Test data modeling method and system
By constructing experimental data models using SysML graphs, the problems of unsuitable and insufficient expressive methods for enterprise experimental data modeling are solved, providing an efficient data sharing platform and improving the operational efficiency and innovation capabilities of enterprises.
Patent Information
- Application Number
- CN202511214380.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-28
- Publication Date
- 2025-12-02
AI Technical Summary
Existing technologies have problems such as inapplicability of methods, insufficient expressiveness, and poor model reusability in enterprise experimental data modeling, making it difficult to effectively manage and share experimental data.
The experimental data model is constructed using SysML's module definition diagram, activity diagram, requirement diagram, and parameter diagram. The structural, behavioral, requirement, and parameter relationships of the experimental data are clearly expressed through structural, behavioral, requirement, and parameter models, respectively.
It has enabled an efficient and intuitive data sharing platform, which has improved operational efficiency and problem-solving capabilities, prevented the loss of experience, and promoted the inheritance and innovation of enterprise experiments.
Smart Images
Figure CN121052718A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of data modeling technology, and specifically relates to an experimental data modeling method. Background Technology
[0002] With the rapid development of information technology, the digital economy era, with response speed and innovation capabilities as its core metrics, has fully arrived. In this context, data has become significantly more important as a key production factor, and experimental data management has become a crucial means for enterprises to enhance their core competitiveness under information-based conditions. As a fundamental component of data management, data modeling plays an increasingly important role in the organization, sharing, and application of data within enterprises.
[0003] In the process of enterprise test management, building a systematic and structured test data model has the following significant advantages:
[0004] (1) Achieve the orderliness and data sharing of experimental information. Organize scattered and disordered experimental data through digital and structured means, provide employees with a unified data sharing platform, improve work efficiency and innovation capabilities, and optimize experimental processes and result quality. For example, some R&D companies have introduced task-oriented data modeling tools, which can dynamically adapt to changes in experimental plans and improve the reusability and maintainability of experimental data.
[0005] (2) Promote the explicit application of implicit experimental experience and strengthen organizational memory. By using experimental data modeling systems, the experiential data accumulated by front-line researchers or testers can be systematized and documented, and transformed into organizational assets. This reduces the loss of data and experience caused by personnel turnover, and new personnel can also quickly learn and reproduce previous experiments.
[0006] (3) Supporting experimental decision-making and strategic adjustments. By constructing experimental data models of the internal and external environments of enterprises, collecting and integrating information on key technological changes and industry trends, enterprises can dynamically perceive external changes and optimize experimental strategies and resource allocation.
[0007] (4) Promote efficient matching and collaboration between people and data. Establish a data carrier model covering the entire experimental process, link experimental data with expert capabilities, achieve the organic unity of "finding data" and "finding experts", and improve the speed of problem response and problem-solving capabilities.
[0008] (5) Support the accumulation, reuse, and innovation of experimental data. By modeling and preserving the experimental results and technological achievements accumulated by predecessors, the system forms a data foundation for the company's sustainable development and provides a reference for subsequent experimental optimization and technological innovation.
[0009] Current common data modeling methods include various forms such as first-order predicate logic-based, frame-based, rule-based, and ontology-based methods. However, these traditional modeling methods still have many limitations in the application of enterprise experimental data modeling: first-order predicate logic-based methods are not suitable for modeling experimental data with high uncertainty, and are prone to "data explosion" during inference, resulting in low system efficiency when there are a large number of experimental facts. Frame-based methods lack the ability to express procedural experimental data and are difficult to describe complex experimental steps and sequences. Rule-based methods, while suitable for expressing experimental control logic, have insufficient support for structured experimental data and are difficult to reflect the organizational relationships between static elements such as equipment and experimental parameters. Ontology-based methods lack intuitiveness and operability in some scenarios, have complex model structures, poor readability, and are not conducive to understanding and maintenance by front-line experimental personnel. Summary of the Invention
[0010] To address the problems of inapplicability, insufficient expressiveness, and poor model reusability in existing technologies for experimental data modeling, this invention provides an experimental data modeling method, comprising the following steps:
[0011] The structural model is built by constructing a module definition diagram.
[0012] The steps for building a behavior model include constructing a behavior model using activity diagrams.
[0013] The steps for building a requirement model include constructing a requirement model using requirement diagrams.
[0014] The steps for establishing a parametric model involve constructing a parametric model using parametric diagrams.
[0015] Furthermore, the structural model building steps include:
[0016] The enterprise test data is abstracted, and the corresponding test data objects are represented by SysML block elements. The structural and behavioral characteristics of its components are defined. For the relationships between test data objects, the hierarchical structure of the test data system and the interrelationships between data objects in the system are represented by modeling methods such as association, generalization and dependency.
[0017] Furthermore, the steps for building a behavioral model include:
[0018] The behavior of the experimental process based on function and state in the enterprise is abstracted. The behavior content is represented by basic action nodes and object nodes in the activity diagram. The activity process is partitioned by edge connection and the behavior is assigned to block to describe the behavioral characteristics of the experimental data in the dynamic process.
[0019] Furthermore, the steps for establishing a demand model include:
[0020] The hierarchical structure of experimental data requirements, the traceability relationships between requirements, and the design's fulfillment of requirements are modeled. The requirement element in SysML is used to represent specific experimental data requirements. The hierarchical structure of requirements is represented by the containment relationship, and the dependencies between requirements are established by the trace relationship, the inheritance relationship, and the fulfillment relationship. Among them, the trace relationship is used to express the impact path of requirement changes, the inheritance relationship indicates that the child requirement inherits the content of the parent requirement, and the fulfillment relationship indicates that the corresponding module can fulfill the requirement.
[0021] Further steps for establishing the parametric model:
[0022] The constraint relationships between experimental data systems or data objects within a system are abstracted and defined through constraint attributes and constraint parameters in a parametric diagram. Experimental data objects are represented by value attributes, and constraint parameters are bound to value attributes using binding connectors, thereby establishing connections between data objects with constraint logic.
[0023] A SysML-based experimental data modeling system includes:
[0024] The structural model establishes subsystems and constructs the structural model through module definition diagrams.
[0025] The behavior model building subsystem constructs the behavior model through activity diagrams.
[0026] The requirement model building subsystem constructs the requirement model using requirement diagrams.
[0027] The parametric model establishment subsystem constructs the parametric model through parametric diagrams.
[0028] Furthermore, the structural model building subsystem abstracts the enterprise test data, uses SysML block elements to represent the corresponding test data objects, and defines the structural and behavioral characteristics of its components. For the relationships between data objects, the hierarchical structure of the test data system and the relationships between different objects are represented by modeling methods such as association, generalization and dependency.
[0029] Furthermore, the behavior model building subsystem abstracts the experimental process behaviors based on functions or states in the enterprise, uses basic action nodes and object nodes in the activity diagram to represent specific behaviors, divides the behavior flow by connecting nodes with edges, and assigns behaviors to corresponding blocks to characterize the dynamic behavior characteristics of experimental data in the process execution.
[0030] Furthermore, the requirement model establishment subsystem models the hierarchical structure of experimental data requirements, the traceability relationships between requirements, and the design's fulfillment of requirements. It uses the requirement element in SysML to represent specific requirements, expresses the hierarchical structure of requirements through inclusion relationships, and represents various dependencies through trace relationships, inheritance relationships, and fulfillment relationships. Among them, the trace relationship reflects the impact chain of requirement changes, the inheritance relationship is used to express the inheritance and refinement of requirements, and the fulfillment relationship indicates that the corresponding module can meet specific experimental data requirements.
[0031] Furthermore, the parameter modeling subsystem abstracts and models the parameter constraint relationships between the experimental data system or its internal data objects, defining the relationship between constraint attributes and constraint parameters in the parameter graph; experimental data objects are represented by value attributes, and constraint parameters are associated with value attributes using binding connectors, thereby constructing the connection between data objects with parameter constraint relationships.
[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0033] This invention possesses excellent versatility and scalability, making it suitable for manufacturing enterprises or experiment-intensive organizations of different sizes and types. It has broad application prospects in areas such as enterprise data modeling, digital experiment management, and intelligent manufacturing. Through the experiment data model constructed using this invention, enterprises can provide employees with an efficient and intuitive data sharing platform, improving their operational efficiency and problem-solving capabilities. This enables the transformation of experiment data from individual experience into organizational assets, effectively preventing experience loss and promoting the inheritance, innovation, and continuous improvement of enterprise experiments. Furthermore, the modeling process of this invention is simple and has a low application threshold. Enterprise technical personnel only need to master basic SysML graphical data and combine it with modeling tools to quickly establish experiment data models, reducing modeling costs and improving feasibility of implementation. Attached Figure Description
[0034] Figure 1 This is a schematic diagram of the experimental data structure model constructed using a module definition diagram in this invention.
[0035] Figure 2 This is a schematic diagram of the experimental data demand model constructed using a demand graph in this invention.
[0036] Figure 3 This is a schematic diagram of the experimental data parameter model constructed by the parameter diagram in this invention. Detailed Implementation
[0037] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.
[0038] like Figure 1As shown, the construction steps of the experimental data structure model are completed using SysML's Block Definition Diagram (BDD). Specifically, this includes: abstractly modeling the enterprise experimental data; representing experimental data objects using SysML block elements; and defining their structural and behavioral characteristics (such as experimental processes, equipment, operating procedures, and experimental parameters). Through relationships such as association, generalization, dependency, and annotation, the hierarchical structure of the experimental data system and the interactions and connections between modules are clearly expressed, constructing an experimental data structure model with logical and visual characteristics.
[0039] like Figure 2 As shown, the steps for establishing the experimental data requirement model are completed using a SysML requirement diagram. Specifically, this includes: abstractly modeling user requirements, requirement hierarchy, requirement tracing, and the design's fulfillment of these requirements during the experimental process. The `requirement` element represents various experimental requirements, with containment relationships characterizing the requirement hierarchy; trace relationships, derived requirement relationships, and satisfaction relationships represent requirement dependencies.
[0040] (1) Tracking relationship: The arrow points out that when the requirement changes, it will affect the tail requirement.
[0041] (2) Inheritance of requirements: The tail requirement inherits the content of the arrow requirement.
[0042] (3) Satisfying the relationship: The tail end test module can meet this requirement.
[0043] like Figure 3 As shown, the steps for establishing the experimental data parameter model are completed using SysML's parametric diagram, specifically including:
[0044] (1) Abstract model the constraint relationships between the experimental data system or its internal data objects.
[0045] (2) Use constraint property / parameter to define formulas or constraints, and value property to represent specific parameter values;
[0046] (3) By binding the constraint parameters to the value attributes through the binding connector, the constraint relationship between the experimental data objects is formed, ensuring that the dependencies between different modules and parameters within the system are accurately reflected.
[0047] In this invention, the focus of each model is as follows:
[0048] (1) The structural model focuses on describing the hierarchical structure of the experimental data system and the relationships between different data modules;
[0049] (2) Behavioral models mainly characterize experimental behaviors based on functions and states, and describe the dynamics of the experimental process;
[0050] (3) The requirements model emphasizes the hierarchy, traceability, and satisfaction relationships of experimental requirements;
[0051] (4) The parameter model represents the parameter constraints and dependencies between experimental data systems or internal modules.
[0052] Based on the above model construction method, this invention provides an experimental data modeling system based on SysML, including at least one of a structural model building subsystem, a behavioral model building subsystem, a demand model building subsystem, and a parameter model building subsystem.
[0053] (1) Structural model building subsystem
[0054] The structural model is constructed using a block diagram, including:
[0055] The experimental data is abstracted and represented by SysML blocks, defining the structural and behavioral characteristics of its components.
[0056] The hierarchical structure of the experimental system and the interrelationships between different modules are represented by association, generalization, dependency and annotation.
[0057] (2) Behavioral Model Building Subsystem
[0058] Building behavioral models using Activity Diagrams includes:
[0059] Abstract the function-based and state-based behaviors during the experiment;
[0060] The experimental behavior is represented by basic actions and object nodes in the activity graph. The activity is partitioned by connecting nodes, and the behavior is assigned to blocks to represent the dynamic behavior information of the experimental system, ensuring the real-time and dynamic nature of the experimental process.
[0061] (3) Demand Model Establishment Subsystem
[0062] Build a requirement model using a requirement diagram, including:
[0063] The hierarchical relationship of experimental requirements, the traceability relationship between requirements, and the satisfaction status of requirements are abstracted.
[0064] SysML uses requirements to represent corresponding requirements, containment relationships to represent hierarchical relationships between experimental requirements, and trace, derivereqt, and satisfaction relationships to represent dependencies between requirements.
[0065] The tracking relationship indicates that when the requirement at the arrow end changes, it will cause the requirement at the tail end to change. The inheritance relationship indicates that the requirement at the tail end inherits the requirement at the arrow end. The fulfillment relationship indicates that the module at the tail end can fulfill the requirement.
[0066] (4) Parametric Model Establishment Subsystem
[0067] Constructing a parametric model using a parametric diagram includes:
[0068] Abstract the constraint relationships between the test system or its internal modules;
[0069] The constraint relationships between test objects are defined by the constraint attributes and constraint parameters in the parametric diagram, and the test data objects are represented by the value attributes in the parametric diagram.
[0070] By binding connectors, constraint parameters and value attributes are bound together, ensuring that the constraint relationships between experimental data objects are accurately modeled.
[0071] (5) Implementation method
[0072] Those skilled in the art can implement the system and its apparatus, modules, and units provided by this invention using pure software code, or by solidifying the method steps into hardware components (logic gates, application-specific integrated circuits, PLCs, embedded microcontrollers, etc.) through logic programming to realize the system functions. Therefore, the system and its internal modules of this invention can be regarded as hardware structures, as software modules performing the required functions, or as a combination of hardware and software.
[0073] This invention combines the characteristics of enterprise experimental data with the limitations of existing modeling methods, and innovatively introduces SysML, originally applied to the field of system modeling, into experimental data modeling. It fully leverages the expressive capabilities of various modeling graphics in SysML, solves the problem of the lack of effective modeling tools for enterprise experimental data, improves the systematicness and efficiency of experimental data organization and sharing, and better meets the actual needs of experimental management and inheritance.
[0074] The structural model constructed in this invention is based on a module definition graph (BDD), which can provide a clear structured expression framework for experimental data, accurately define the compositional relationships, functional attributes, and associations between experimental data objects and other objects, and provide a stable foundation for subsequent behavior and parameter modeling.
[0075] The behavioral model of this invention is constructed using an activity diagram, which enables functional division and module allocation of operational behaviors during the experiment, realizing the mapping between the operational process and the experimental data objects, improving the clarity and readability of behavioral modeling, and facilitating the optimization and analysis of dynamic processes.
[0076] The requirement model constructed in this invention is based on SysML requirement diagrams, which can systematically describe the hierarchy and dependencies between requirements such as test specifications, quality standards, and user expectations, and clearly reflect the degree to which the model modules meet the relevant requirements, thereby enhancing the consistency between the model and actual business objectives.
[0077] The parametric model of this invention is constructed through a parametric diagram, which can effectively represent the constraints and dependencies between parameters within an experimental data system or subsystem. It supports unified management of key experimental variables through parametric formulas, thereby improving the accuracy and logic of modeling.
[0078] In summary, these are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. All equivalent changes and modifications made in accordance with the scope of the present invention and the contents of the specification are within the scope of the present invention.
Claims
1. A method for modeling experimental data, characterized in that, Includes the following steps: The structural model is built by constructing a module definition diagram. The steps for building a behavior model include constructing a behavior model using activity diagrams. The steps for building a requirement model include constructing a requirement model using requirement diagrams. The steps for establishing a parametric model involve constructing a parametric model using parametric diagrams.
2. The method according to claim 1, characterized in that, The steps for establishing a structural model include: The enterprise test data is abstracted, and the corresponding test data objects are represented by SysML block elements. The structural and behavioral characteristics of its components are defined. For the relationships between test data objects, the hierarchical structure of the test data system and the interrelationships between data objects in the system are represented by modeling methods such as association, generalization and dependency.
3. The method according to claim 1, characterized in that, The steps for building a behavioral model include: The behavior of the experimental process based on function and state in the enterprise is abstracted. The behavior content is represented by basic action nodes and object nodes in the activity diagram. The activity process is partitioned by edge connection and the behavior is assigned to block to describe the behavioral characteristics of the experimental data in the dynamic process.
4. The method according to claim 1, characterized in that, The steps for establishing a demand model include: The hierarchical structure of experimental data requirements, the traceability relationships between requirements, and the design's fulfillment of requirements are modeled. The requirement element in SysML is used to represent specific experimental data requirements. The hierarchical structure of requirements is represented by the containment relationship, and the dependencies between requirements are established by the trace relationship, the inheritance relationship, and the fulfillment relationship. Among them, the trace relationship is used to express the impact path of requirement changes, the inheritance relationship indicates that the child requirement inherits the content of the parent requirement, and the fulfillment relationship indicates that the corresponding module can fulfill the requirement.
5. The method according to claim 1, characterized in that, Steps for establishing a parametric model: The constraint relationships between experimental data systems or data objects within a system are abstracted and defined through constraint attributes and constraint parameters in a parametric diagram. Experimental data objects are represented by value attributes, and constraint parameters are bound to value attributes using binding connectors, thereby establishing connections between data objects with constraint logic.
6. An experimental data modeling system, characterized in that, include: The structural model establishes subsystems and constructs the structural model through module definition diagrams. The behavior model building subsystem constructs the behavior model through activity diagrams. The requirement model building subsystem constructs the requirement model using requirement diagrams. The parametric model establishment subsystem constructs the parametric model through parametric diagrams.
7. The system according to claim 6, characterized in that: The structural modeling subsystem abstracts enterprise experimental data, uses SysML block elements to represent corresponding experimental data objects, and defines the structural and behavioral characteristics of its components. For the relationships between data objects, the hierarchical structure of the experimental data system and the relationships between different objects are represented by modeling methods such as association, generalization and dependency.
8. The system according to claim 6, characterized in that: The behavior model building subsystem abstracts the experimental process behaviors based on functions or states in an enterprise, uses basic action nodes and object nodes in an activity diagram to represent specific behaviors, divides the behavior flow by connecting nodes with edges, and assigns behaviors to corresponding blocks to characterize the dynamic behavioral characteristics of experimental data during process execution.
9. The system according to claim 6, characterized in that: The requirements modeling subsystem models the hierarchical structure of experimental data requirements, the traceability relationships between requirements, and the design's fulfillment of requirements. It uses the requirement element in SysML to represent specific requirements, expresses the hierarchical structure of requirements through inclusion relationships, and represents various dependencies through trace relationships, inheritance relationships, and fulfillment relationships. Among them, trace relationships reflect the impact chain of requirement changes, inheritance relationships are used to express the inheritance and refinement of requirements, and fulfillment relationships indicate that the corresponding module can meet specific experimental data requirements.
10. The system according to claim 6, characterized in that: The parameter modeling subsystem abstractly models the parameter constraint relationships between the experimental data system or its internal data objects, defining the relationship between constraint attributes and constraint parameters in the parameter diagram; experimental data objects are represented by value attributes, and binding connectors are used to associate constraint parameters with value attributes, thereby constructing the connection between data objects with parameter constraint relationships.