Research and development service platform, service method, electronic equipment and storage medium

Abstract

The invention provides a research and development service platform and method, electronic equipment and a storage medium, and the method comprises a data processing module which is used for receiving data from a plurality of data sources; the task execution engine is connected with the data processing module and is used for receiving a user request, performing task decomposition on the user request to obtain a plurality of sub-tasks, extracting related data from the data processing module according to the type of each sub-task, and scheduling corresponding function sub-units, enabling the function subunits to execute corresponding subtasks based on the related data; the security control module is connected with the data processing module and the task execution engine and is used for providing a model calling strategy for the task execution engine; wherein the task execution engine selects to call the public model or the private model to execute the sub-task according to the model call strategy when the task execution engine schedules the function sub-unit to execute the corresponding sub-task. By means of the method and device, the research and development efficiency is improved, and it is ensured that the core data is safe and controllable in the process.

Description

Technical Field

[0001] This application relates to the field of enterprise R&D technology, and more specifically, to an R&D service platform, service method, electronic device and storage medium. Background Technology

[0002] In enterprise R&D, existing technical solutions typically build independent data interfaces, model services, and permission systems for specific departments or single scenarios. This model leads to different business units repeatedly building basic capabilities, resulting in significant resource waste. At the same time, due to the closed architecture, dedicated models trained in a certain scenario or business logic developed in another scenario are difficult to directly migrate or reuse in other scenarios, resulting in isolated capabilities and a narrow scope of application. Summary of the Invention

[0003] In view of the above, the purpose of this application is to provide a research and development service platform, service method, electronic device and storage medium, which aims to overcome at least one of the above-mentioned defects.

[0004] Firstly, this application provides a research and development service platform, the platform comprising: The data processing module is used to receive data from multiple data sources; The task execution engine, connected to the data processing module, is used to receive user requests, decompose the user requests into multiple sub-tasks, extract relevant data from the data processing module according to the type of each sub-task, and schedule the corresponding functional sub-units to execute the corresponding sub-tasks based on the relevant data. A security control module, connected to the data processing module and the task execution engine, is used to provide a model invocation strategy for the task execution engine; When scheduling the functional subunits to execute corresponding subtasks, the task execution engine selects to call a public model or a private model to execute the subtask according to the model invocation strategy.

[0005] In one possible implementation, the task execution engine includes: The request analysis unit is used to identify the task type requested by the user. A task orchestration unit, connected to the request analysis unit, is used to parse the user request into multiple sub-tasks according to the task type; A functional unit includes multiple functional sub-units, each of which is used to perform at least one sub-task among data retrieval, statistical analysis, logical reasoning, scheme generation, or simulation prediction. The scheduling and integration unit is connected to the task orchestration unit and the functional unit respectively, and is used to schedule the functional sub-units to execute corresponding sub-tasks and integrate the output results of all functional sub-units into the final result.

[0006] In one possible implementation, the security control module includes: A data security classification unit, connected to the data processing module, is used to perform sensitivity marking on the data received by the data processing module; The model invocation decision unit is connected to the data security classification unit and the task execution engine respectively, and is used to select to invoke a public model or a local private model to execute the subtask based on the sensitivity label of the relevant data involved in each subtask. The operation permission management unit is used to control the user's access permissions and call permissions to the data processing module according to the user's role permissions.

[0007] In one possible implementation, the model invocation decision unit selects to invoke either a public model or a local private model in the following manner: If the data involved in the subtask contains highly sensitive tags, then the local private model is invoked; If the data involved in the subtask contains low-sensitivity tags or the data is publicly available, then the public model is invoked.

[0008] In one possible implementation, the data processing module includes: The data access unit is used to collect data from multiple internal and external data sources; The data governance unit, connected to the data access unit, is used to perform structuring, vectorization, or graphing processing on the collected data to provide data services for different sub-task types.

[0009] In one possible implementation, the task execution engine further includes: The intelligent agent generation unit, connected to the scheduling and integration unit, is used to encapsulate the task flow executed by the scheduling and integration unit into a reusable dedicated intelligent agent.

[0010] Secondly, this application provides a research and development service method, the method comprising: The data processing module is used to receive data from multiple data sources; The task execution engine is connected to the data processing module and is used to receive user requests, decompose the user requests into multiple sub-tasks, extract relevant data from the data processing module according to the type of each sub-task, and schedule the corresponding functional sub-units to execute the corresponding sub-tasks based on the relevant data. The security control module is connected to the data processing module and the task execution engine, and is used to provide a model invocation strategy for the task execution engine. When scheduling the functional sub-unit to execute the corresponding sub-task, the task execution engine selects to call a public model or a private model to execute the sub-task according to the model invocation strategy.

[0011] In one possible implementation, the task execution engine includes: The request analysis unit is used to identify the task type requested by the user. A task orchestration unit, connected to the request analysis unit, is used to parse the user request into multiple sub-tasks according to the task type; A functional unit includes multiple functional sub-units, each of which is used to perform at least one sub-task among data retrieval, statistical analysis, logical reasoning, scheme generation, or simulation prediction. The scheduling and integration unit is connected to the task orchestration unit and the functional unit respectively, and is used to schedule the functional sub-units to execute corresponding sub-tasks and integrate the output results of all functional sub-units into the final result.

[0012] Thirdly, this application also provides an electronic device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method described above are performed.

[0013] Fourthly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the method described above.

[0014] This application provides a research and development service platform, service method, electronic device, and storage medium. The method includes: a data processing module for receiving data from multiple data sources; a task execution engine connected to the data processing module for receiving user requests, decomposing the user requests into multiple sub-tasks, extracting relevant data from the data processing module according to the type of each sub-task, and scheduling corresponding functional sub-units to execute the corresponding sub-tasks based on the relevant data; and a security control module connected to the data processing module and the task execution engine for providing a model invocation strategy for the task execution engine. When scheduling functional sub-units to execute corresponding sub-tasks, the task execution engine selects to invoke either a public model or a private model to execute the sub-task according to the model invocation strategy. This application improves research and development efficiency and ensures the security and controllability of core data during the process.

[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the structure of a research and development service platform provided in an embodiment of this application; Figure 2 A flowchart illustrating a research and development service method provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0019] First, the applicable scenarios for this application will be introduced. This application can be applied to the field of enterprise R&D technology.

[0020] Research has revealed that in enterprise R&D processes, existing technical solutions typically build independent data interfaces, model services, and permission systems for specific departments or single scenarios. This model leads to different business units repeatedly building basic capabilities, resulting in significant resource waste. Furthermore, due to closed architectures, specialized models trained for a particular scenario or business logic developed for another scenario are difficult to directly migrate or reuse, resulting in isolated capabilities and a narrow application scope. In addition, most platforms are designed from the outset to serve specific domain R&D tasks, lacking underlying versatility and flexible scalability, and thus failing to efficiently support cross-departmental and diverse business needs. From a data perspective, R&D activities rely on diverse data sources and formats (including structured data, unstructured documents, and external intelligence), posing significant technical challenges to existing solutions when integrating multi-source heterogeneous data. More critically, when handling sensitive data involving core enterprise knowledge, a unified, built-in security and compliance framework is often lacking, making it difficult to achieve widespread application of intelligent capabilities while ensuring data privacy and security.

[0021] Based on this, embodiments of this application provide a research and development service platform, service method, electronic device, and storage medium, aiming to overcome at least one of the above-mentioned defects.

[0022] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of a research and development service platform provided in an embodiment of this application. Figure 1 As shown in the figure, the R&D service platform provided in this application embodiment includes a data processing module 101, a task execution engine 102, and a security control module 103.

[0023] Here, the service platform of this application serves as a fundamental, universal, and scalable operating system-level platform supporting the operation of dedicated AI employees. Its operation process is as follows: the user initiates a request; the data processing module 101 retrieves relevant data according to the permission scope; the task execution engine 102 calls the corresponding functional module to perform the task; during this process, if sensitive data is involved, a private model is enabled for local processing according to the policy of the security control module 103; if it is a general query, a public model is called; finally, the system returns the synthesized results (such as text summaries and charts) to the user.

[0024] Specifically, the data processing module 101 is used to receive data from multiple data sources.

[0025] Here, data sources include, but are not limited to: online structured data (such as connecting to the group's data lake to automatically synchronize team project data), online unstructured data (such as integrating with the workbench system to capture uploaded PDF, Word, Excel and other attachments), offline imported data, and external input data (such as industry standards, patent data, academic papers, market intelligence, etc.).

[0026] In a preferred embodiment of this application, the data processing module 101 includes a data access unit and a data governance unit.

[0027] The data access unit is used to collect data from multiple internal and external data sources.

[0028] The data governance unit is connected to the data access unit and is used to perform structuring, vectorization, or graphing processing on the collected data.

[0029] Specifically, for structured data, a unified metadata model is constructed and field mapping relationships are established; for unstructured text, natural language processing (NLP) techniques are used to extract keywords, entities, and summaries, which are then stored in a vector database; for tabular data, the semantics of its rows and columns are parsed to construct lightweight knowledge graph nodes. In this way, multi-source heterogeneous data is transformed into easily retrieved and computed knowledge, enabling cross-source data association and knowledge accumulation, and improving data value density.

[0030] return Figure 1 The task execution engine 102 is connected to the data processing module 101 and is used to receive user requests, decompose user requests into tasks to obtain multiple sub-tasks, extract relevant data from the data processing module 101 according to the type of each sub-task, and schedule the corresponding functional sub-units to execute the corresponding sub-tasks based on the relevant data.

[0031] Here, the task execution engine 102 serves as the platform's capability hub. Its functional sub-units are not bound to specific business scenarios, but only encapsulate basic cognitive and computing capabilities. They are designed for use by dedicated AI employees at the upper level, thereby enabling the rapid development of dedicated intelligent agents and avoiding the problems of non-reusable capabilities and limited application scenarios.

[0032] In a preferred embodiment of this application, the task execution engine 102 includes a request analysis unit, a task orchestration unit, a functional unit, a scheduling and integration unit, and an agent generation unit.

[0033] The request analysis unit is used to identify the type of task requested by the user.

[0034] The task orchestration unit is connected to the request analysis unit and is used to parse user requests into multiple sub-tasks according to the task type.

[0035] The functional unit includes multiple functional sub-units, each used to perform at least one sub-task among data retrieval, statistical analysis, logical reasoning, scheme generation, or simulation prediction.

[0036] Specifically, the functional sub-units can be embodied as follows: a data question-answering sub-unit, which supports natural language questions, automatically parses intent, retrieves data, and generates answers; a data analysis sub-unit, which performs statistical analysis, anomaly detection, and other operations on datasets; a data reasoning sub-unit, which performs logical deduction and hypothesis verification based on rule engines or large models; a data design sub-unit, which assists in generating initial drafts of design schemes based on input parameters and constraints; and a data simulation sub-unit, which inputs parameters into a proxy model to quickly output performance prediction results.

[0037] The scheduling and integration unit is connected to the task orchestration unit and the functional unit, respectively. It is used to schedule the functional sub-units to execute the corresponding sub-tasks and integrate the output results of all functional sub-units into the final result.

[0038] Here, for complex requests that include multiple atomic tasks, the scheduling and integration unit can schedule the corresponding functional sub-units to execute sequentially, and use the output of the previous functional sub-unit as the input of the next functional sub-unit to form a processing pipeline, thereby efficiently completing complex tasks.

[0039] The agent generation unit, connected to the scheduling and integration unit, is used to encapsulate the task processes executed by the scheduling and integration unit into reusable dedicated agents.

[0040] return Figure 1 The security control module 103 is connected to the data processing module 101 and the task execution engine 102, and is used to provide the task execution engine 102 with model invocation strategies to implement a strict security and compliance framework and solve the security problem of sensitive data processing.

[0041] When the task execution engine 102 schedules the corresponding sub-tasks, it selects to call either the public model or the private model to execute the sub-task according to the model calling strategy.

[0042] In a preferred embodiment of this application, the security control module 103 includes a data security classification unit, a model invocation decision unit, and an operation permission management unit.

[0043] The data security classification unit is connected to the data processing module 101 and is used to perform sensitivity marking on the data received by the data processing module 101.

[0044] The model invocation decision unit is connected to the data security classification unit and the task execution engine 102, respectively, and is used to select to invoke a public model or a local private model to execute the subtask based on the sensitivity label of the relevant data involved in each subtask.

[0045] Specifically, the model invocation decision unit selects to invoke either the public model or the local private model in the following ways: If the data involved in the subtask contains highly sensitive tags, the local private model is invoked; if the data involved in the subtask contains low-sensitivity tags or the data is public data, the public model is invoked.

[0046] The operation permission management unit is used to control the user's access and calling permissions to the data processing module 101 according to the user's role permissions, so as to realize fine-grained permission control, and together with data isolation and compliance audit, it forms a complete security and compliance system.

[0047] To more clearly illustrate how this application platform works, a specific scenario of "vehicle disassembly and assembly manual development" is used as an example below. This embodiment demonstrates how the platform, as the underlying support, coordinates various modules to complete a complex R&D task: User operation: The user inputs the specified vehicle model.

[0048] Platform behavior: 1. Data preparation: The data processing module 101 automatically associates multiple sources of data, including the vehicle's 3D CAD model, work time library, SBOM (Bill of Materials), parts catalog, and assembly relationship table. The data governance unit can construct a lightweight knowledge graph to express the topological relationships between parts.

[0049] 2. Task Execution and Scheduling: After receiving a request, the task execution engine 102 decomposes it into multiple sub-tasks, such as virtual verification, structure generation, and content generation, through the request analysis unit and task orchestration unit. The scheduling and integration unit then schedules these sub-tasks sequentially. (1) Data reasoning function subunit, based on 3D spatial relationship, performs virtual disassembly simulation to generate disassembly and assembly structure tree and work schedule.

[0050] (2) Data design function subunit, which calls external CAD interface to generate exploded view and special tool list.

[0051] (3) Data analysis function sub-unit, which compares historical cases to identify high-risk operation points.

[0052] (4) Data question answering function subunit, which converts structured data into natural language step descriptions.

[0053] 2. Security Control: During this process, the model call decision unit of the security control module 103 decides whether to use a private model to execute the sub-tasks locally based on the sensitivity of the processed 3D model, SBOM and other data.

[0054] 3. Results Integration: The scheduling and integration unit automatically integrates and typesets the structure tree, exploded diagram, timetable, tool list, safety prompt text, and other elements output by the above functional sub-units by calling the document template engine, and finally outputs repair manuals and timetable documents that conform to enterprise standards.

[0055] Ultimately, through this platform, the automated and intelligent generation of standardized documents from multi-source data was achieved, significantly improving R&D efficiency and ensuring the security and controllability of core data in the process.

[0056] Compared with existing R&D service platforms, this application, by constructing a unified intelligent foundation consisting of a data processing module, a task execution engine, and a security control module, achieves the following beneficial effects: Avoiding redundant construction and improving reusability: By providing common data interfaces and reusable basic functional sub-units, each department does not need to build underlying capabilities separately, avoiding "reinventing the wheel" and realizing cross-scenario reuse of capabilities.

[0057] Enhance the platform's versatility and scalability: The hierarchical architecture design allows the platform to be independent of specific business domains, flexibly supporting the diverse R&D needs of different departments, and enabling the rapid development of specialized applications through the intelligent agent generation unit.

[0058] Achieving efficient data integration and value enhancement: Employing diversified data governance methods (structured, vectorized, and graph-based) to uniformly process multi-source heterogeneous data breaks down data silos, enables data correlation analysis and knowledge accumulation, and enhances data value density.

[0059] Ensuring end-to-end security and compliance: By integrating data classification, dynamic model routing, and fine-grained access control into a single security control module, the entire process from data access and processing to output is ensured to be carried out within a framework that meets security and compliance requirements, and in particular, it effectively protects sensitive R&D data.

[0060] Please see Figure 2 , Figure 2 A flowchart illustrating the research and development service method provided in the embodiments of this application. For example... Figure 2 As shown in the figure, the R&D service method provided in this application embodiment includes: S101, the data processing module is used to receive data from multiple data sources.

[0061] S102. The task execution engine is connected to the data processing module to receive user requests and decompose the user requests into multiple sub-tasks. Based on the type of each sub-task, relevant data is extracted from the data processing module, and the corresponding functional sub-unit is scheduled to execute the corresponding sub-task based on the relevant data.

[0062] S103, the security control module is connected to the data processing module and the task execution engine to provide a model invocation strategy for the task execution engine. When the task execution engine schedules the corresponding sub-tasks, it selects to call a public model or a private model to execute the sub-task according to the model invocation strategy.

[0063] In a preferred embodiment of this application, the task execution engine includes: The request analysis unit is used to identify the type of task requested by the user.

[0064] The task orchestration unit, connected to the request analysis unit, is used to parse user requests into multiple subtasks based on task type.

[0065] A functional unit includes multiple functional sub-units, each of which is used to perform at least one sub-task among data retrieval, statistical analysis, logical reasoning, scheme generation, or simulation prediction.

[0066] The scheduling and integration unit is connected to the task orchestration unit and the functional unit, respectively. It is used to schedule the functional sub-units to execute the corresponding sub-tasks and integrate the output results of all functional sub-units into the final result.

[0067] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 3 As shown, the electronic device 200 includes a processor 210, a memory 220, and a bus 230.

[0068] The memory 220 stores machine-readable instructions that can be executed by the processor 210. When the electronic device 200 is running, the processor 210 and the memory 220 communicate via the bus 230. When the machine-readable instructions are executed by the processor 210, the steps of the method described above can be performed. For specific implementation details, please refer to the method embodiment, which will not be repeated here.

[0069] This application also provides a computer-readable storage medium storing a computer program. When the computer program is run by a processor, it can execute the steps of the method described above. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0070] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0071] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0072] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0073] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0074] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A research and development service platform, characterized in that, The platform includes: The data processing module is used to receive data from multiple data sources; The task execution engine, connected to the data processing module, is used to receive user requests, decompose the user requests into multiple sub-tasks, extract relevant data from the data processing module according to the type of each sub-task, and schedule the corresponding functional sub-units to execute the corresponding sub-tasks based on the relevant data. A security control module, connected to the data processing module and the task execution engine, is used to provide a model invocation strategy for the task execution engine; When scheduling the functional subunits to execute corresponding subtasks, the task execution engine selects to call a public model or a private model to execute the subtask according to the model invocation strategy.

2. The platform according to claim 1, characterized in that, The task execution engine includes: The request analysis unit is used to identify the task type requested by the user. A task orchestration unit, connected to the request analysis unit, is used to parse the user request into multiple sub-tasks according to the task type; A functional unit includes multiple functional sub-units, each of which is used to perform at least one sub-task among data retrieval, statistical analysis, logical reasoning, scheme generation, or simulation prediction. The scheduling and integration unit is connected to the task orchestration unit and the functional unit respectively, and is used to schedule the functional sub-units to execute corresponding sub-tasks and integrate the output results of all functional sub-units into the final result.

3. The platform according to claim 1, characterized in that, The security control module includes: A data security classification unit, connected to the data processing module, is used to perform sensitivity marking on the data received by the data processing module; The model invocation decision unit is connected to the data security classification unit and the task execution engine respectively, and is used to select to invoke a public model or a local private model to execute the subtask based on the sensitivity label of the relevant data involved in each subtask. The operation permission management unit is used to control the user's access permissions and call permissions to the data processing module according to the user's role permissions.

4. The platform according to claim 3, characterized in that, The model invocation decision unit selects to invoke either a public model or a local private model in the following ways: If the data involved in the subtask contains highly sensitive tags, then the local private model is invoked; If the data involved in the subtask contains low-sensitivity tags or the data is publicly available, then the public model is invoked.

5. The platform according to claim 1, characterized in that, The data processing module includes: The data access unit is used to collect data from multiple internal and external data sources; The data governance unit, connected to the data access unit, is used to perform structuring, vectorization, or graphing processing on the collected data to provide data services for different sub-task types.

6. The platform according to claim 2, characterized in that, The task execution engine also includes: The intelligent agent generation unit, connected to the scheduling and integration unit, is used to encapsulate the task flow executed by the scheduling and integration unit into a reusable dedicated intelligent agent.

7. A research and development service method, characterized in that, The method includes: The data processing module is used to receive data from multiple data sources; The task execution engine is connected to the data processing module and is used to receive user requests, decompose the user requests into multiple sub-tasks, extract relevant data from the data processing module according to the type of each sub-task, and schedule the corresponding functional sub-units to execute the corresponding sub-tasks based on the relevant data. The security control module is connected to the data processing module and the task execution engine, and is used to provide a model invocation strategy for the task execution engine. When scheduling the functional sub-unit to execute the corresponding sub-task, the task execution engine selects to call a public model or a private model to execute the sub-task according to the model invocation strategy.

8. The method according to claim 7, characterized in that, The task execution engine includes: The request analysis unit is used to identify the task type requested by the user. A task orchestration unit, connected to the request analysis unit, is used to parse the user request into multiple sub-tasks according to the task type; A functional unit includes multiple functional sub-units, each of which is used to perform at least one sub-task among data retrieval, statistical analysis, logical reasoning, scheme generation, or simulation prediction. The scheduling and integration unit is connected to the task orchestration unit and the functional unit respectively, and is used to schedule the functional sub-units to execute corresponding sub-tasks and integrate the output results of all functional sub-units into the final result.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the method as described in claim 7 or 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the method as described in claim 7 or 8.

Images