user requirement embedding and the hydrological geographic information service embedding stored in the vector database.

[0030] The data service matching module is used to select hydrological and geographic information data services that meet the conditions from the service registry center based on the input parameter information of the target hydrological and geographic information processing service or hydrological and geographic information composite service and the parsed user needs. The service registry center pre-stores a large number of hydrological and geographic information data services.

[0031] The hydrological analysis service chain combination module is used to combine target hydrological geographic information processing services or hydrological geographic information composite services, as well as hydrological geographic information data services, into a hydrological analysis service chain.

[0032] The hydrological analysis results acquisition module is used to input the hydrological analysis service chain into the service chain execution engine to obtain the hydrological analysis results.

[0033] Based on the same inventive concept, a third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the automatic generation method for hydrological analysis service chains based on large models as described in the first aspect.

[0034] Based on the same inventive concept, the fourth aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the automatic generation method for hydrological analysis service chains based on large models as described in the first aspect.

[0035] Compared with the prior art, the advantages and beneficial technical effects of the present invention are as follows:

[0036] To address the limited research on automated hydrological analysis and the high degree of human intervention, this paper proposes an automatic generation method for hydrological analysis service chains supported by a large-scale model. This method enables the direct deriving of hydrological analysis results from natural language-described requirements. First, considering the diversity and non-standardization of user requirements, a requirement understanding method based on a large-scale model is proposed. Guided by a designed requirement template, user requirements are automatically parsed into standardized requirement solutions. Then, considering the service matching problem under zero-sample conditions, services and user requirements are embedded as vectors using the large-scale model. The most suitable hydrological geographic information processing service or hydrological geographic information composite service is obtained by calculating similarity, and hydrological geographic information data services are selected under these constraints, thus obtaining a complete hydrological analysis service chain. Finally, the hydrological analysis service chain is input into the service chain execution engine to obtain the hydrological analysis results, effectively reducing human intervention and improving the efficiency of hydrological analysis. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart of the method for automatically generating hydrological analysis service chains based on large models according to an embodiment of the present invention;

[0039] Figure 2 This is a framework diagram of the hydrological analysis service chain automatic generation system based on a large model, according to an embodiment of the present invention. Detailed Implementation

[0040] This invention proposes an automatic generation method for hydrological analysis service chains based on a large model, enabling the direct deriving of hydrological analysis results from natural language descriptions of user needs. First, a hydrological geographic information service expression model is proposed, and the hydrological analysis model and data are published as hydrological geographic information services. Then, based on the large model and prompt templates, the user's natural language description of their needs is parsed, and the most matching hydrological geographic information service is obtained by calculating similarity. Finally, the service chain is run through a service chain execution engine to obtain the hydrological analysis results. This invention automatically parses user needs and constructs service chains based on a large model, and obtains hydrological analysis results through a service chain execution engine, thus effectively solving the problem of automated generation of hydrological analysis service chains and significantly improving the automation level of hydrological analysis.

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example 1

[0043] This invention discloses an automatic generation method for hydrological analysis service chains based on large models, comprising:

[0044] S1: Publish hydrological analysis models as hydrological geographic information processing services, publish hydrological analysis data as hydrological geographic information data services, and publish combined models of hydrological analysis models that complete specific tasks as hydrological geographic information composite services.

[0045] S2: Based on the attribute information of hydrological geographic information processing service and hydrological geographic information composite service, the embedding representation of the service is generated through a large model and stored in a vector database. The embedding representation of hydrological geographic information processing service and hydrological geographic information composite service includes the semantic association and similarity between the services.

[0046] S3: Design a standardized requirements template for hydrological analysis, including administrative region, requirements, and year;

[0047] S4: Utilize standardized demand templates for hydrological analysis to prompt and guide large-scale models, transforming user demands described in natural language into structured demands;

[0048] S5: Input the transformed structured demand into the large model, generate the user demand embedding, and select the target hydrological geographic information processing service or hydrological geographic information composite service based on the similarity between the user demand embedding and the hydrological geographic information service embedding stored in the vector database.

[0049] S6: Based on the input parameter information of the target hydrological geographic information processing service or hydrological geographic information composite service, and the parsed user requirements, select hydrological geographic information data services that meet the conditions from the service registration center. The service registration center pre-stores a large number of hydrological geographic information data services.

[0050] S7: Combine the target hydrological geographic information processing service or hydrological geographic information composite service with hydrological geographic information data service into a hydrological analysis service chain;

[0051] S8: Input the hydrological analysis service chain into the service chain execution engine to obtain the hydrological analysis results.

[0052] Specifically, step S1 can utilize service publishing strategies to publish hydrological analysis models, hydrological analysis data, and combined models of hydrological analysis models as hydrological geographic information services, thereby improving the sharing and interoperability of models and data. Considering that hydrological analysis involves multiple models such as hydrological analysis models, image cropping, and projection, publishing them as hydrological geographic information processing services through service publishing strategies enhances model sharing and interoperability. Considering that hydrological analysis involves various data such as remote sensing images, DEMs, and administrative regions, publishing them as hydrological geographic information data services through service publishing strategies improves data sharing capabilities. Considering that hydrological analysis tasks typically require multiple models to complete, publishing them as composite hydrological geographic information services through service publishing strategies enhances the functionality of the service registry.

[0053] Step S2 generates embedding representations of hydrological geographic information processing services and hydrological geographic information composite services through a large model based on the attribute information, and stores these embeddings in a vector database.

[0054] Step S3 considers the diversity and non-standardization of user needs derived from large models, and designs a standardized requirement template for the field of hydrological analysis, which includes key information such as administrative region, requirement, and year.

[0055] Step S4 addresses the user's needs described in natural language by using the requirement template prompts and guiding models designed in step S3, thereby transforming the natural language-described requirements into a structured requirement form.

[0056] Step S5 generates the embedding of user needs parsed in step S3 using a large model, calculates the similarity between it and the embedding of hydrological geographic information services stored in the vector database, and selects the hydrological geographic information processing service or hydrological geographic information composite service with the highest similarity.

[0057] Step S6 selects hydrological data services that meet the constraints from the service registry based on the input parameter information of the hydrological geographic information processing service or the hydrological geographic information composite service, as well as the parsed user requirements.

[0058] Step S7 combines the hydrological geographic information processing service or hydrological geographic information composite service obtained in step S5 with the hydrological geographic information data service obtained in step S6 into a hydrological analysis service chain.

[0059] Step S8 inputs the hydrological analysis service chain obtained in step S7 into the service chain execution engine to obtain the hydrological analysis results.

[0060] In a specific implementation process, a flowchart of an automatic generation method for hydrological analysis service chains based on large models is shown below. Figure 1 As shown, the main steps include the following:

[0061] Step 1: Define the hydrological geographic information processing service representation model. Based on the service publishing strategy, publish the hydrological analysis-related models as hydrological geographic information processing services. The definition of the hydrological geographic information processing service representation model is as follows:

[0062]

[0063] In equation (1), ID PS A unique identifier representing a hydrological geographic information processing service; URL PS Name represents the link address for the hydrological geographic information processing service; PS Indicates the name of the hydrological geographic information processing service; Title FS Indicates the title of the hydrological geographic information processing service; Descp PS Input represents descriptive information for hydrological geographic information processing services. PS This represents the set of input parameters for hydrological geographic information processing services, including parameter names, parameter types, and parameter descriptions; Output PS This represents the set of output parameters for hydrological and geographic information processing services. PS Indicates the type of hydrological geographic information processing service (such as WPS, RESTful, and SOAP services), Ver PS This indicates the version of the hydrological geographic information processing service, Prov. PS GCon represents the provider of hydrological geographic information processing services. PS Represents the constraint set of hydrological geographic information processing services, QoS PS This indicates the service quality information of hydrological and geographic information processing services.

[0064] Step two involves defining the hydrological geographic information data service representation model and, based on the service publishing strategy, publishing the hydrological analysis-related models as hydrological geographic information data services. Hydrological geographic information data services are divided into hydrological raster data services and hydrological vector data services. The definition of the hydrological raster data service representation model is as follows:

[0065]

[0066] In equation (2), ID RDS A unique identifier for a hydrological raster data service; URL RDS Name represents the link address to the hydrological raster data service. RDSIndicates the name of the hydrological raster data service; Title RDS This indicates the title of the hydrological raster data service; Descp RDS Descriptive information representing hydrological raster data services; TFeat RDS Indicates time characteristics, including time reference, time, year, season, and date; SFeat RDS Represents spatial characteristics, including spatial reference, spatial extent, and spatial resolution; Type RDS This indicates the type of hydrological raster data service, including services such as WCS and WMS; SType RDS This indicates the data source type for hydrological raster data services, including UAV imagery and satellite remote sensing imagery; Ver RDS Indicates the version of the hydrological raster data service; Prov RDS Indicates the provider of hydrological raster data services; Format RDS Indicates the formats supported by the hydrological raster data service; QoS RDS This indicates the service quality information of the hydrological raster data service.

[0067] The definition of the hydrological vector data service representation model is as follows:

[0068]

[0069] In equation (3), ID VDS A unique identifier representing a hydrological vector data service; URL VDS Name represents the link address of the hydrological vector data service; VDS Indicates the name of the hydrological vector data service; Title VDS Indicates the title of the hydrological vector data service; Descp VDS Descriptive information representing hydrological vector data services; TFeat VDS Indicates time characteristics, including time reference, time, year, season, and date; SFeat VDS Represents spatial characteristics, including spatial reference, spatial extent, and spatial resolution; Type VDS This indicates the type of hydrological data service, including services such as WFS and WMS; Ver VDS Indicates the version of the hydrological vector data service; Prov VDS Indicates the provider of hydrological vector data services; Format VDS Indicates the formats supported by the hydrological vector data service; QoS VDS This indicates the service quality information for hydrological vector data services.

[0070] Step three: Considering that hydrological analysis tasks typically require multiple models to complete, a hydrological-geographic information composite service representation model is defined. This model is then published as a hydrological-geographic information composite service using a service publishing strategy. The definition of the hydrological-geographic information composite service representation model is as follows:

[0071]

[0072] In equation (4), ID CS A unique identifier representing a composite hydrological and geographic information service; Name CS Indicates the name of the hydrological and geographic information composite service; Title CS Indicates the title of the hydrological and geographic information composite service; Descp CS Descriptive information representing hydrological and geographic information composite services; Input CS This represents the set of input parameters for the hydrological and geographic information composite service, including parameter name, parameter type, and parameter description; Output CS This represents the output parameter set of the hydrological and geographic information composite service; QoS CS This refers to the service quality information of hydrological and geographic information composite services, HAPSs. CS ParamRels represents the collection of hydrological and geographic information processing services included in a hydrological and geographic information composite service. CS A set representing the relationships between parameters in a hydrological and geographic information composite service.

[0073] Step four involves generating embedding representations of hydrological geographic information processing services and composite hydrological geographic information services using a large model, based on their attribute information. These embeddings encapsulate the semantic relationships and similarities between the hydrological geographic information services, providing a foundation for subsequent service matching.

[0074] Step 5: Design a standardized requirements template for hydrological analysis, {“Administrative Region”: “$”, “Requirement”: “$”, “Year”: “$”}. This template contains key information such as administrative region, requirement, and year for subsequent requirement understanding.

[0075] Step Six: In response to the user's hydrological analysis needs described in natural language, the requirements template designed in Step Five is used to prompt and guide the large model, thereby transforming the natural language description of the requirements into a structured requirement template form.

[0076] Step 7: Generate the structured user demand embeddings parsed in Step 6 using the large model, calculate the cosine similarity between these embeddings and the service embeddings stored in the vector database, and select the hydrological and geographic information processing service or the hydrological and geographic information composite service with the highest similarity. The definition of cosine similarity is as follows:

[0077]

[0078] In equation (5), A and B represent two vectors, and ||A|| / and ||B|| represent the lengths of vectors A and B, respectively.

[0079] Step 8: Based on the input parameter information of the hydrological geographic information processing service or hydrological geographic information composite service obtained in Step 7, and the user requirements parsed in Step 6, select hydrological geographic information data services that meet the constraints from the service registry. The service constraints mainly include the data type of the input parameters.

[0080] Step 9: Combine the hydrological geographic information processing service or hydrological geographic information composite service obtained in Step 7 with the hydrological geographic information data service obtained in Step 8 into a hydrological analysis service chain.

[0081] Step nine: Bind the hydrological geographic information data service obtained in step eight to the input parameters corresponding to the hydrological geographic information processing service or hydrological geographic information composite service obtained in step seven, thereby forming a hydrological analysis service chain.

[0082] Step 10: Input the hydrological analysis service chain obtained in Step 9 into the service chain execution engine, and call the corresponding service through the service chain execution engine to obtain the hydrological analysis results.

[0083] Example 2

[0084] Based on the same inventive concept, this embodiment discloses an automatic generation system for hydrological analysis service chains based on large models. Please refer to [link to relevant documentation]. Figure 2 ,include:

[0085] Service publishing module 201 is used to publish hydrological analysis models as hydrological geographic information processing services, hydrological analysis data as hydrological geographic information data services, and combined models of hydrological analysis models that complete specific tasks as hydrological geographic information composite services.

[0086] The processing service and composite service embedding generation module 202 is used to generate the embedding representation of the hydrogeographic information processing service and the hydrogeographic information composite service through a large model based on the attribute information of the hydrogeographic information processing service and the hydrogeographic information composite service, and store it in the vector database. The embedding representation of the hydrogeographic information processing service and the hydrogeographic information composite service includes the semantic association and similarity between the services.

[0087] Standardized Requirements Template Design Module 203 is used to design standardized requirements templates for hydrological analysis, including administrative regions, requirements, and years.

[0088] The User Requirement Structured Transformation Module 204 is used to transform user requirements described in natural language into structured requirements by using standardized requirement templates for hydrological analysis to prompt and guide the large model.

[0089] The service matching module 205 is used to input the transformed structured demand into a large model, generate the user demand embedding, and select the target hydrogeographic information processing service or hydrogeographic information composite service based on the similarity between the user demand embedding and the service embedding stored in the vector database.

[0090] The data service matching module 206 is used to select hydrological and geographic information data services that meet the conditions from the service registration center based on the input parameter information of the target hydrological and geographic information processing service or hydrological and geographic information composite service and the parsed user requirements. The service registration center pre-stores a large number of hydrological and geographic information data services.

[0091] The hydrological analysis service chain combination module 207 is used to combine the target hydrological geographic information processing service or hydrological geographic information composite service and hydrological geographic information data service into a hydrological analysis service chain.

[0092] The hydrological analysis result acquisition module 208 is used to input the hydrological analysis service chain into the service chain execution engine to obtain the hydrological analysis results.

[0093] Since the system described in Embodiment 2 of this invention is the same system used to implement the automatic generation method of hydrological analysis service chain based on a large model in Embodiment 1 of this invention, those skilled in the art can understand the specific structure and variations of this system based on the method described in Embodiment 1 of this invention, and therefore will not be repeated here. All systems used in the method of Embodiment 1 of this invention fall within the scope of protection of this invention.

[0094] Example 3

[0095] Based on the same inventive concept, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in Embodiment 1.

[0096] Since the computer-readable storage medium described in Embodiment 3 of this invention is the same computer-readable storage medium used in implementing the automatic generation method of hydrological analysis service chain based on a large model in Embodiment 1 of this invention, those skilled in the art can understand the specific structure and variations of this computer-readable storage medium based on the method described in Embodiment 1 of this invention, and therefore will not be repeated here. All computer-readable storage media used in the method of Embodiment 1 of this invention fall within the scope of protection of this invention.

[0097] Example 4

[0098] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in Embodiment 1.

[0099] Since the computer device described in Embodiment 4 of this invention is the same computer device used to implement the automatic generation method of hydrological analysis service chain based on large model in Embodiment 1 of this invention, those skilled in the art can understand the specific structure and variations of this computer device based on the method described in Embodiment 1 of this invention, and therefore will not be described again here. All computer devices used in the method of Embodiment 1 of this invention fall within the scope of protection of this invention.

[0100] Those skilled in the art will understand that embodiments of the present invention can be provided as 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.

[0101] 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, and 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 illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0102] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various modifications and variations to the embodiments of the invention without departing from the spirit and scope of the invention. Thus, if these modifications and variations of the embodiments of the invention fall within the scope of the claims of the invention and their equivalents, the invention also intends to include these modifications and variations.

Claims

1. A method for automatically generating hydrological analysis service chains based on large models, characterized in that, include: S1: Publish hydrological analysis models as hydrological geographic information processing services, publish hydrological analysis data as hydrological geographic information data services, and publish combined models of hydrological analysis models that complete specific tasks as hydrological geographic information composite services. S2: Based on the attribute information of hydrological geographic information processing service and hydrological geographic information composite service, the embedding representation of the service is generated through a large model and stored in a vector database. The embedding representation of hydrological geographic information processing service and hydrological geographic information composite service includes the semantic association and similarity between the services. S3: Design a standardized requirements template for hydrological analysis, including administrative region, requirements, and year; S4: Utilize standardized demand templates for hydrological analysis to prompt and guide large-scale models, transforming user demands described in natural language into structured demands; S5: Input the transformed structured demand into the large model, generate the user demand embedding, and select the target hydrological geographic information processing service or hydrological geographic information composite service based on the similarity between the user demand embedding and the hydrological geographic information service embedding stored in the vector database. S6: Based on the input parameter information of the target hydrological geographic information processing service or hydrological geographic information composite service, and the parsed user requirements, select hydrological geographic information data services that meet the conditions from the service registration center. The service registration center pre-stores a large number of hydrological geographic information data services. S7: Combine the target hydrological geographic information processing service or hydrological geographic information composite service with hydrological geographic information data service into a hydrological analysis service chain; S8: Input the hydrological analysis service chain into the service chain execution engine to obtain the hydrological analysis results.

2. The method for automatically generating hydrological analysis service chains based on large models as described in claim 1, characterized in that, Hydrological geographic information data services include hydrological raster data services and hydrological vector data services.

3. The method for automatically generating hydrological analysis service chains based on large models as described in claim 2, characterized in that, The attributes of a hydrological raster data service include its unique identifier, name, description, data source type, and supported formats. The attributes of a hydrological vector data service include its unique identifier, name, description, and supported formats.

4. The method for automatically generating hydrological analysis service chains based on large models as described in claim 1, characterized in that, The attribute information of a hydrological geographic information processing service includes its unique identifier, name, description, input parameter set, output parameter set, and type. The attribute information of a hydrological geographic information composite service includes its unique identifier, name, description, input parameter set, output parameter set, and the set of hydrological geographic information processing services it contains.

5. The method for automatically generating hydrological analysis service chains based on large models as described in claim 1, characterized in that, Based on the similarity between the user's required embedding and the service embeddings stored in the vector database, the target hydrological and geographic information processing service or hydrological and geographic information composite service is selected, including: Calculate the cosine similarity between the user demand embedding and the service embedding stored in the vector database; Select the hydrological geographic information processing service or hydrological geographic information composite service with the highest similarity as the target hydrological geographic information processing service or hydrological geographic information composite service.

6. The method for automatically generating hydrological analysis service chains based on large models as described in claim 1, characterized in that, The input parameter information for the target hydrological geographic information processing service or hydrological geographic information composite service includes the name, data type, and parameter description of the input parameter.

7. An automatic generation system for hydrological analysis service chains based on large models, characterized in that, include: The service publishing module is used to publish hydrological analysis models as hydrological geographic information processing services, hydrological analysis data as hydrological geographic information data services, and combined models of hydrological analysis models that complete specific tasks as hydrological geographic information composite services. The module for generating embeddings for processing services and composite services is used to generate embedding representations of hydrogeographic information processing services and hydrogeographic information composite services through a large model based on the attribute information of the hydrogeographic information processing services and hydrogeographic information composite services, and store them in a vector database. The embedding representations of hydrogeographic information processing services and hydrogeographic information composite services include the semantic association and similarity between the services. The standardized requirements template design module is used to design standardized requirements templates for hydrological analysis, which include administrative regions, requirements, and years. The user demand structure transformation module is used to transform user demands described in natural language into structured demands by using standardized demand templates for hydrological analysis to prompt and guide the large model. The processing service and composite service matching module is used to input the transformed structured requirements into the large model, generate the user requirement embedding, and select the target hydrogeographic information processing service or hydrogeographic information composite service based on the similarity between the user requirement embedding and the service embedding stored in the vector database. The data service matching module is used to select hydrological and geographic information data services that meet the conditions from the service registry center based on the input parameter information of the target hydrological and geographic information processing service or hydrological and geographic information composite service and the parsed user needs. The service registry center pre-stores a large number of hydrological and geographic information data services. The hydrological analysis service chain combination module is used to combine target hydrological geographic information processing services or hydrological geographic information composite services, as well as hydrological geographic information data services, into a hydrological analysis service chain. The hydrological analysis results acquisition module is used to input the hydrological analysis service chain into the service chain execution engine to obtain the hydrological analysis results.

8. The automatic generation system for hydrological analysis service chains based on large models as described in claim 7, characterized in that, The processing service and composite service matching module is specifically used for: Calculate the cosine similarity between the user demand embedding and the service embedding stored in the vector database; Select the hydrological geographic information processing service or hydrological geographic information composite service with the highest similarity as the target hydrological geographic information processing service or hydrological geographic information composite service.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the automatic generation method for hydrological analysis service chains based on large models as described in any one of claims 1 to 6.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the automatic generation method for hydrological analysis service chains based on large models as described in any one of claims 1 to 6.

Images