A system and method for designing an architecture of an integrated algorithm model
Through the architectural design system of integrated algorithm models, the problem of lack of outstanding intelligent characteristics in intelligent equipment and intelligent activity design is solved, efficient design and verification of the intelligent system is realized, and design quality and efficiency are improved.
Patent Information
- Application Number
- CN202410939104.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-15
AI Technical Summary
In the existing design of intelligent equipment and intelligent activity systems, the intelligent characteristics are not outstanding, the intelligent effect is single, and the intelligent application is weak, resulting in low quality and efficiency of intelligent system design.
A system design system integrating algorithm models is proposed, including intelligent element design module, architecture modeling module, intelligent auxiliary design module and model verification module. By configuring intelligent algorithms and basic activities and basic components, the equipment model is constructed and carried out. Static and dynamic verification to realize the comprehensive design and verification of intelligent systems.
It improves the quality and efficiency of intelligent system design, highlights the characteristics of intelligence, supports the design of intelligent element systems, realizes the rapid transformation and application of intelligent technologies, and improves the security and comprehensive effect of design.
Smart Images

Figure CN118466943B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of architecture design, and in particular to an architecture design system and method for integrating an algorithm model. Background Art
[0002] Intelligent equipment and intelligent activities have the characteristics of different intelligent system attributes and elements, intelligent interaction, etc., and have broad application prospects in future battlefields. The design of intelligent systems involving intelligent equipment requires a comprehensive description and detailed characterization to promote the implementation of concepts and demand response, and drive the construction of specific systems and equipment.
[0003] In the existing system design process, individual elements with integrated intelligent features such as intelligent equipment and intelligent activities rely on different algorithm models, but there are problems such as lack of prominent intelligent features, single intelligent effects, and weak intelligent applications. Summary of the invention
[0004] In order to solve the above technical problems, the present invention proposes an architecture design system and method of an integrated algorithm model to solve the existing technical problems of single intelligent effect and weak intelligent application.
[0005] The present invention proposes an architecture design system integrating an algorithm model, the system comprising:
[0006] Intelligent element design module, architecture modeling module, intelligent auxiliary design module and model verification module;
[0007] The intelligent element design module determines the basic activities and basic components corresponding to the equipment model based on the capability requirements and architecture requirements of the equipment model, configures intelligent algorithms for the basic activities and basic components, configures interactive interfaces for the basic activities and basic components to interact with other equipment models, and forms configured basic activities and basic components based on user input; the basic activities are the smallest units of behavior that can be executed by nodes of the equipment model to change the state of itself or related entities, and the basic components are the smallest components that constitute the equipment model; the equipment model is an abstract model of equipment, and equipment is an instantiation of the equipment model;
[0008] The intelligent auxiliary design module stores auxiliary design data for providing the architecture modeling module with relevant data of the auxiliary design function;
[0009] The architecture modeling module obtains configured basic activities and configured basic elements from the intelligent element design module based on user input, and constructs the equipment model in combination with the auxiliary design data provided by the intelligent auxiliary design module;
[0010] The model verification module performs static verification and dynamic verification on the equipment model. The static verification includes a structure consistency check and a legality check, and the dynamic verification includes a connectivity check.
[0011] Preferably, the intelligent element design module includes an activity autonomy degree design submodule, an intelligent technology list design submodule, an intelligent technology and activity mapping design submodule, an autonomous collaborative relationship design submodule, and a collaborative rule design submodule;
[0012] The activity autonomy degree design submodule is used to manage various attributes of basic activities and basic elements; based on user input, corresponding attributes are set for each basic activity and each basic element related to the user input;
[0013] The intelligent technology list design submodule is used to display the basic activity list in the form of a list, and the basic activity list includes basic activities, categories to which basic activities belong, applicable equipment for basic activities, several intelligent algorithms supporting basic activities, and technical effects;
[0014] The intelligent technology and activity mapping design submodule is used to display the association mapping between intelligent algorithms and basic activities in the form of a matrix. By checking the elements in the matrix, the selected intelligent algorithm is configured to the basic activity;
[0015] The autonomous collaborative relationship design submodule is used to set collaborative relationships for each basic activity and each basic element related to the user input based on the user input;
[0016] The collaboration rule design submodule is used to set collaboration rules for each basic activity and each basic element related to the user input based on the user input.
[0017] Preferably, the intelligent auxiliary design module includes a data relationship graph analysis submodule, an integrated intelligent algorithm display submodule, and a model view generation submodule;
[0018] The data relationship graph analysis submodule is used to determine the object to be analyzed and the analysis perspective according to the user's input, generate a relationship graph for the object to be analyzed, and display the relationship graph based on the analysis perspective;
[0019] The integrated intelligent algorithm display submodule is used to display the dynamic timing diagram and dynamic state diagram of each intelligent algorithm, and to display the dislocation detection of the dynamic timing diagram and dynamic state diagram of each intelligent algorithm and the content output in real time during the compilation process;
[0020] The model view generation submodule is used to display the global architecture overview model of the constructed equipment model.
[0021] Preferably, the architecture modeling module includes a service perspective building submodule, a system perspective building submodule, a capability perspective building submodule, an information perspective building submodule and a standard specification perspective building submodule;
[0022] The service perspective construction submodule extracts tasks based on user input, determines the process and constraints of task execution, decomposes tasks, and determines the equipment model to be designed; based on the mapping relationship with the system perspective construction submodule, the capability perspective construction submodule, the information perspective construction submodule, and the standard specification perspective construction submodule, the completed basic activities and configured basic components are obtained from the intelligent element design module, and the equipment model is constructed in combination with the auxiliary design data provided by the intelligent auxiliary design module;
[0023] The system perspective construction submodule extracts tasks based on user input, determines the systems required to perform the tasks, determines the system architecture and interfaces with other systems, and maps the system architecture and interfaces with other systems to the equipment model;
[0024] The capability perspective construction submodule determines the capabilities required to complete the task, decomposes the required capabilities, and maps the required capabilities to the equipment model;
[0025] The information perspective construction submodule determines the information flow mode to complete the task and maps it with each basic activity;
[0026] The standard specification perspective construction submodule determines the standard specifications required to complete the task and maps the required standard specifications with the equipment model.
[0027] Preferably, the model verification module includes a model static verification module and a model dynamic verification module;
[0028] The model static verification module includes an architecture consistency check submodule, an architecture integrity check submodule, an architecture legality check submodule and an architecture connectivity check submodule;
[0029] The model dynamic module performs detection and verification and connectivity verification on the dynamic graph generated in the modeling process of the architecture modeling module based on the auxiliary design data provided by the intelligent auxiliary design module;
[0030] The model static verification module includes an architecture consistency check submodule, an architecture integrity check submodule and an architecture legality check submodule;
[0031] The architecture consistency check submodule performs a consistency check on the equipment model based on a consistency check rule;
[0032] The architecture integrity check submodule performs a functional and structural integrity check on the equipment model;
[0033] The architecture legality check submodule performs a legality check on the equipment model based on the full legality check rules.
[0034] The present invention provides a method for designing an architecture of an integrated algorithm model. The method is based on the above-mentioned architecture design system, and the method comprises:
[0035] Step S1: The user inputs task information and constraint conditions, and the architecture design system of the integrated algorithm model determines the equipment model to be designed;
[0036] Step S2: The architecture design system of the integrated algorithm model has a visual interface, which displays basic components, basic activities, and intelligent algorithms. The basic components and basic activities are configured by dragging the basic components and basic activities, setting properties on the basic components and basic activities, and superimposing intelligent algorithms. The architecture design system of the integrated algorithm model generates an equipment model for the configured basic components and basic activities, verifies the equipment model, and uses the verified equipment model as output.
[0037] The present invention provides an electronic device, comprising:
[0038] at least one processor; and
[0039] a memory communicatively connected to the at least one processor; wherein,
[0040] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.
[0041] The present invention provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the method as described above.
[0042] The present application proposes an architecture design system and method for an integrated algorithm model. The present invention focuses on mission tasks and is oriented towards the design requirements of intelligent systems. By providing the functions of the integrated algorithm model to support the acquisition of the cultivation algorithm list, the binding of single algorithms and the verification of integrated algorithms, the system meets the needs of intelligent element design and intelligent body design at the single level, promotes the rapid transformation and application of intelligent technology, connects the algorithm model output by the intelligent body training and cultivation system, provides auxiliary means of support, and improves the quality and efficiency of intelligent system design.
[0043] The present invention adopts an architecture design system that integrates intelligent algorithm models to provide system intelligent design capabilities. On the basis of being able to assist in the design of various types of perspective models such as tree diagrams, flow charts, tables, etc. in a human-computer interactive manner, it realizes the function of highlighting the intelligent design of individual units, connects to the algorithm library of the training and cultivation system, that is, it can obtain the list of intelligent algorithms of the training and cultivation system, and provide functions such as real-time acquisition, update, and viewing of intelligent algorithms; it can bind and view intelligent algorithms for individual design elements; it can use intelligent algorithms to perform logical verification of activity processes to meet the design requirements of intelligent systems.
[0044] The system of the present invention has the functions of connecting with the intelligent agent training and cultivation system and using intelligent algorithm verification to assist in supporting optimized design, thereby achieving comprehensive improvement of design quality, efficiency and safety while effectively supporting the implementation of intelligent system design tasks.
[0045] The present invention is connected with the training and cultivation system to obtain a list of intelligent algorithms in the cultivation system, and combines it with the system design model to achieve the highlighting of intelligent characteristics in the system design process and support the design of the intelligent element system.
[0046] This invention focuses on the characteristics of the monomer intelligent integration algorithm and proposes a system architecture design method for the integrated algorithm model through the requirements of the intelligent element monomer design level. At the same time, relying on the system architecture optimization design system, the proposed system architecture design method is functionally implemented for its intelligent auxiliary design module, solving the problem of the associated design connection with the intelligent algorithm integration in the monomer intelligent design, realizing the model dynamic verification technology of the integrated algorithm model, supporting the monomer level element design in the intelligent system design, highlighting the intelligent characteristics, assisting the design process, and improving the design efficiency and design quality.
[0047] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which:
[0049] Figure 1 A schematic diagram of the system structure design for the integrated algorithm model proposed in the present invention;
[0050] Figure 2 A schematic diagram of the integration of the intelligent algorithm of the present invention into the activity process;
[0051] Figure 3 A schematic diagram of a class implementation for intelligent verification of the present invention;
[0052] Figure 4 Interaction diagram for intelligent verification of the present invention. DETAILED DESCRIPTION
[0053] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0054] Figure 1 The schematic diagram of the architecture design system structure of the integrated algorithm model is shown. Figure 1 As shown, the system comprises:
[0055] Intelligent element design module, architecture modeling module, intelligent auxiliary design module and model verification module;
[0056] The intelligent element design module determines the basic activities and basic components corresponding to the equipment model based on the capability requirements and architecture requirements of the equipment model, configures intelligent algorithms for the basic activities and basic components, configures interactive interfaces for the basic activities and basic components to interact with other equipment models, and forms configured basic activities and basic components based on user input; the basic activities are the smallest units of behavior that the equipment model can execute to change the state of itself or related entities, and the basic components are the smallest components that constitute the equipment model; the equipment model is an abstract model of equipment, and equipment is an instantiation of the equipment model;
[0057] The intelligent auxiliary design module stores auxiliary design data for providing the architecture modeling module with relevant data of the auxiliary design function;
[0058] The architecture modeling module obtains configured basic activities and configured basic elements from the intelligent element design module based on user input, and constructs the equipment model in combination with the auxiliary design data provided by the intelligent auxiliary design module;
[0059] The model verification module performs static verification and dynamic verification on the equipment model. The static verification includes a structure consistency check and a legality check, and the dynamic verification includes a connectivity check.
[0060] In the present invention, the related data is, for example, data relationship analysis data, design style data, etc.
[0061] Furthermore, the intelligent element design module includes an activity autonomy degree design submodule, an intelligent technology list design submodule, an intelligent technology and activity mapping design submodule, an autonomous collaborative relationship design submodule, and a collaborative rule design submodule.
[0062] The activity autonomy degree design submodule is used to manage various attributes of basic activities and basic components; based on user input, corresponding attributes are set for each basic activity and each basic component related to the user input.
[0063] The intelligent technology list design submodule is used to display the basic activity list in the form of a list, and the basic activity list includes basic activities, categories to which basic activities belong, applicable equipment for basic activities, several intelligent algorithms supporting basic activities, and technical effects.
[0064] The intelligent technology and activity mapping design submodule is used to display the association mapping between intelligent algorithms and basic activities in the form of a matrix. By checking the elements in the matrix, the selected intelligent algorithm is configured to the basic activity.
[0065] The autonomous collaborative relationship design submodule is used to set collaborative relationships for each basic activity and each basic element related to the user input based on the user input.
[0066] The collaboration rule design submodule is used to set collaboration rules for each basic activity and each basic element related to the user input based on the user input.
[0067] In the present invention, the activity autonomy degree design submodule designs the attributes of the activity process modeling, designs the autonomy degree of the activity elements such as remote control, autonomy, semi-autonomy, etc., and realizes the characterization of the autonomy degree. By adding attribute values, the autonomy degree attribute is bound and associated with the activity process at the metamodel level, and the activity autonomy degree is designed in the attribute view during the modeling process.
[0068] The intelligent technology list design submodule uses intelligent technology classes to model different intelligent technology entities and describe their different attribute contents. The intelligent technology data content to be modeled is obtained by acquiring the intelligent algorithm interface, and the intelligent technologies involved in the intelligent algorithm list in the external system are mapped to different attribute contents to construct the corresponding intelligent technology modeling class; the intelligent technology list model class is used to obtain, edit, and display all intelligent technology operations in the form of a list of constructed intelligent technologies.
[0069] The intelligent technology and activity mapping design submodule builds a mapping binding relationship between the intelligent algorithm and the basic activities at the data layer to implement operations such as editing and displaying the mapping relationship. The mapping relationship attribute class is used to store the corresponding relationship, and the intelligent technology and activity mapping model class is used to store, display and organize the model.
[0070] The autonomous collaborative relationship design submodule designs the collaborative relationships and collaborative rules involved in the system. By entering the basic information of the model in the new organizational structure dialog box, a new organizational structure model is created, and the collaborative relationship between organizational structures is constructed by dragging and dropping lines in the toolbar, and the new coordination relationship is created in the data layer and the coordination relationship is bound to the previous and next organizational structure element data with coordination relationship objects.
[0071] The collaborative rule design submodule is used to model the coordination rule related content and data, and supports the construction of collaborative design rule models. The design of collaborative rules is to use the collaborative design rule model class to create new collaborative rule data and a system architecture model presented in a table form, supporting the editing of new rules and attributes.
[0072] Furthermore, the intelligent auxiliary design module includes a data relationship graph analysis submodule, an integrated intelligent algorithm display submodule, and a model view generation submodule.
[0073] The data relationship graph analysis submodule is used to determine the object to be analyzed and the analysis perspective according to the user's input, generate a relationship graph for the object to be analyzed, and display the relationship graph based on the analysis perspective.
[0074] The integrated intelligent algorithm display submodule is used to display the dynamic timing diagram and dynamic state diagram of each intelligent algorithm, and to display the dislocation detection of the dynamic timing diagram and dynamic state diagram of each intelligent algorithm and the content output in real time during the compilation process.
[0075] The model view generation submodule is used to display the global architecture overview model of the constructed equipment model.
[0076] In the present invention, the data relationship graph analysis submodule determines the analysis object by selecting the perspective, the element type in the perspective, and the specific element name, supports the dynamic generation of the relationship graph of the analysis object, and displays all related elements and related relationships of the analysis object.
[0077] The integrated intelligent algorithm display submodule is used for model routine misalignment detection of dynamic timing diagrams and dynamic state diagrams based on intelligent algorithms and dynamic model verification functions of real-time output during the compilation process. The integrated knowledge graph content is used to detect and verify the dynamic timing diagrams and state diagrams in the modeling process, and the compilation process is output in real time to realize the dynamic verification function.
[0078] The model view generation submodule is used to automatically generate global architecture overview models, data dictionaries and other functions based on the extraction, analysis and utilization of existing data relationships. At the same time, this submodule can provide semi-automatic generation functions of various types of models.
[0079] Furthermore, the architecture modeling module includes a service perspective building submodule, a system perspective building submodule, a capability perspective building submodule, an information perspective building submodule and a standard specification perspective building submodule.
[0080] The service perspective construction submodule extracts tasks based on user input, determines the process and constraints of task execution, and decomposes tasks to determine the equipment model to be designed; based on the mapping relationship with the system perspective construction submodule, the capability perspective construction submodule, the information perspective construction submodule and the standard specification perspective construction submodule, the completed basic activities and configured basic components are obtained from the intelligent element design module, and the equipment model is constructed in combination with the auxiliary design data provided by the intelligent auxiliary design module.
[0081] The system perspective construction submodule extracts tasks based on user input, determines the systems required to perform the tasks, determines the system architecture and interfaces with other systems, and maps the system architecture and interfaces with other systems to the equipment model.
[0082] The capability perspective construction submodule determines the capabilities required to complete the task, decomposes the required capabilities, and maps the required capabilities to the equipment model.
[0083] The information perspective construction submodule determines the information flow mode to complete the task and maps it with each basic activity.
[0084] The standard specification perspective construction submodule determines the standard specifications required to complete the task and maps the required standard specifications with the equipment model.
[0085] In the present invention, the service perspective construction submodule extracts tasks, obtains the process and constraints describing the task execution, as well as the organization and grouping of task completion and the collaborative relationship, mainly including conceptual model construction, task decomposition model, state diagram, and timing diagram.
[0086] The system perspective construction submodule is used to describe the system composition structure of the architecture and the operating relationship between systems, including the construction of system composition model, system interaction model, system function decomposition model, system and activity mapping relationship model, etc.
[0087] The capability perspective construction submodule is used to describe the completion capability requirements of the system architecture, including the construction of capability decomposition model, capability and organization mapping relationship, effect attribute model, etc. The role of the capability perspective is to design the overall goal vision of the system into quantifiable effect requirements.
[0088] The information perspective construction submodule is used to describe the information activities and activity processes of the system, and to outline the operation structure of the system with the information activity process as the main body, including the construction of information activity process model, business activity and information activity mapping model, etc.
[0089] The standard specification perspective construction submodule is used to design the standard specifications such as the technical system, interaction, data format, etc. followed by the system architecture. It focuses on the construction of the technical standard model, and models, digitizes, and visualizes the technical standard classification and standard specifications suitable for the current architecture design.
[0090] Furthermore, the model verification module includes a model static verification module and a model dynamic verification module.
[0091] The model static verification module includes an architecture consistency check submodule, an architecture integrity check submodule, an architecture legality check submodule and an architecture connectivity check submodule.
[0092] The model dynamic module performs detection and verification and connectivity verification on the dynamic graph generated during the modeling process of the architecture modeling module based on the auxiliary design data provided by the intelligent auxiliary design module.
[0093] The model static verification module includes an architecture consistency check submodule, an architecture integrity check submodule and an architecture legality check submodule.
[0094] The architecture consistency check submodule performs a consistency check on the equipment model based on a consistency check rule.
[0095] The architecture integrity check submodule performs a functional and structural integrity check on the equipment model.
[0096] The architecture legality check submodule performs a legality check on the equipment model based on the full legality check rules.
[0097] In the present invention, the model dynamic module has the function of regular misalignment detection of the model based on the dynamic timing diagram and dynamic state diagram of the intelligent algorithm and the dynamic model verification function of real-time output during the compilation process. The integrated knowledge graph content is used to detect and verify the dynamic timing diagram and state diagram in the modeling process, and the compilation process is output in real time to realize the dynamic verification function. Model checking is performed on the possible problems of multi-type models such as process-type and structural-type in the system architecture design process. Metamodel technology is used to formulate model connectivity check rules to realize rule-based model static verification, thereby supporting the connectivity check of the system architecture.
[0098] The architecture consistency check submodule is used to perform model checks on possible problems in multiple types of models such as process-type and structural-type models during the architecture design process. It adopts metamodel technology and formulates model consistency check rules to achieve rule-based model static verification, thereby supporting architecture consistency checks.
[0099] The architecture integrity check submodule is used to perform model checks on possible problems in multiple types of models such as process-type and structural-type models during the architecture design process. It adopts metamodel technology and formulates model integrity check rules to achieve rule-based model static verification, thereby supporting architecture integrity checks.
[0100] The architecture validity check submodule is used to perform model checks on possible problems in multiple types of models such as process-type and structural-type models during the architecture design process. It adopts metamodel technology and formulates model validity check rules to achieve rule-based model static verification, thereby supporting architecture validity checks.
[0101] In the present invention, the capability requirements and architecture requirements are determined based on tasks. The tasks are determined by a concept-assisted design subsystem of an architecture design system that is independent of the integrated algorithm model, and the capability requirements and architecture requirements are determined based on tasks by a capability requirement-assisted generation subsystem of an architecture design system that is independent of the integrated algorithm model. The generated equipment model is sent to a simulation deduction test system of an architecture design system that is independent of the integrated algorithm model for use.
[0102] The architecture design of the integrated algorithm model of the present invention is driven by the concept formed by the concept-assisted design subsystem, guided by the capability requirements and system requirements formed by the capability requirements to assist in generating the subsystem, and carries out the architecture model design work. On the basis of static and dynamic verification and analysis of the system design model, it realizes iterative optimization of the architecture design.
[0103] In the present invention, the method of configuring the algorithm for the basic activity is as follows Figure 2Select an activity, right-click to open the function menu, click Integrated Intelligent Algorithm, select Bind Intelligent Algorithm, View Bind Algorithm, or Unbind Intelligent Algorithm according to the function, and after binding the intelligent algorithm, right-click to open the function menu to execute the model verification function. The underlying implementation of the verification is as follows Figure 3 shown.
[0104] In the present invention, the description of the intelligent verification class is shown in Table 1.
[0105] Table 1 Description of intelligent verification classes
[0106]
[0107] The interactive diagram of the model static verification module for intelligent verification is as follows: Figure 4 As shown. First, the intelligent algorithm is integrated; second, the intelligent algorithm is verified and the verification result is obtained. Then, the business logic is verified based on the algorithm and the business logic verification result is obtained.
[0108] In the present invention, select the activity element, right-click to open the function menu, select "Integrate Intelligent Algorithm", select "Bind Intelligent Algorithm", "View Bound Algorithm", or "Unbind Intelligent Algorithm" as needed, and the algorithm selection window, binding algorithm window, and unbinding algorithm confirmation window will pop up respectively, open the acquired intelligent algorithm list, check the bound intelligent algorithm, and display the single intelligent design node. Each single intelligent design node is a basic activity or basic object.
[0109] Dynamic model verification integrates intelligent algorithms into the architecture model to implement business logic verification based on algorithm-based dynamic sequence diagrams and dynamic state diagrams. Sequence diagrams and state diagrams are used to describe business process behaviors. During the operation of the business process, check whether the interactions in the sequence diagram meet the requirements and specifications, and verify the correctness of the sequence diagram; check whether the state transitions in the state diagram meet the requirements and specifications, and verify the correctness of the state diagram.
[0110] Traditional inspection and analysis methods for various design models mainly include manual review methods, rule methods, constraint transformation methods and formal methods. The technical implementation characteristics and advantages and disadvantages of these model inspection methods are shown in the following table.
[0111] Table 2 Main model verification implementation methods
[0112]
[0113] The intelligent auxiliary design module is based on the visualization model construction realized by the architecture modeling module, and generates the timing diagram and state diagram corresponding to the activity process and the node resource flow transmission process; the structure verification module obtains the timing diagram and / or state diagram to be verified as the diagram to be verified, and generates the graph structure data G=(V, E) corresponding to the diagram to be verified based on the diagram to be verified, wherein V is a node in the diagram to be verified, and E is a link relationship or transmission relationship in the diagram to be verified, and the connection line is an edge; V is used as a node of the graph structure data, and E is used as an edge of the graph structure data. Encode the graph structure data, verify the generated code based on the GNN model, and output the coding compilation process result of the verification process.
[0114] The present invention takes into account the structural characteristics of timing diagrams and state diagrams, and integrates the graph neural network (GNN) intelligent algorithm in the architecture model to detect abnormal situations where the attributes or structures in the diagram do not conform to the normal pattern, thereby realizing the detection of common errors in timing diagram and state diagram models.
[0115] GNN is a deep learning model for processing graph-structured data, which can extract important information from graphs and make useful predictions. GNN has been widely studied and has been successfully applied in complex machine learning tasks such as node classification, link prediction, and graph classification, thanks to its high expressiveness achieved through message passing in effectively learning graph representations. GNN-based graph anomaly detection methods use information about graph attributes or structures to learn and predict anomalies. Specifically, it is necessary to first convert the timing graph and state graph into a graph structure, and then verify the correctness of the model by learning and predicting the graph structure.
[0116] First, the elements, attributes, and relationships in the model are converted into graph structure data.
[0117] A graph consists of a set of nodes and the edges between them, defined as G=(V, E), where V is a set of nodes and E is the edges between nodes. During the conversion, nodes are used to represent element entities, and edges are used to represent the relationships between element entities. The edges between node sets can represent the structural information of the graph, which can be stored using an adjacency matrix.
[0118] Second, the graph structure is encoded.
[0119] Graphs have different structures and sizes, so they usually do not conform to the matrix form expected by neural networks. Graphs also have other characteristics that make them different from the information types designed by classical neural networks. For example, as long as the relationship between the nodes remains unchanged, changing the order and position of the nodes will not have an impact. The information of graph data contains three levels: node information, edge information, and overall graph information. Graph data must be converted into a format that can be processed by neural networks, generally represented as a series of matrices. Techniques such as node embedding and edge embedding are used to encode the graph structure. Node embedding is to map each node to a low-dimensional vector space, and edge embedding is to map each edge to a vector space. On this basis, GNN can use the information of nodes and edges to encode and learn the graph structure.
[0120] Again, GNN is used to learn and predict the graph structure.
[0121] GNN performs an optimized transformation on all attributes on the graph. After obtaining the optimized attribute vector, it connects the fully connected neural network to perform classification and regression tasks at the node, edge or graph level. It uses intelligent algorithm models such as graph convolutional neural network GCN and graph attention network GAT to represent the graph as a low-dimensional, real-valued, dense vector form, integrate the attribute information of the graph for learning and prediction, and verify the correctness of the model, such as detecting deadlocks, conflicts and other problems.
[0122] Finally, the model validation log and result information are output.
[0123] According to the learning prediction results of the algorithm, the model verification log information is output in real time, and the prediction results are converted into a model verification result list for output and display.
[0124] The static verification verifies the constructed digitized model based on preset rules, which are integrity and consistency rules set or input by the system. When performing dynamic verification, the automatically generated state diagram and timing diagram are obtained, the current compiled code and model data attribute status are obtained, and the logical relationship content of the human in the loop is verified for the current model structure based on the compilation logic and timing status combined with intelligent algorithms.
[0125] In the present invention, the concept of Human in the Loop (HITL) mainly comes from the field of artificial intelligence and is a way of artificial intelligence learning. With the support of big data and computing power, the fields of artificial intelligence and machine learning continue to advance, and services and automation systems dominated by artificial intelligence are gradually becoming an indispensable part of daily life. At present, machine learning performs well in supervised learning tasks, but the trained models are prone to devastating disasters on new data sets. The growing demand and contradictions for human-computer interaction have led researchers to explore solutions for human-in-the-loop, hoping to rely on human intelligence to help machines become more intelligent.
[0126] Human-in-the-loop emphasizes the introduction of the "human" factor into the loop, making people part of the loop. Human-in-the-loop means integrating personal judgment into the AI system process, forming a closed loop of process operation, and supporting repeated operation with relatively fixed steps and states. In essence, human-in-the-loop redefines the problem of automation as a human-computer interaction design problem, with the goal of achieving interactive machine learning that can enhance or improve the ability of humans to complete tasks. By involving people as an important part of the loop and participating in the closed-loop control of the system, the operation of the system is monitored and adjusted, thereby effectively improving the correctness and reliability of the model. Human-in-the-loop allows users to change the results of events or processes, intelligently track changes over time and intermediate results, and achieve rapid iteration and rapid response feedback.
[0127] In order to introduce people into the loop of the business process, a list of trigger events is automatically generated based on the architecture view, and manual event triggering is used to achieve dynamic interaction between people and business processes through ECA (Event-Condition-Action) rules and event generators. Then, timing diagrams and state diagrams are dynamically generated based on the triggered events, and their logic is verified based on intelligent algorithms.
[0128] ECA rules are a method that combines event-triggered rules with an object-oriented, event-driven environment. They consist of three parts: events, conditions, and actions. The form of ECA rules is:
[0129] WHEN Events
[0130] IF Conditions THEN
[0131] Action
[0132] ENDIF
[0133] ENDWHEN
[0134] Event: A specific behavior or action that can produce a change in state or condition during execution. Events are atomic and either occur or not occur at a certain moment. Events are divided into simple events and compound events. A simple event is an event that occurs at a certain moment, and a compound event is composed of several simple events through event algebra.
[0135] The main properties of an event include:
[0136] name
[0137] describe
[0138] Condition: used to determine whether the rule's activity can be executed. In general, the conditional expression can be any logical expression in the rule prediction, and its return value is TRUE or FALSE. If the condition calculation result is TRUE, then the action is executed; if the condition returns FALSE, then the action is not executed, and the next condition in the rule body of the same event is determined to be satisfied.
[0139] Action: When a rule is triggered and the conditions are met, the action will be executed. When an event is manually triggered, a set of rules related to the event are triggered and the related actions are executed.
[0140] The model verification module can generate statistical results based on the verification results. It provides a display view to perform statistical analysis on the model's integrity, consistency association, legality, connectivity, human-in-the-loop verification and other verification results, and displays them in a visual way, so that users can quickly understand the overall quality and problem distribution of the network information system architecture design model, and make it easier for users to formulate targeted strategies to improve the quality and efficiency of the architecture design model in future work.
[0141] In the present invention, statistical analysis of the verification results can be achieved through a variety of methods such as bar charts, pie charts, tables, etc., which is convenient for users to intuitively see relevant statistical information and supports exporting statistical display result charts as pictures.
[0142] In addition to supporting statistics, analysis and visual display of verification results, the intelligent assisted design module will also provide a report generation function, which supports the output of verification and analysis results into a Word format report in the specified format, making it convenient for users to summarize, edit and submit model verification results.
[0143] The present invention provides a method for designing an architecture of an integrated algorithm model. The method is based on the aforementioned system for designing an architecture of an integrated algorithm model. The method comprises:
[0144] Step S1: The user inputs task information and constraint conditions, and the architecture design system of the integrated algorithm model determines the equipment model to be designed;
[0145] Step S2: The architecture design system of the integrated algorithm model has a visual interface, which displays basic components, basic activities, and intelligent algorithms. The basic components and basic activities are configured by dragging the basic components and basic activities, setting properties on the basic components and basic activities, and superimposing intelligent algorithms. The architecture design system of the integrated algorithm model generates an equipment model for the configured basic components and basic activities, verifies the equipment model, and uses the verified equipment model as output.
[0146] The above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).
[0147] The above modules can be connected or communicated with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, etc., or any combination thereof. The wireless connection may include a connection in the form of a LAN, a WAN, Bluetooth, a ZigBee, or NFC, or any combination thereof. Two or more modules may be combined into a single module, and any one module may be divided into two or more units. Those skilled in the art will clearly understand that for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application.
[0148] It should be noted that the above modules can be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more microprocessors (DSP), or one or more field programmable gate arrays (FPGA). For example, when a module is implemented in the form of a processing element scheduling program code, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).
[0149] The present application also provides a program product, such as a computer-readable storage medium, including a program, which is used to execute the above method embodiment when executed by a processor.
[0150] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0151] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0152] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.
[0153] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (English: processor) to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (English: Read-Only Memory, referred to as: ROM), random access memory (English: Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.
Claims
1. An architecture design system integrating an algorithm model, characterized in that: The system comprises: Intelligent element design module, architecture modeling module, intelligent auxiliary design module and model verification module; The intelligent element design module determines the basic activities and basic components corresponding to the equipment model based on the capability requirements and architecture requirements of the equipment model, configures intelligent algorithms for the basic activities and basic components, configures interactive interfaces for the basic activities and basic components to interact with other equipment models, and forms configured basic activities and basic components based on user input; the basic activities are the smallest behavioral units that can change the state of the equipment model itself or related entities, and the basic components are the smallest components that constitute the equipment model; the equipment model is an abstract model of the equipment, and the equipment is an instantiation of the equipment model; The intelligent auxiliary design module stores auxiliary design data for providing the architecture modeling module with relevant data of the auxiliary design function; The architecture modeling module obtains configured basic activities and configured basic elements from the intelligent element design module based on user input, and constructs the equipment model in combination with the auxiliary design data provided by the intelligent auxiliary design module; The model verification module performs static verification and dynamic verification on the equipment model, wherein the static verification includes a structural consistency check and a legality check, and the dynamic verification includes a connectivity check; The model verification module includes a model static verification module and a model dynamic verification module; The model static verification module includes an architecture consistency check submodule, an architecture integrity check submodule, an architecture legality check submodule and an architecture connectivity check submodule; The model dynamic verification module performs detection and verification and connectivity verification on the dynamic graph generated in the modeling process of the architecture modeling module based on the auxiliary design data provided by the intelligent auxiliary design module; the model dynamic verification module has a dynamic model verification function that is output in real time during the compilation process of the dynamic timing diagram and dynamic state diagram based on the intelligent algorithm; the dynamic timing diagram and state diagram in the modeling process are detected and verified by using the integrated knowledge graph content, and the compilation process is output in real time to realize the dynamic verification function; The model static verification module includes an architecture consistency check submodule, an architecture integrity check submodule and an architecture legality check submodule; the model static verification module performs model checks on the process type and structure type in the system architecture design process, adopts metamodel technology, and implements rule-based model static verification by formulating model connectivity check rules; The architecture consistency check submodule performs a consistency check on the equipment model based on a consistency check rule; The architecture integrity check submodule performs a functional and structural integrity check on the equipment model; The architecture legality check submodule performs a legality check on the equipment model based on the full legality check rules; The intelligent element design module includes an activity autonomy degree design submodule, an intelligent technology list design submodule, an intelligent technology and activity mapping design submodule, an autonomous collaborative relationship design submodule, and a collaborative rule design submodule; The activity autonomy degree design submodule is used to manage various attributes of basic activities and basic elements; based on user input, corresponding attributes are set for each basic activity and each basic element related to the user input; The intelligent technology list design submodule is used to display the basic activity list in the form of a list, and the basic activity list includes basic activities, categories to which basic activities belong, applicable equipment for basic activities, several intelligent algorithms supporting basic activities, and technical effects; The intelligent technology and activity mapping design submodule is used to display the association mapping between intelligent algorithms and basic activities in the form of a matrix. By checking the elements in the matrix, the selected intelligent algorithm is configured to the basic activity; The autonomous collaborative relationship design submodule is used to set collaborative relationships for each basic activity and each basic element related to the user input based on the user input; The collaboration rule design submodule is used to set collaboration rules for each basic activity and each basic element related to the user input based on the user input.
2. The system according to claim 1, characterized in that The intelligent auxiliary design module includes a data relationship graph analysis submodule, an integrated intelligent algorithm display submodule, and a model view generation submodule; The data relationship graph analysis submodule is used to determine the object to be analyzed and the analysis perspective according to the user's input, generate a relationship graph for the object to be analyzed, and display the relationship graph based on the analysis perspective; The integrated intelligent algorithm display submodule is used to display the dynamic timing diagram and dynamic state diagram of each intelligent algorithm, and to display the dislocation detection of the dynamic timing diagram and dynamic state diagram of each intelligent algorithm and the content output in real time during the compilation process; The model view generation submodule is used to display the global architecture overview model of the constructed equipment model.
3. The system according to claim 1, characterized in that The architecture modeling module includes a service perspective construction submodule, a system perspective construction submodule, a capability perspective construction submodule, an information perspective construction submodule and a standard specification perspective construction submodule; The service perspective construction submodule extracts tasks based on user input, determines the process and constraints of task execution, decomposes tasks, and determines the equipment model to be designed; based on the mapping relationship with the system perspective construction submodule, the capability perspective construction submodule, the information perspective construction submodule, and the standard specification perspective construction submodule, the completed basic activities and configured basic components are obtained from the intelligent element design module, and the equipment model is constructed in combination with the auxiliary design data provided by the intelligent auxiliary design module; The system perspective construction submodule extracts tasks based on user input, determines the systems required to perform the tasks, determines the system architecture and interfaces with other systems, and maps the system architecture and interfaces with other systems to the equipment model; The capability perspective construction submodule determines the capabilities required to complete the task, decomposes the required capabilities, and maps the required capabilities to the equipment model; The information perspective construction submodule determines the information flow mode to complete the task and maps it with each basic activity; The standard specification perspective construction submodule determines the standard specifications required to complete the task and maps the required standard specifications with the equipment model.
4. A method for designing an architecture of an integrated algorithm model, the method being based on the system for designing an architecture of an integrated algorithm model as claimed in any one of claims 1 to 3, the method comprising: Step S1: The user inputs task information and constraint conditions, and the architecture design system of the integrated algorithm model determines the equipment model to be designed; Step S2: The architecture design system of the integrated algorithm model has a visual interface, which displays basic components, basic activities, and intelligent algorithms. The basic components and basic activities are configured by dragging the basic components and basic activities, setting properties on the basic components and basic activities, and superimposing intelligent algorithms. The architecture design system of the integrated algorithm model generates an equipment model for the configured basic components and basic activities, verifies the equipment model, and uses the verified equipment model as output.
5. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of claim 4.
6. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to make the computer execute the method according to claim 4.
Citation Information
Patent Citations
Intelligent human platform establishing method for equipment design and evaluation
CN103699735A
Digital twin industrial software platform makeTwin
CN115826937A
Model-based system research and development process method and management system
CN116796498A
Equipment predictive maintenance decision system and method based on industrial configuration
CN117993577A
Equipment maintenance support resource configuration optimization system
CN118114560A
