28 results about "Requirements model" patented technology

The invention discloses a system demand intelligent analysis and model construction method based on natural language processing, and relates to the technical field of system engineering and model-driven engineering, and the method comprises the following steps: S1, analyzing field metadata; s2, constructing a domain knowledge heterogeneous graph; s3, introducing a theme-guided attention mechanism and a graph-guided dynamic sparse attention mechanism to improve a Longform model; s4, calling an LLM model to generate an entry demand model; s5, the LLM and GNN double-engine analysis is executed; s6, introducing a PPO model as a reinforcement learning agent, and optimizing the semantic mapping rule base; and S7, inputting the design description text into the improved Longform model to generate a behavior model, a structure model and a parameter model. According to the method, the limitations of information extraction generalization, key element omission and weak semantic association of a traditional natural language processing technology in professional field text understanding are overcome, and an efficient and accurate solution is provided for system demand intelligent analysis and model automatic construction.

The application provides a civil aircraft strength evaluation requirement model construction and verification method based on system engineering, which comprises the following steps: S1, establishing a stakeholder requirement database; S2, constructing a top-level requirement model; S3, establishing an application scenario database based on the top-level requirement model, and establishing a corresponding relationship between the top-level requirement and the scenario; S4, constructing a civil aircraft strength evaluation requirement model based on the corresponding relationship between the top-level requirement and the scenario; S5, verifying the requirement traceability chain in the civil aircraft strength evaluation requirement model, and rectifying the requirement traceability chain that does not meet the requirements. The method can establish a stakeholder requirement-top-level requirement-scenario-architecture-verification requirement traceability chain, so that the source of a verification requirement from a stakeholder requirement can be quickly understood, when a stakeholder requirement is changed, the strength evaluation method that needs to be modified related to the stakeholder requirement can be quickly located, and the related model can be modified, and various scenario requirements can be met.

The invention discloses a resource scheduling method and device, terminal equipment and a storage medium, and relates to the technical field of power systems, and the method comprises the steps: converting collected resource data into a resource sensing state value; identifying a feature parameter data set of each task, analyzing a matching relationship between the feature parameter data set and a task demand, screening out a parameter combination matched with the current resource sensing state value from the feature data set according to the matching relationship, and generating a demand modeling parameter set according to the parameter combination; collecting sample data related to the demand modeling parameter set, performing time sequence analysis on the sample data to obtain sample analysis data, comparing difference conditions of task completion rates under different sample analysis data, and adjusting parameters of the modeling parameter set according to the difference conditions to obtain a parameter docking result; and screening out a parameter combination matched with the current sample condition from a parameter docking result, and converting the parameter combination into a resource scheduling strategy. According to the invention, the resource scheduling effect can be effectively improved.

The invention provides a micro-service code automatic generation method, system and equipment based on a multi-level model and a medium, and relates to the technical field of software development, the method comprises the following steps: according to system development stage characteristics and information dimensions, the system is subdivided into a multi-level model description framework comprising a demand model, an architecture model, a design model and a deployment model; and based on a mapping rule between a predefined model and a code, converting the multi-level model into an executable template code based on a Spring micro-service framework through a hierarchical code generation mechanism, so as to realize an automatic multi-level code generation process from demand to a final system. According to the method, the problems that in existing model driving code generation, input is not standard, generated codes are not complete, a complete information system is difficult to generate during large language model code automatic generation, and the codes are possibly inconsistent are solved, and a more efficient, standard and complete code automatic generation scheme is provided.

The application relates to the technical field of software development tools, and discloses a micro application generation system and method based on natural language interaction, which comprises four core components of a natural language understanding module, a requirement modeling module, a code generation engine and a deployment and integration platform; the natural language understanding module generates a structured semantic representation through intent recognition, entity extraction and context analysis; the requirement modeling module constructs a data model, a process logic and an interface prototype based on the semantic representation, and optimizes model parameters by using reinforcement learning; the code generation engine generates cross-platform code of a micro service architecture through template matching and component assembly; and the deployment and integration platform realizes seamless integration of external systems and Internet of Things devices through standardized interfaces. The application can solve the problems of traditional low-code platforms, such as dependence on visual dragging operation, rough requirement analysis and insufficient cross-field adaptation.

The application discloses an intelligent splitting and assembling method of a database model based on a business scene, belongs to the technical field of database management and application, and relates to the technical field of database management and application. The splitting method comprises the following steps: converting an abstract or unstructured business model into a calculable or structured graph representation form, mapping a business function unit into a node in the graph, mapping the relationship between businesses into an edge, and constructing a business topology structure. The assembling method comprises the following steps: acquiring the business topology structure, automatically analyzing a natural language business requirement into a structured requirement model, performing feature extraction on the analyzed requirement, generating a multi-dimensional feature representation, and assembling business model slices in a progressive matching mode. The application can improve the database operation efficiency, reduce the database construction and management cost, and enhance the adaptability and flexibility of the database to business changes.

The invention discloses an intelligent engineering integrated construction and operation and maintenance management method suitable for multiple types of buildings, and relates to the technical field of building intellectualization, and the method is realized through the following core steps: firstly, carrying out building type adaptation and demand modeling, and constructing an intelligent engineering demand model adaptive to a target building type; generating a modular construction scheme based on the demand model and implementing dynamic management and control; after construction is completed, cross-system integrated debugging is carried out, and heterogeneous system data interaction adaptation is achieved; an integrated operation and maintenance management platform is built and intelligent monitoring is carried out; and finally, operation and maintenance decision and full-life-cycle management are realized based on the monitoring data. The method can be adapted to various building types, complete-process collaborative management and control of construction and operation and maintenance are achieved, the construction quality and operation and maintenance efficiency of intelligent engineering are improved, and data support is provided for full-life-cycle optimization of a building intelligent system.

The invention discloses an LLM-based system-level demand modeling and analysis method and platform, and belongs to the field of demand engineering and system modeling. The invention provides an LLM (Logical Language Model)-driven automatic demand behavior model construction and conflict detection method aiming at the problem that accurate modeling and consistency verification of natural language system-level software demands are difficult. According to the method, control flow logic in a demand text is recursively analyzed by simulating an expert stratified analysis process, and a behavior model is automatically generated. The modeling process comprises the steps of recognizing atomic activities, extracting a hierarchical relation and constructing a behavior model. On the basis, a system-level behavior tree is constructed based on preposition and postposition conditions of behavior nodes, a plurality of demand models are comprehensively analyzed, and behavior consistency analysis and conflict detection on the system level are achieved. According to the method, automatic expansion from the demand-level behavior model to the system-level behavior model can be realized, and an effective technical scheme is provided for automatic modeling and analysis of system-level demands.

The invention relates to the technical field of digital twinning, and particularly discloses an MBSE-based production line digital twinning system design method, which comprises the following steps of: constructing a physical operation scene model and a digital operation scene model of a production line; constructing a physical system demand model and a digital operation scene model of the production line; constructing a physical operation activity model and a digital operation activity model of the production line; constructing a physical validity measurement model and a digital validity measurement model of the production line; constructing a physical system module model and a digital system module model of the production line; constructing a physical system function model and a digital system function model of the production line; constructing a physical module component model and a digital module component model of the production line; and constructing a physical system parameter model and a digital system parameter model of the production line.

The invention discloses a software product multi-version parallel development method based on a large language model, and the method comprises the steps: receiving a software multi-version development request, and generating a multi-version demand model through natural language processing and analysis; constructing a training data set containing version difference information based on the multi-version demand model, and performing fine tuning on the large language model in combination with the instruction template to obtain a target large language model adapted to multi-version development; double storage spaces are configured for each software design stage, and initial design achievements and final design achievement storage of all versions are processed in parallel; generating, complementing and optimizing intelligent codes of all versions by utilizing a target large language model; design and code conflicts among multiple versions are detected in real time and intelligently coordinated and solved; and integration verification of all versions is completed through combination of automatic testing and manual testing. According to the method, parallel efficient development of multi-version software is realized, the code generation accuracy is improved, and the development cost caused by demand change and version conflict is reduced.

The invention provides a client heterogeneous federated learning method for a mobile computing environment, and relates to the technical field of federated learning methods, and the method comprises the following steps: screening a client most suitable for participating in federated learning from federated learning candidates through a prediction-based heterogeneous federated learning active configuration strategy; in combination with an interactive knowledge propagation strategy based on channel distillation and decoupling knowledge distillation, knowledge migration is realized through channel distillation and decoupling knowledge distillation technologies; a client-side in-turn chairman strategy is introduced, that is, each client side is allowed to serve as a chairman teacher in turn, and transmission of knowledge in the client sides is promoted. According to the method, on the premise of ensuring data privacy and security, core challenges such as strong dynamic property of client equipment, data isomerism and client personalized requirements (model isomerism) in a mobile computing environment can be effectively handled, so that the model performance in federated learning is remarkably improved, and an excellent comprehensive effect is realized.

The application provides a software test case generation method and system based on domain knowledge constraints. A domain knowledge base covering domain basic knowledge and industry testing criteria is constructed; the original requirement text is normalized, the semantic understanding ability of the knowledge base and the large language model is combined, the requirement type is identified and the structured decomposition rule is dynamically matched, and a structured requirement model is constructed; the test focus features in the requirement are extracted, and the test item decomposition requirements are formulated; the large language model is driven to expand the scene and systematically disassemble the requirement, and the systematic test items are generated according to the test item decomposition requirements; the test cases are generated in multiple dimensions with the test items as the input, and the traceable relationship among the requirements, the test items and the test cases is established; the integrity check and consistency check are performed, and the test design result is output. The application enables the large language model to perform the test design task in a controlled and professional knowledge environment, and realizes the systematic generation from the natural language requirement to the executable test case.

This invention proposes a knowledge mapping system and analysis method for software reliability requirements, belonging to the field of software reliability requirements analysis. This method constructs a standardized requirements modeling system comprising an external cross-linking interface model, a functional logic model, a functional combination model, and a state scenario model to achieve a formal representation of software requirements. It then decomposes and summarizes individual functional requirements, identifies potential failure modes based on a pre-defined classification and comparison system, and establishes a three-layer mapping system of software functional requirements – failure modes – reliability requirements. This completes the forward derivation and transformation of reliability requirements, outputting standardized analysis results. This invention achieves standardized and traceable processing of software reliability requirements analysis, providing prerequisite input and logical basis for the development of highly reliable software, ensuring system stability and task success.

The invention belongs to the technical field of system engineering and software engineering, and provides a function logic confirmation method, system, equipment and medium based on a demand model.The method comprises the steps that firstly, a minimum function module framework organized according to levels is constructed for a system to be confirmed; secondly, processing an original natural language demand to form a demand subset corresponding to each functional unit; thirdly, a single-function logic model is established for each minimum function module and tested, and original requirements are corrected if logic defects are found; and finally, integrating a plurality of confirmed single-function models into a multifunctional module according to a hierarchical relationship and an interface relationship, executing an integration test, and if an interaction logic defect is found, correcting the demand again until the test passes, and completing closed-loop confirmation of the system-level function logic. According to the method, a'modeling-testing-correcting 'closed-loop process is constructed, the dependence of demand confirmation on personnel experience is effectively reduced, and the correctness and integrity of demand logic are remarkably improved.

This application relates to a code generation method, apparatus, device, storage medium, and program product. The method includes: obtaining a requirement model; obtaining an initial class diagram of a design model based on the requirement model; optimizing the initial class diagram of the design model using a multi-objective genetic algorithm and a large model to obtain an optimized design model; and generating monolithic system code matching the requirement model based on the optimized design model. This application can improve code generation efficiency and code generation quality.

The invention belongs to the technical field of software development, and discloses an intelligent auxiliary method and system for software development, and the method comprises the following steps: obtaining a software development demand text, and generating a structured demand model based on a natural language processing model and a software development domain knowledge graph; generating a plurality of sets of candidate technical schemes based on the structured demand model in combination with a technical stack adaptation rule, and performing scoring; on the basis of a pre-training coding model, in combination with context information of a current development scene, real-time generation of codes is achieved; based on the structured demand model and the generated code, automatically generating a test case and generating a test report; data of all links in the whole development process is collected, and continuous parameter optimization is conducted through a reinforcement learning algorithm. By the adoption of the method and system, intelligent assistance of whole-process software development from demand analysis to iterative optimization is achieved, the development efficiency and code quality are improved, the development threshold is lowered, and the technical problems that an existing auxiliary tool is fragmented and poor in adaptability are solved.

The invention relates to the technical field of component assembling, and discloses a component assembling system and method, electronic equipment, a storage medium and a program product, and the component assembling system comprises a component market, a visual assembling designer and an application assembling model; the component market is used for storing a plurality of first target components; the visual assembly designer is used for acquiring demand information input by a user and analyzing the demand information to obtain a demand model; the application assembly model is used for matching the demand model with a plurality of first target components in the component market to obtain a plurality of second target components which are successfully matched, and sending the plurality of second target components to a developer for auditing; and the application assembly model is further used for assembling the multiple second target components passing the audit of the developer and sending the assembly result to the developer for audit, the matched components are analyzed and selected according to the user requirements, component assembly is carried out, and the flexibility of component assembly is improved.

The invention discloses a large-model-assisted iterative user demand completion method and system, and belongs to the field of demand engineering. The method comprises the steps of generating an activity graph based on an initial demand of a user, and presenting a core interaction path by a user journey; guiding the large model to identify a function hook list, extracting a use case for each function hook, constructing a structured use case model, extracting a data entity and generating an E-R graph, and further establishing a'use case-entity-operation 'triple; integrity detection is carried out based on the triple, a report is generated, if missing is found, a complementation suggestion is provided, and the model is iteratively updated after user confirmation; if the function hooks are complete, continuing to process the remaining function hooks; and finally, converting the complete demand model into a structured demand set for output. According to the method, through triple-driven closed-loop detection and iterative completion, the demand integrity and the engineering efficiency are remarkably improved, and the demand engineering is promoted to develop to intelligence.

The application provides an MBSE-oriented requirement model rapid query and visualization method, obtains a query keyword weight vector according to the weight of each query keyword in an MBSE requirement model library, calculates the similarity of the query keyword weight vector and the weight vector of each requirement model lightweight XML file, obtains a requirement model lightweight XML file with the highest similarity, obtains a corresponding requirement model in the MBSE requirement model library according to the obtained requirement model lightweight XML file with the highest similarity, and performs lightweight display on the obtained requirement model, the requirement model file is parsed and a text index is formed, so that the rapid query of the requirement model is realized, the requirement model most conforming to the query intention is obtained by defining a similarity score algorithm, and the precision rate is ensured.

This invention belongs to the field of computer technology, specifically relating to a scheduling and management method for serverless ensemble learning inference services. It includes: S11, acquiring historical request data, base learner models, and system resource status from a cloud application system; S12, performing offline model preference analysis on the historical request data and allocating static weights to the models; S13, classifying online requests and fine-tuning their weights based on static weights, system resource status, and request requirements, scheduling requests based on ensemble inference results; S14, dynamically optimizing the scheduling model; and so on. This invention achieves dynamic scheduling of requests by considering multiple factors such as demand, models, and resources, improving the balance between service accuracy, throughput, and resource utilization. Furthermore, this method, by combining the characteristics of serverless ensemble learning inference services, employs a reinforcement learning-based multi-model instance scaling algorithm to improve service performance and resource consumption.

The invention provides a product full life cycle model construction method and system. The method comprises the following steps: obtaining a meta-meta model from a database; creating a meta-model oriented to the demand model based on the meta-meta-model combination; creating a product structured demand model based on the meta-model combination; searching related instances in a design demand instance library for the product structured demand model based on the meta-model, the attributes of the meta-meta-model and the relation characteristics; judging whether the instance is matched with the demand model or not through domain matching degree calculation, and if yes, incorporating the instance into a demand instance set to be pushed; outputting a demand instance set and a demand list; and if not, constructing a new product design demand instance until the new product design demand instance is matched with the demand model. According to the method, an efficient cross-unit, cross-stage and cross-level cooperation mechanism is met, system-level multi-scale definition of complex products is supported, visualization and traceability of the whole design process are achieved, and the product quality is improved.

The invention relates to a system suitable for field demand modeling, and belongs to the field of software modeling. Comprising a front-end module used for interface display and user interaction; the demand template customization module is used for creating a new demand template, the demand modeling module is used for demand modeling, the checking module is used for checking the correctness of the newly created demand template or model, and the publishing module is used for generating a code of a demand modeling system based on the newly created demand template and automatically deploying the code; the data management module is used for storing the new demand template and data corresponding to the model into a database; and the operation management module is used for configuring the operation environment of the demand modeling system. The system supports a new demand template, the code of the demand modeling system is incrementally generated based on the new demand template through the publishing module, and automatic incremental deployment is carried out, so that the online demand modeling system supports new demand change to realize continuous integration, the iterative development cost brought by the demand change is greatly reduced, and the demand modeling efficiency is improved.

The invention discloses an electromechanical system architecture forward design method based on a servitization thought, and belongs to the field of airplane electromechanical system design. The connection of requirements, functions, services, logic and physical architectures is realized through model-based system engineering (MBSE). Specifically, the method comprises the following steps of: establishing a demand structured model by using a SysML language, analyzing a target system function to obtain a service scheme meeting a function demand, realizing service logic through a logic layer state machine, and performing joint simulation with a physical layer simulation model to form'demand-function-service-logic-physics' full-link mapping taking the model as a clue; and the forward design of the system architecture is realized.

The invention discloses a data storage device for supporting large model training and a use method. The data storage device comprises a system control center, and the system control center is in communication connection with a model construction module, a storage module generation module, a hierarchical architecture construction and strategy generation module and a strategy verification module. According to the method, data types, access frequency and security requirements are collected in a multi-dimensional mode through the model building module, parameters are dynamically adjusted in combination with a training scene, and the single thinking that traditional storage only pays attention to capacity is avoided; the storage module generation module ensures that the hardware performance of the storage unit is adaptive to the training demand through multi-scene performance test and iterative correction on the basis of the demand model; the layered architecture classifies and utilizes storage media with different performances, the performance and the cost are balanced, and a double-layer control strategy realizes dynamic evolution of storage scheduling and adapts to training of full-process load change; the strategy verification module avoids the strategy deployment risk in advance through digital twin simulation and multi-scene optimization, and guarantees the stable operation of the system.

The application discloses a kind of system-level demand modeling and analysis method and platform based on LLM, belong to demand engineering and system modeling field.Aiming at the problem that natural language system-level software demand is difficult to accurately model and consistent verification, an automatic demand behavior model construction and conflict detection method driven by LLM is proposed.The method simulates the hierarchical analysis process of experts, recursively analyzes the control flow logic in the demand text, and automatically generates the behavior model.The modeling process includes identifying atomic activities, extracting hierarchical relationships and constructing behavior models.On this basis, based on the precondition and postcondition of the behavior node, a system-level behavior tree is constructed, and multiple demand models are analyzed comprehensively to realize behavior consistency analysis and conflict detection at the system level.The application can realize the automatic expansion from the demand-level behavior model to the system-level behavior model, and provides an effective technical solution for system-level demand automatic modeling and analysis.

The application discloses a software requirement modeling method based on an artificial intelligence large model, relates to the technical field of requirement modeling, and comprises the following steps: converting a fused semantic element set into nodes and edges, adding an uncertainty mark to an edge with insufficient association basis in the conversion process, constructing a requirement semantic knowledge graph, constructing a requirement model generation scheme based on the requirement semantic knowledge graph, filling the requirement model generation scheme into a preset structured prompt word template, obtaining a structured prompt, driving a multi-modal large model to generate a requirement model draft by using the structured prompt, performing rule compliance checking on the requirement model draft, executing an iterative verification and correction cycle according to a checking result, and obtaining a target requirement model. The application greatly improves the automation efficiency of software requirement modeling while improving the accuracy, completeness and usability of the target requirement model by generating the requirement model draft.

This application discloses a method and system for bidirectional synchronization and verification of requirements and code. The method includes: performing hierarchical parsing of natural language format requirement data using a pre-trained large language model; generating a requirement model based on the results of the hierarchical parsing, wherein the requirement model conforms to DSL syntax; establishing a bidirectional mapping relationship between the semantic elements of the requirement model and the nodes of the AST of the code; based on the bidirectional mapping relationship, when the requirement model changes, marking or updating the corresponding nodes according to the changed semantic elements, or when the code changes, marking or updating the corresponding semantic elements according to the changed nodes; performing a semantic consistency comparison between the requirement logic of the current requirement model and the actual logic of the current abstract syntax tree, and generating a conflict report when the core semantics of the requirement logic and the actual logic are inconsistent. This method can achieve bidirectional synchronization and consistency verification of requirements and code, improving development efficiency and development quality.

The invention provides an overall configuration method and application of a space-based rotary launching platform. The model-based system engineering method comprises the following steps of: establishing a demand model, a function model and an architecture model of the platform; a parameterized modeling step: based on the architecture model, constructing a digital model of each subsystem of the platform by using parameterized design software; a dynamic simulation analysis step: carrying out dynamic simulation on the constructed digital model, and analyzing the influence of the load launching process on the platform and the motion of the load; and a scheme iterative optimization step: optimizing the platform design scheme based on the dynamic simulation analysis result. According to the method, disturbance suppression breakthrough innovation is achieved, the launching efficiency is greatly improved, and the orbit precision is effectively guaranteed.