Object-driven digital model orthogonal analysis and dual-adaptation fusion method and system

Abstract

The invention relates to a target-driven digital model orthogonal analysis and dual-adaptation fusion method and system, and the method comprises the steps: obtaining an overall demand, and combining the overall demand with empirical data and a knowledge graph to generate a model demand; obtaining a model composition file and a model description file which is determined based on a model demand and is used for dividing model categories, generating a meta-model description file according to the model composition file and the model description file, and generating a meta-model based on the meta-model description file; analyzing input and output data logic of the meta-model, scheduling parameters according to meta-model interface specifications, and redefining an interface by adopting a nonlinear iteration and aggregation optimization method; correlation adaptation and proxy packaging are carried out on a result of transmission logic numerical value fitting, and numerical value verification is completed in combination with a precision constraint condition; and adapting the incidence relation of the meta-agent model, generating a business-level numeralization agent model, scheduling parameters to perform precision verification on the business-level numeralization agent model, and generating a unified and operable system model.

Description

Technical Field

[0001] This application relates to the field of digital modeling technology, and in particular to a target-driven method and system for orthogonal analysis and dual-adaptation fusion of digital models. Background Technology

[0002] Throughout the entire equipment lifecycle, digital models serve as a crucial supporting role, acting as the digital carrier of core information related to equipment design, manufacturing, and delivery. Their development is primarily based on overall unit units, supplemented by subsystem units, with the core objective of supporting equipment research, production, testing, and evaluation, achieving seamless integration of the "design-manufacturing-delivery" process. The content of equipment digital models varies with the application stage: the research and development stage includes models of design requirements, functions, performance, and physics; the production stage focuses on models of process design, manufacturing processes, and quality; and the application stage centers on support-related model elements such as delivery status, support requirements, and usage methods. Equipment digital models possess multi-disciplinary integration characteristics, with complex and diverse modeling tools and formats. They typically consist of constituent models and characteristic models: the constituent models describe the internal and external structure and compositional logic (including overall layout), while the characteristic models describe core attributes such as delivery status (including functional performance); their representation forms include various formats such as 3D structural diagrams and logical views.

[0003] However, current equipment model calculation and simulation evaluation tools have obvious technical bottlenecks and are difficult to meet the needs of systematic application. Most tools are geared towards specific evaluation objects, with rigid structures and single functions. They cannot adapt to systematic and complex calculations, nor do they have a cross-domain model collaborative application environment. This makes it difficult to efficiently analyze and collaborate equipment digital models of multiple disciplines, stages, and formats, which seriously restricts the realization of their full life cycle value.

[0004] Therefore, there is an urgent need to develop a goal-driven digital model orthogonal analysis and dual-adaptation fusion method and system to solve one or more of the above-mentioned problems. Summary of the Invention

[0005] In view of this, in order to solve the above-mentioned technical problems or some of the technical problems, the embodiments of the present invention provide a target-driven digital model orthogonal analysis and dual-adaptation fusion method and system.

[0006] In a first aspect, this application provides a target-driven method for orthogonal analysis and dual-adaptation fusion of digital models, the method comprising: Obtain overall requirements, and combine these overall requirements with empirical data and knowledge graphs to generate model requirements to guide model loading and metamodel decomposition; Obtain the model composition file and the model description file for classifying model categories based on the model requirements; generate a meta-model description file based on the model composition file and the model description file; and generate a meta-model based on the meta-model description file. The input and output data logic of the meta-model is analyzed. Based on the scheduling parameters of the meta-model interface specification, the interface is redefined using nonlinear iteration and aggregation optimization methods, and the numerical fitting of the transmission logic is realized simultaneously. The results of fitting the numerical values ​​of the transmitted logic are associated, adapted, and encapsulated by a proxy, and numerical verification is completed in combination with precision constraints. Adapt the association relationships of the meta-proxy model to generate a business-level numerical proxy model, and use scheduling parameters to verify the accuracy of the business-level numerical proxy model to generate a unified and runnable system model.

[0007] In one possible implementation, the step of obtaining the overall requirements and combining the overall requirements with empirical data and knowledge graphs to generate model requirements includes: Based on the indicator description documents of equipment development and design and the test description documents of test evaluation, the overall requirements for model evaluation and calculation analysis are obtained. By pre-constructing equipment systems and digital model knowledge graphs, the mission-level indicators corresponding to the overall requirements are analyzed and decomposed to form detailed indicator requirement description documents that match the test subjects; Based on the detailed indicator requirement description document, matching the indicator evaluation requirements of professional fields, and combining empirical data and test subject requirements, a multi-level evaluation indicator system description document is constructed. Using the aforementioned multi-level evaluation index system as a constraint and guide, the test resource requirements are matched from the professional dimensions involved in the test subjects to form a meta-model professional description document; Based on the meta-model professional description file, model clustering analysis is performed to classify categories from the perspectives of assessment tasks and application business, generate meta-model category description files, and integrate the meta-model category description files to generate model requirements.

[0008] In one possible implementation, the step of generating a metamodel description file based on the model composition file and the model description file, and generating a metamodel based on the metamodel description file, includes: Based on the model composition file and model description file, the dispatched equipment system model is decomposed and logically analyzed to determine the professions, fields and forms of expression involved in the meta-model, and to form the element analysis results. Analyze the key core parameters and feature factors of each type of model, and generate meta-model construction description files; Based on the tool types and formats of the metamodel, the encapsulation protocols, data connection and transmission specifications of input and output interfaces are parsed, interface matching is completed and metamodel interface description files are generated. The metamodel construction description file and the metamodel interface description file are integrated to generate a metamodel description file, and a metamodel is generated based on the metamodel description file.

[0009] In one possible implementation, the analysis of the input-output data logic of the meta-model, according to the meta-model interface specification scheduling parameters, redefining the interface using nonlinear iteration and aggregation optimization methods, and simultaneously achieving numerical fitting of the transmission logic, includes: The underlying physical mechanism logic, intermediate data representation logic, and top-level business application logic of the meta-model are analyzed in their entirety to generate a standardized data logic graph. Based on the meta-model interface specification and the data logic graph, the associated parameters at each level are matched and scheduled to build a logic parameter interface linkage mechanism. A combination of nonlinear iterative algorithm and aggregation optimization method is adopted. With input and output data as the driving force, the meta-model transitive logic numerical fitting model is constructed by alternating between prior input response and random input response, and the transitive logic numerical model description file is generated simultaneously. Based on the aforementioned transmission logic numerical model description file, the underlying logic interface, intermediate representation interface, and top-level application interface of the meta-model are uniformly and standardizedly redefined to clarify the interface data format, transmission protocol, and interaction rules.

[0010] In one possible implementation, the step of associating, adapting, and encapsulating the result of fitting the transmitted logic numerical values, and performing numerical verification in conjunction with precision constraints, includes: Load the numerical fitting results of the transmission logic, the full element information of the meta-model and the description file of the accuracy constraints, and sort out the full elements of the meta-model from the horizontal model dimension to obtain the full element analysis results of the meta-model. Based on the full-element analysis results of the meta-model, the numerical fitting results of the transmission logic are precisely matched with the meta-model operation logic and interface logic to form an association mapping relationship between the fitting results and the meta-model logic. Based on the aforementioned correlation mapping relationship, the fitting results are reduced in order to obtain the initial meta-proxy model; Based on the accuracy constraint description file, the initial meta-surrogate model is driven to perform numerical calculations, the output data is obtained and compared with the accuracy constraint standard to obtain the error results; The model parameters are dynamically adjusted based on the error results through a feedback control network until the initial meta-surrogate model meets the accuracy constraint requirements. Based on the validated meta-proxy model, a standardized meta-proxy model description file is generated.

[0011] In one possible implementation, the process of adapting the association relationships of the meta-proxy model to generate a business-level numerical proxy model, and using scheduling parameters to verify the accuracy of the business-level numerical proxy model to generate a unified and runnable system model includes: Obtain standardized meta-proxy model description files, instances of each meta-proxy model, and the business flow and information flow specifications of the original digital model; Integrate the data dimension and model dimension proxy model, fuse and match based on interface description protocol, obtain the input, output and time sequence logic between models, and generate a logic adaptation description file between models. Based on the adaptation description file, the meta-proxy models are associated and assembled to generate a business-level numerical proxy model; The system schedules preset verification parameters and datasets to verify the data, computation, timing, and event logic of the proxy model and obtains the verification results. Based on the verification results, optimize the model parameters or solidify the model form to generate a unified, runnable system model.

[0012] Secondly, this application provides a target-driven digital model orthogonal analytic and dual-adaptation fusion system, the system comprising: The evaluation target decomposition and adaptation plugin is used to obtain the overall requirements and combine the overall requirements with empirical data and knowledge graphs to generate model requirements, so as to guide model loading and metamodel decomposition. The model decomposition and logic parsing engine is used to obtain the model composition file and the model description file for classifying the model categories based on the model requirements, generate a meta-model description file based on the model composition file and the model description file, and generate a meta-model based on the meta-model description file. The logic fitting and interface adaptation engine is used to analyze the input and output data logic of the meta-model, schedule parameters according to the meta-model interface specification, redefine the interface using nonlinear iteration and aggregation optimization methods, and simultaneously realize the numerical fitting of the transmitted logic. The proxy model encapsulation and verification engine is used to perform correlation adaptation and proxy encapsulation on the results of the numerical fitting of the transmitted logic, and to complete numerical verification in combination with precision constraints. A business-level model reconstruction plugin is used to adapt the association relationship of the meta-proxy model, generate a business-level numerical proxy model, and perform accuracy verification of the business-level numerical proxy model using scheduling parameters to generate a unified and runnable system model.

[0013] In one possible implementation, the evaluation target decomposition and adaptation plugin includes: The evaluation target acquisition module is used to obtain the overall requirements for model evaluation and calculation analysis based on the indicator description documents of equipment development and design and the test description documents of test evaluation. The target autonomous analysis module is used to analyze and decompose the mission-level indicators corresponding to the overall requirements by pre-constructing equipment system and digital model knowledge graph, and form a detailed indicator requirement description document that matches the test subjects; The indicator system construction module is used to construct a multi-level evaluation indicator system description file based on the detailed indicator requirement description file, matching the indicator evaluation requirements of the professional field, and combining empirical data and test subject requirements. The model professional adaptation module is used to match the test resource requirements from the professional dimensions involved in the test subjects, based on the constraints of the multi-level evaluation index system, and form a meta-model professional description file. The model category segmentation module is used to perform model clustering analysis based on the meta-model professional description file, segment categories from the evaluation task and application business level, generate meta-model category description files, and integrate the meta-model category description files to generate model requirements.

[0014] In one possible implementation, the model decomposition and logic parsing engine includes: The original model element analysis module is used to perform element decomposition and logical analysis on the dispatched equipment system model based on the model composition file and model description file, determine the profession, field and expression form of the meta-model, and form the element analysis result. The metamodel interface parsing module is used to parse the key core parameters and feature factors of various types of models and generate metamodel construction description files. The metamodel tool parsing module is used to parse the encapsulation protocol, data connection and transmission specifications of input and output interfaces based on the tool type and format of the metamodel, complete interface matching and generate metamodel interface description files; The metamodel module generation module is used to integrate the metamodel construction description file and the metamodel interface description file to generate a metamodel description file, and generate a metamodel based on the metamodel description file.

[0015] In one possible implementation, the logic fitting and interface adaptation engine includes: The meta-model parameter scheduling module is used to perform full-link analysis of the underlying physical mechanism logic, intermediate data representation logic and top-level business application logic of the meta-model, generate a standardized data logic graph, and match and schedule the associated parameters at each level according to the meta-model interface specification and data logic graph to build a logical parameter interface linkage mechanism. The transit logic numerical fitting module is used to construct a meta-model transit logic numerical fitting model by combining nonlinear iterative algorithms and aggregation optimization methods, guided by input and output data, and by iteratively alternating between prior input response and random input response, and simultaneously generating a transit logic numerical model description file. The metamodel interface redefinition module is used to redefine the underlying logic interface, intermediate representation interface and top-level application interface of the metamodel in a unified and standardized manner based on the transmitted logic numerical model description file, so as to clarify the interface data format, transmission protocol and interaction rules.

[0016] Compared with the prior art, the technical solutions provided in this application have the following advantages: The method provided in this application obtains independent meta-models by decomposing the model structure and building tools, and achieves model dimension and data dimension adaptation and modular organization analysis through horizontal and vertical orthogonal analysis, thereby improving analysis efficiency; it breaks through the single meta-model division method, fits the actual equipment, drives design from top to bottom based on the overall goal, and realizes index decomposition, intelligent matching and meta-model modular division by using experience data and knowledge graphs; it breaks through the proxy model implementation method based on relationship fitting, and uses vertical multi-level data logic to drive and integrate multi-level data logic from bottom to top to form an efficient numerical reduced-order proxy model construction framework and technical system; and it provides a general adaptation tool, realizes meta-model parsing adaptation through standardized open interfaces, and completes model numerical and functional logic construction and model system integration by relying on the hybrid data model analysis engine. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0020] Figure 1 A flowchart illustrating a target-driven orthogonal analysis and dual-adaptation fusion method for digital models provided in this application embodiment; Figure 2 This application provides a working link diagram of a target-driven digital model orthogonal analysis and dual-adaptation fusion system. Figure 3 This application provides a schematic diagram of the structure of a target-driven digital model orthogonal analytic and dual-adaptive fusion system. Figure 4 This is a flowchart illustrating the workflow of a target-driven digital model orthogonal analysis and dual-adaptation fusion system provided in an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The following disclosure provides numerous different embodiments or examples for implementing various structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the scope of the invention. Furthermore, reference numerals and / or letters may be repeated in different examples. Such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed.

[0023] Equipment digital models are digital descriptions of equipment design schemes, manufacturing processes, and delivery status during the overall design, manufacturing, and delivery process. The design, development, and research of equipment digital models are primarily undertaken by the overall development unit, supplemented by subsystem development units, supporting equipment development, production, testing, and evaluation, realizing the "design-manufacturing-delivery" process of the digital model. The content of the equipment digital model differs at different stages. The development stage mainly includes design requirement models, functional models, performance models, and physical models; the production stage mainly includes process design models, manufacturing process models, and quality models; and the application stage mainly includes model elements such as equipment delivery status, support requirements, operation methods, operating procedures, and maintenance characteristics.

[0024] Equipment digital models are characterized by their multi-disciplinary nature, involving a complex variety of modeling tools and formats. Generally, they consist of a constituent model and a characteristic model. The constituent model describes the internal and external structure and compositional logic, including the overall layout, system components, and deliverable materials. The characteristic model describes the equipment's delivery status, physical characteristics, internal mechanisms, and other information, including functional performance and target characteristics. The modeling tools and representations vary across different models, generally including 3D structural diagrams, logical views, structured text, specialized analysis algorithms, simulation training software, interactive electronic manuals, source code, dynamic link libraries, and data acquisition programs.

[0025] Most current model computation and simulation evaluation tools are geared towards specific evaluation objects, with limited structure and functionality, making it difficult to meet the needs of system operation and complex calculations, and lacking an environment that supports cross-domain models. Providing an integrated model parsing and adaptation service in equipment model computation and simulation evaluation tools has become a necessary means for the systematic application of models. The integrated model parsing and adaptation service, based on a parsing and adaptation engine with an autonomous scheduling network, enables efficient parsing and system reconstruction of complex models in both the horizontal component dimension and the vertical representation dimension, thereby forming a generalized computational-level model.

[0026] To address the technical problem that existing digital models lack adaptability in sparse and anomalous data scenarios, making it difficult to accurately reflect the true state and evolution of equipment systems during stable operation, and leading to discrepancies between model evaluation results and actual business needs, this application provides a target-driven digital model orthogonal analysis and dual-adaptation fusion method and system. Through the synergistic effect of the above modules, it can effectively improve the accuracy, adaptability, and engineering practicality of digital models in equipment system evaluation and computational analysis, providing strong technical support for the entire life cycle of equipment development, testing, and evaluation.

[0027] Figure 1 This is a flowchart illustrating a target-driven orthogonal analytic and dual-adaptation fusion method for digital models provided in an embodiment of this application, as shown below. Figure 1 As shown, the method includes: S101. Obtain the overall requirements and combine the overall requirements with empirical data and knowledge graphs to generate model requirements to guide model loading and metamodel decomposition.

[0028] In this embodiment, the requirements for equipment digital model evaluation and computational analysis are dynamically acquired. Through key steps such as evaluation target acquisition, target autonomous analysis, indicator system construction, model professional adaptation, and model category classification, mission-level indicators are intelligently analyzed and decomposed to form a complete multi-level evaluation indicator system. Meta-model categories are divided from the evaluation task and application business levels, and meta-model category description files are generated, providing equipment model selection requirements and model category classification definitions.

[0029] S102. Obtain the model composition file and the model description file for classifying model categories based on the model requirements; generate a meta-model description file based on the model composition file and the model description file; and generate a meta-model based on the meta-model description file.

[0030] In this embodiment, the equipment system model is scheduled, element decomposed, and logically parsed according to the metamodel category description file. Through key steps such as metamodel element parsing, metamodel tool parsing, metamodel interface parsing, and metamodel module generation, a general-purpose adaptation tool is used to be compatible with modeling tools and multiple formats of various types of models such as 3D structure, mathematical logic, and behavioral simulation. A structured metamodel module is generated, and the model is orthogonally decomposed. Interface requirements are given from the data dimension and composition requirements are given from the model dimension.

[0031] S103. Analyze the input and output data logic of the meta-model, schedule parameters according to the meta-model interface specification, redefine the interface using nonlinear iteration and aggregation optimization methods, and simultaneously achieve numerical fitting of the transmission logic.

[0032] In this embodiment, the logic of each metamodel module is analyzed from the vertical data dimension, including the underlying logic, intermediate representation and top-level application of the model. The multi-dimensional integrated adaptation method of "mechanism interpretation + numerical fitting + application verification" is adopted. Through key steps such as metamodel parameter scheduling, transfer logic numerical fitting, and metamodel interface redefinition, the numerical logic fitting model is constructed for the metamodel module with input and output as the driving force, realizing the data integration of the metamodel, generating the metamodel interface numerical logic description file, and giving the metamodel interface logic.

[0033] S104. The result of the numerical fitting of the transmitted logic is associated, adapted, and proxied, and numerical verification is completed in combination with the accuracy constraint conditions.

[0034] In this embodiment, a full meta-model element analysis is performed on each meta-model module from the horizontal model dimension, including meta-model specialization, meta-model type, meta-model tools, etc. Through key steps such as meta-proxy model encapsulation and runtime accuracy constraint verification, the original form, format and independent specific interface of the model are stripped away, the meta-model operation logic and interface logic are matched, and the meta-model is reduced in order to obtain a meta-proxy model that has completed encapsulation and numerical accuracy verification.

[0035] S105. Adapt the association relationship of the meta-proxy model to generate a business-level numerical proxy model. Use scheduling parameters to verify the accuracy of the business-level numerical proxy model and generate a unified and runnable system model.

[0036] In this embodiment, the complete system model is combined and reconstructed. Through key steps such as logical adaptation between models, business model generation, and model accuracy verification, the logical association and assembly of the meta-proxy model are realized, the data dimension and model dimension are integrated, and the business-level proxy model is verified at the underlying level through the "loading / scheduling / running integrated engine" to perform data logic, operation logic, timing logic, and event logic, generate model composition files, and finally form a unified and runnable digital model that meets business constraints.

[0037] The target-driven digital model orthogonal analysis and dual-adaptation fusion method provided in this application starts from the overall needs of equipment development in a target-driven manner. It relies on knowledge graphs and experience data to realize the intelligent decomposition of mission-level indicators and the construction of a multi-level evaluation indicator system. Then, it accurately matches professional domain model resources, completes the modular division and category definition of meta-models, and ensures that the model decomposition process closely fits the actual business scenarios and evaluation task requirements of the equipment.

[0038] For example, a given multi-disciplinary digital model of equipment is analyzed, parsed, fitted, and reconstructed from both the model and data levels to obtain a workable model in a unified format, specifically including: Step 1: Evaluation Target Decomposition and Adaptation Plugin. The overall requirements for model evaluation and computational analysis are obtained through the structured file interface. Based on empirical data and knowledge graph, intelligent decomposition of targets and rapid matching of equipment models are achieved, generating domain model requirements and guiding model loading and meta-model decomposition. Step 2: Model decomposition and logic parsing engine, loads and schedules model composition files, obtains description files for model category classification, and parses, decomposes and forms metamodel description files from three levels: elements, tools and interfaces, according to model composition, format and application requirements, and generates metamodel modules. Step 3: Logic Fitting and Interface Adaptation Engine. This is a vertical data dimension analysis engine. It obtains the meta-model description file, analyzes the input and output data logic of the meta-model, schedules parameters according to the meta-model interface specification, and uses nonlinear iteration and aggregation optimization methods to achieve numerical fitting of the transmission logic and redefine the interface. Step 4: Proxy model encapsulation and verification engine, which is a horizontal model dimension analysis engine. It obtains the metamodel description file and the metamodel interface logic description file. Based on the application capabilities of the metamodel module, it performs correlation adaptation and proxy encapsulation on the numerical fitting results of the metamodel transmission logic, and performs numerical verification based on the metamodel accuracy requirement constraints. Step 5: The business-level model reconstruction plugin obtains the encapsulated meta-proxy model, adopts a normalized and standardized model output format, adapts the relationship between meta-proxy models, generates a business-level, numerical proxy model, realizes the fusion of meta-proxy models, schedules parameter data for accuracy verification, and generates a unified and runnable system model.

[0039] At the model analysis level, through the analysis of original model elements, interfaces, and tools, a deep deconstruction of the equipment system model is achieved across disciplines, tools, and formats. This accurately extracts the core parameters, logical interfaces, and tool features of the meta-model, laying the foundation for subsequent model fusion. In the logic fitting and interface adaptation stage, an innovative strategy combining nonlinear iterative algorithms and aggregation optimization methods is adopted. Using input and output data as a guide, a numerical fitting model for the meta-model's logic transmission is constructed. Furthermore, the underlying logic interfaces, intermediate representation interfaces, and top-level application interfaces of the meta-model are uniformly and standardizedly redefined, effectively solving the problems of heterogeneous interfaces and logical conflicts in multi-source meta-models, and achieving smooth connection of model logic and efficient data flow.

[0040] Building upon this foundation, by encapsulating the meta-proxy model and verifying its operational accuracy constraints, redundant information in the model is stripped away, achieving a reduction in the meta-model's order and efficient numerical proxying. Then, by adapting the relationships within the meta-proxy model, a business-level numerical proxy model is generated and rigorously verified for accuracy. Finally, these are integrated to form a unified and operational system model. This approach overcomes the limitations of traditional model parsing and adaptation, which relies on single-dimensional, static associations. Through horizontal and vertical orthogonal parsing and a dual-adaptation fusion mechanism, it achieves deep adaptation and modular organization analysis of the model dimension and data dimension. This significantly improves the parsing efficiency, adaptation accuracy, and engineering practicality of complex equipment digital models in systematic evaluation and computational analysis, providing strong technical support for model applications throughout the entire equipment lifecycle.

[0041] In an optional embodiment of the present invention, the step of obtaining the overall requirements and combining the overall requirements with empirical data and knowledge graphs to generate model requirements includes: Based on the indicator description documents for equipment development and design, and the test description documents for testing and evaluation, the overall requirements for model evaluation and computational analysis are obtained. By pre-constructing a knowledge graph of the equipment system and digital model, the mission-level indicators corresponding to the overall requirements are analyzed and decomposed to form detailed indicator requirement description documents matching the test subjects. Based on the detailed indicator requirement description documents, the evaluation requirements of professional fields are matched, and combined with empirical data and test subject requirements, a multi-level evaluation indicator system description document is constructed. Using the multi-level evaluation indicator system as a constraint, test resource requirements are matched from the professional dimensions involved in the test subjects to form a meta-model professional description document. Based on the meta-model professional description document, model clustering analysis is performed, and categories are divided from the evaluation task and application business levels to generate meta-model category description documents. Finally, the meta-model category description documents are integrated to generate model requirements.

[0042] In this embodiment, the evaluation target decomposition and adaptation plugin obtains the overall requirements for model evaluation and computational analysis through a structured file interface. Based on empirical data and knowledge graphs, it achieves intelligent target decomposition and rapid matching of equipment models, generates domain model requirements, and guides model loading and meta-model decomposition. Specifically, it includes: Step 1: Identifying Assessment Objectives; Step 2: Goal self-analysis; Step 3: Load the indicator system; Step 4: Evaluation indicator requirements are called by the model decomposition and logic parsing engine and the proxy model encapsulation and verification engine to generate corresponding description files; Step 5: Obtain multi-disciplinary digital model description files from the model decomposition and logic parsing engine; Step 6: Professional model adaptation; Step 7: Model category segmentation, generating meta-model description files categorized by category, which are then called by the model decomposition and logic parsing engine to carry out subsequent parsing work.

[0043] In an optional embodiment of the present invention, the step of generating a metamodel description file based on the model composition file and the model description file, and generating a metamodel based on the metamodel description file, includes: Based on the model composition file and model description file, the dispatched equipment system model is decomposed and logically analyzed to determine the professions, fields and forms of expression involved in the meta-model, and to form the element analysis results. Analyze the key core parameters and feature factors of each type of model, and generate meta-model construction description files; Based on the tool types and formats of the metamodel, the encapsulation protocols, data connection and transmission specifications of input and output interfaces are parsed, interface matching is completed and metamodel interface description files are generated. The metamodel construction description file and the metamodel interface description file are integrated to generate a metamodel description file, and a metamodel is generated based on the metamodel description file.

[0044] In this embodiment, the model decomposition and logic parsing engine loads and schedules the model composition file, obtains the model category classification description file, and parses, decomposes, and forms a metamodel description file from three levels: elements, tools, and interfaces, according to the model composition, format, and application requirements, and generates a metamodel module; specifically including: Step 1: Call the evaluation indicator requirements generated by the evaluation target decomposition and adaptation plugin, obtain multi-professional models based on the requirements, and generate multi-professional digital model description files. Send these files to the evaluation target decomposition and adaptation plugin to guide the requirement analysis work. Simultaneously generate integrated model application requirements and send them to the proxy model encapsulation and verification engine to guide the constraint analysis work. Step 1: Based on the category-based metamodel description files, conduct multi-dimensional analysis of the metamodel; Step 2: Metamodel element analysis; Step 3: Metamodel tool analysis; Step 4: Generate meta-model features and modeling tool description files on the model dimension, and send them to the proxy model encapsulation and verification engine; Step 5: Metamodel Interface Parsing; Step 6: Generate a meta-model interface description file and parameter list on the data dimension, and send them to the logic fitting and interface adaptation engine; Step 7: Generate the basic meta-model module.

[0045] In an optional embodiment of the present invention, the step of analyzing the input-output data logic of the meta-model, redefining the interface using nonlinear iteration and aggregation optimization methods according to the meta-model interface specification scheduling parameters, and simultaneously achieving numerical fitting of the transmission logic includes: The underlying physical mechanism logic, intermediate data representation logic, and top-level business application logic of the meta-model are analyzed in their entirety to generate a standardized data logic graph. Based on the meta-model interface specification and the data logic graph, the associated parameters at each level are matched and scheduled to build a logic parameter interface linkage mechanism. A combination of nonlinear iterative algorithm and aggregation optimization method is adopted. With input and output data as the driving force, the meta-model transitive logic numerical fitting model is constructed by alternating between prior input response and random input response, and the transitive logic numerical model description file is generated simultaneously. Based on the aforementioned transmission logic numerical model description file, the underlying logic interface, intermediate representation interface, and top-level application interface of the meta-model are uniformly and standardizedly redefined to clarify the interface data format, transmission protocol, and interaction rules.

[0046] In this embodiment, the logic fitting and interface adaptation engine is a vertical data dimension analysis engine. It acquires the meta-model description file, analyzes the input and output data logic of the meta-model, schedules parameters according to the meta-model interface specification, and uses nonlinear iteration and aggregation optimization methods to achieve numerical fitting of the transmitted logic, thus redefining the interface. Specifically, this includes: Step 1: Obtain the metamodel interface description file and parameter list sent by the model decomposition and logic parsing engine, and carry out metamodel parameter scheduling; Step 2: Obtain the basic meta-model module sent by the model decomposition and logic parsing engine, perform logical numerical fitting, generate a meta-model data dimension logic description file, and send it to the proxy model encapsulation and verification engine. Step 3: Redefine the meta-model interface, including the underlying logic interface, intermediate representation interface, and top-level application interface. The underlying logic parameter relationships are embedded in the model's physical mechanism model, explaining the mathematical implementation of the physical mechanism. The intermediate representation parameter relationships are embedded in the model interface protocol, describing the data transmission and output methods. The top-level application parameter relationships are embedded in the model's business application, clarifying the mutual support between the model and other business modules. Step 4: Generate the redefined metamodel module and send it to the proxy model encapsulation and validation engine.

[0047] In one optional embodiment of the present invention, the step of associating and adapting the result of fitting the transmitted logic numerical value and encapsulating it by proxy, and completing the numerical verification in conjunction with precision constraints, includes: Load the numerical fitting results of the transmission logic, the full element information of the meta-model and the description file of the accuracy constraints, and sort out the full elements of the meta-model from the horizontal model dimension to obtain the full element analysis results of the meta-model. Based on the full-element analysis results of the meta-model, the numerical fitting results of the transmission logic are precisely matched with the meta-model operation logic and interface logic to form an association mapping relationship between the fitting results and the meta-model logic. Based on the aforementioned correlation mapping relationship, the fitting results are reduced in order to obtain the initial meta-proxy model; Based on the accuracy constraint description file, the initial meta-surrogate model is driven to perform numerical calculations, the output data is obtained and compared with the accuracy constraint standard to obtain the error results; The model parameters are dynamically adjusted based on the error results through a feedback control network until the initial meta-surrogate model meets the accuracy constraint requirements. Based on the validated meta-proxy model, a standardized meta-proxy model description file is generated.

[0048] In this embodiment, the proxy model encapsulation and verification engine is a horizontal model dimension analysis engine. It obtains the metamodel description file and the metamodel interface logic description file. Based on the application capabilities of the metamodel module, it performs correlation adaptation and proxy encapsulation on the numerical fitting results of the metamodel transmission logic, and performs numerical verification based on the metamodel accuracy requirement constraints. Specifically, this includes: Step 1: Schedule and load the evaluation target decomposition and adaptation plugin-generated evaluation indicator requirements, as well as the integrated model application requirements generated by the model decomposition and logic parsing engine; Step 2: Calculate and obtain the running accuracy constraints; Step 1: Schedule and load the meta-model elements and modeling tool description files generated by the model decomposition and logic parsing engine to achieve horizontal reconstruction of the meta-model; Step 3: Generate the metamodel description file after horizontal reconstruction and send it to the business-level model reconstruction plugin; Step 4: Schedule and load the meta-model parameters and fitting logic description files generated by the logic fitting and interface adaptation engine to achieve vertical reconstruction of the meta-model; Step 5: Generate the metamodel description file after vertical reconstruction and send it to the business-level model reconstruction plugin; Step 6: Schedule and load the meta-model module after the interface redefinition generated by the logic fitting and interface adaptation engine to complete the overall encapsulation of the meta-model; Step 7: Meta-model verification based on accuracy constraints; Step 8: Generate the reconstructed metamodel description file and send it to the business-level model reconstructing plugin.

[0049] In an optional embodiment of the present invention, the process of adapting the association relationship of the meta-proxy model to generate a business-level numerical proxy model, and using scheduling parameters to verify the accuracy of the business-level numerical proxy model to generate a unified and runnable system model includes: Obtain standardized meta-proxy model description files, instances of each meta-proxy model, and the business flow and information flow specifications of the original digital model; Integrate the data dimension and model dimension proxy model, fuse and match based on interface description protocol, obtain the input, output and time sequence logic between models, and generate a logic adaptation description file between models. Based on the adaptation description file, the meta-proxy models are associated and assembled to generate a business-level numerical proxy model; The system schedules preset verification parameters and datasets to verify the data, computation, timing, and event logic of the proxy model and obtains the verification results. Based on the verification results, optimize the model parameters or solidify the model form to generate a unified, runnable system model.

[0050] In this embodiment, the business-level model reconstruction plugin obtains the encapsulated meta-proxy model, adopts a normalized and standardized model output format, adapts the relationships between meta-proxy models, generates a business-level, numerical proxy model, realizes the fusion of meta-proxy models, performs accuracy verification on scheduling parameter data, and generates a unified and runnable system model, specifically including: Step 1: Obtain the metamodel description file after horizontal reconstruction generated by the proxy model encapsulation and verification engine; Step 2: Model dimensional logic verification; Step 3: Obtain the vertically reconstructed metamodel description file generated by the proxy model encapsulation and verification engine; Step 4: Data dimension logic validation; Step 5: Obtain the reconstructed metamodel description file generated by the proxy model encapsulation and verification engine; Step 6: Logic adaptation between models; Step 7: Business model generation; Step 8: Model accuracy verification.

[0051] Figure 2 The working link diagram of a target-driven digital model orthogonal analytic and dual-adaptive fusion system provided in this application embodiment is as follows: Figure 2 As shown, the target-driven digital model orthogonal analytic and dual-adaptation fusion system includes: The evaluation target decomposition and adaptation plugin is used to achieve evaluation target acquisition, autonomous target analysis, indicator system construction, professional model adaptation, and model category classification. Specifically, it enables the rapid, efficient, and accurate acquisition of model evaluation targets, matching with domain indicator systems, and classifying digital model meta-models. The model decomposition and logic parsing engine is used to parse meta-model elements, tools, interfaces, and generate meta-model modules. Specifically, based on meta-model classification descriptions and requirements, it enables automatic analysis of specific digital models, parses the comprehensive logic of meta-models for each category, and divides them into meta-model modules. The logic fitting and interface adaptation engine is used to implement metamodel parameter scheduling, logical numerical fitting, and metamodel interface redefinition. Specifically, in the vertical data dimension, it fits the metamodel logic from the input and output levels of the specified metamodel, constructs the input and output numerical mapping, and adapts the interface. The proxy model encapsulation and validation engine is used to encapsulate the meta-proxy model and validate runtime accuracy constraints. Specifically, in the horizontal model dimension, proxy model fusion is used to eliminate the isolation caused by differences in different meta-model building tools and file formats, and to build a reduced-order meta-proxy model based on runtime accuracy constraints. The business-level model refactoring plugin is used to achieve logical adaptation between models, business model generation, and model accuracy verification. Specifically, it integrates the meta-proxy model based on interface logic, reconstructs a systematic digital model, and verifies the refactored and encapsulated model, using standard prior data to verify the numerical logic.

[0052] Figure 3 This is a schematic diagram of the structure of a target-driven digital model orthogonal analytic and dual-adaptive fusion system provided in an embodiment of this application, as shown below. Figure 3 As shown, the system includes: The evaluation target decomposition and adaptation plugin 301 is used to obtain the overall requirements and combine the overall requirements with empirical data and knowledge graphs to generate model requirements to guide model loading and metamodel decomposition. The model decomposition and logic parsing engine 302 is used to obtain the model composition file and the model description file for classifying the model categories based on the model requirements, generate a meta-model description file based on the model composition file and the model description file, and generate a meta-model based on the meta-model description file. The logic fitting and interface adaptation engine 303 is used to analyze the input and output data logic of the meta-model, schedule parameters according to the meta-model interface specification, redefine the interface using nonlinear iteration and aggregation optimization methods, and simultaneously realize the numerical fitting of the transmitted logic. The proxy model encapsulation and verification engine 304 is used to perform correlation adaptation and proxy encapsulation on the result of the numerical fitting of the transmitted logic, and to complete the numerical verification in combination with the precision constraint conditions. The business-level model reconstruction plugin 305 is used to adapt the association relationship of the meta-proxy model, generate a business-level numerical proxy model, and perform accuracy verification of the business-level numerical proxy model using scheduling parameters to generate a unified and runnable system model.

[0053] In an optional embodiment of the present invention, the evaluation target decomposition and adaptation plugin includes: The evaluation target acquisition module is used to obtain the overall requirements for model evaluation and calculation analysis based on the indicator description documents of equipment development and design and the test description documents of test evaluation. The target autonomous analysis module is used to analyze and decompose the mission-level indicators corresponding to the overall requirements by pre-constructing equipment system and digital model knowledge graph, and form a detailed indicator requirement description document that matches the test subjects; The indicator system construction module is used to construct a multi-level evaluation indicator system description file based on the detailed indicator requirement description file, matching the indicator evaluation requirements of the professional field, and combining empirical data and test subject requirements. The model professional adaptation module is used to match the test resource requirements from the professional dimensions involved in the test subjects, based on the constraints of the multi-level evaluation index system, and form a meta-model professional description file. The model category segmentation module is used to perform model clustering analysis based on the meta-model professional description file, segment categories from the evaluation task and application business level, generate meta-model category description files, and integrate the meta-model category description files to generate model requirements.

[0054] In one optional embodiment of the present invention, the model decomposition and logic parsing engine includes: The original model element analysis module is used to perform element decomposition and logical analysis on the dispatched equipment system model based on the model composition file and model description file, determine the profession, field and expression form of the meta-model, and form the element analysis result. The metamodel interface parsing module is used to parse the key core parameters and feature factors of various types of models and generate metamodel construction description files. The metamodel tool parsing module is used to parse the encapsulation protocol, data connection and transmission specifications of input and output interfaces based on the tool type and format of the metamodel, complete interface matching and generate metamodel interface description files; The metamodel module generation module is used to integrate the metamodel construction description file and the metamodel interface description file to generate a metamodel description file, and generate a metamodel based on the metamodel description file.

[0055] In one optional embodiment of the present invention, the logic fitting and interface adaptation engine includes: The meta-model parameter scheduling module is used to perform full-link analysis of the underlying physical mechanism logic, intermediate data representation logic and top-level business application logic of the meta-model, generate a standardized data logic graph, and match and schedule the associated parameters at each level according to the meta-model interface specification and data logic graph to build a logical parameter interface linkage mechanism. The transit logic numerical fitting module is used to construct a meta-model transit logic numerical fitting model by combining nonlinear iterative algorithms and aggregation optimization methods, guided by input and output data, and by iteratively alternating between prior input response and random input response, and simultaneously generating a transit logic numerical model description file. The metamodel interface redefinition module is used to redefine the underlying logic interface, intermediate representation interface and top-level application interface of the metamodel in a unified and standardized manner based on the transmitted logic numerical model description file, so as to clarify the interface data format, transmission protocol and interaction rules.

[0056] Figure 4 A flowchart illustrating the process of a target-driven orthogonal analytical and dual-adaptive fusion system for digital models provided in this application is shown below. Figure 4 As shown: The model decomposition and logic parsing engine is used to parse meta-model elements, tools, interfaces, and generate meta-model modules. Specifically, based on meta-model classification descriptions and requirements, it enables automatic analysis of specific digital models, parses the comprehensive logic of meta-models for each category, and divides them into meta-model modules. The logic fitting and interface adaptation engine is used to implement metamodel parameter scheduling, logical numerical fitting, and metamodel interface redefinition. Specifically, in the vertical data dimension, it fits the metamodel logic from the input and output levels of the specified metamodel, constructs the input and output numerical mapping, and adapts the interface. The proxy model encapsulation and validation engine is used to encapsulate the meta-proxy model and validate runtime accuracy constraints. Specifically, in the horizontal model dimension, proxy model fusion is used to eliminate the isolation caused by differences in different meta-model building tools and file formats, and to build a reduced-order meta-proxy model based on runtime accuracy constraints. The business-level model refactoring plugin is used to achieve logical adaptation between models, business model generation, and model accuracy verification. Specifically, it integrates the meta-proxy model based on interface logic, reconstructs a systematic digital model, and verifies the refactored and encapsulated model, using standard prior data to verify the numerical logic.

[0057] Specifically, in the evaluation target decomposition and adaptation plugin, the evaluation target acquisition is based on semi-structured and unstructured documents such as indicator description documents in equipment research and design and test description documents in test evaluation. Natural language recognition technology is used to dynamically acquire the equipment digital model evaluation and calculation analysis requirements and targets based on empirical data judgment. The target autonomous analysis, through the pre-built and adaptively improved knowledge graph of the equipment system and digital model, intelligently analyzes and decomposes mission-level indicators such as digital model evaluation and calculation analysis requirements and targets, forming a detailed indicator requirement description document that matches the test subjects; The aforementioned indicator system is constructed based on a detailed indicator requirement description document, which quickly matches the indicator evaluation requirements of professional fields. Based on experience data and test subject requirements, a complete multi-level evaluation indicator system description document is formed. The model is professionally adapted to obtain a multi-level evaluation index system description file. Guided by index constraints, the model matches the test resource requirements from the perspective of the professional aspects involved in the test subjects, and forms a model meta-model professional description file. The model category classification is based on the meta-model professional description file, and further model clustering analysis is performed to break through tool limitations. The categories are divided from the perspective of evaluation tasks and application business, and a meta-model category description file is generated.

[0058] In the model decomposition and logical parsing engine, the meta-model element parsing involves scheduling, element decomposition, and logical parsing of the equipment system model according to the meta-model category description file, and scheduling and parsing model elements such as the profession, field, and form of representation involved in the meta-model. The meta-model tool analyzes the digital model according to its composition and structure. It designs and implements a universal adaptable tool compatible with multiple types of models, such as 3D structures, mathematical logic, and behavioral simulation, and provides a variety of formats. It analyzes the key core parameters and factors of each type of model, including the appearance, material, and composition factors of 3D structures, the operation factors and numerical domain factors of mathematical logic, and the triggering simulation logic, timing simulation logic, scheduling simulation logic, and constraint simulation logic of behavioral simulation, and generates a meta-model construction description file. The metamodel interface parsing process, based on metamodel category description files and metamodel tools and formats, parses input and output interfaces to achieve parsing and matching of interface encapsulation protocols, data connection and transmission specifications, and generates metamodel interface description files. The metamodel module generation integrates the metamodel construction description file and the metamodel interface description file, further refines the metamodel description file, generates a structured metamodel module, and provides a generation framework for the given model dimension and data dimension.

[0059] In the logic fitting and interface adaptation engine, the meta-model parameter scheduling realizes the parameter relationship between the underlying logic, intermediate representation and top-level application of the decomposed model, supports the multi-dimensional integrated adaptation method of "mechanism explanation + numerical fitting + application verification", and provides parameter scheduling and driving force for meta-model data fitting. The transfer logic numerical fitting performs logical analysis on each meta-model module from the vertical data dimension. With input and output as the driving force, it constructs a numerical logic fitting model for the meta-model module, realizes data integration of the meta-model, and establishes the transfer logic numerical model by alternating between prior input response and random input response, and generates a transfer logic numerical model description file. The meta-model interface redefinition, based on the transfer logic numerical model description file, adopts a multi-dimensional integrated adaptation method of "mechanism interpretation + numerical fitting + application verification" to redefine the component-level meta-model interface for specific fields, specific professions, specific businesses, and specific forms, and generate a meta-model interface numerical logic description file.

[0060] In the proxy model encapsulation and verification engine, the meta-proxy model encapsulation performs a full meta-model element analysis on each meta-model module from the horizontal model dimension, including meta-model specialization, meta-model type, meta-model tools, etc. Based on the meta-model interface numerical logic description file, it strips the original form, format and independent specific interface of the meta-model, matches the meta-model operation logic and interface logic, and performs a reduction process on the meta-model to obtain the meta-proxy model. The operational accuracy constraint verification is based on the operational accuracy constraint description file of the professional component-level meta-model in this professional field combined with business needs. The operational accuracy constraint verification adopts the prior input response method. The prior input data drives the meta-proxy model to perform numerical calculations, obtain output data and perform constraint verification, and through the feedback control network, the determined error is used for feedback adjustment to obtain the meta-proxy model output that meets the constraints.

[0061] In the business-level model reconstruction plugin, the logical adaptation between models integrates data dimensions and model meta-proxy models, and completes fusion matching based on the interface description protocol. It also performs logical adaptation between meta-proxy models based on the business flow and information flow of the original digital model, matching and adaptively streamlining the logic of the preceding model input and the following model output, and generating a logical adaptation description file between models. The business model generation, based on the logical adaptation description file between models, combines and reconstructs the complete system model, realizes the logical association and assembly of the meta-proxy model, and generates a business-level proxy model. The model accuracy verification is based on the business-level accuracy constraint description file. The "integrated engine" loads, schedules, and runs the business-level proxy model for verification, including underlying logic such as data logic, operation logic, timing logic, and event logic, to achieve logic adaptation between models, business model generation, and model accuracy verification.

[0062] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0063] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also include the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0064] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A goal-driven method for fusing orthogonal analytic and dual-adaptation digital models, characterized in that, The method includes: Obtain overall requirements, and combine these overall requirements with empirical data and knowledge graphs to generate model requirements to guide model loading and metamodel decomposition; Obtain the model composition file and the model description file for classifying model categories based on the model requirements; generate a meta-model description file based on the model composition file and the model description file; and generate a meta-model based on the meta-model description file. The input and output data logic of the meta-model is analyzed. Based on the scheduling parameters of the meta-model interface specification, the interface is redefined using nonlinear iteration and aggregation optimization methods, and the numerical fitting of the transmission logic is realized simultaneously. The results of fitting the numerical values ​​of the transmitted logic are associated, adapted, and encapsulated by a proxy, and numerical verification is completed in combination with precision constraints. Adapt the association relationships of the meta-proxy model to generate a business-level numerical proxy model, and use scheduling parameters to verify the accuracy of the business-level numerical proxy model to generate a unified and runnable system model.

2. The method according to claim 1, characterized in that, The process of obtaining overall requirements and combining these requirements with empirical data and knowledge graphs to generate model requirements includes: Based on the indicator description documents of equipment development and design and the test description documents of test evaluation, the overall requirements for model evaluation and calculation analysis are obtained. By pre-constructing equipment systems and digital model knowledge graphs, the mission-level indicators corresponding to the overall requirements are analyzed and decomposed to form detailed indicator requirement description documents that match the test subjects; Based on the detailed indicator requirement description document, matching the indicator evaluation requirements of professional fields, and combining empirical data and test subject requirements, a multi-level evaluation indicator system description document is constructed. Using the aforementioned multi-level evaluation index system as a constraint and guide, the test resource requirements are matched from the professional dimensions involved in the test subjects to form a meta-model professional description document; Based on the meta-model professional description file, model clustering analysis is performed to classify categories from the perspectives of assessment tasks and application business, generate meta-model category description files, and integrate the meta-model category description files to generate model requirements.

3. The method according to claim 1, characterized in that, The step of generating a metamodel description file based on the model composition file and the model description file, and generating a metamodel based on the metamodel description file, includes: Based on the model composition file and model description file, the dispatched equipment system model is decomposed and logically analyzed to determine the professions, fields and forms of expression involved in the meta-model, and to form the element analysis results. Analyze the key core parameters and feature factors of each type of model, and generate meta-model construction description files; Based on the tool types and formats of the metamodel, the encapsulation protocols, data connection and transmission specifications of input and output interfaces are parsed, interface matching is completed and metamodel interface description files are generated. The metamodel construction description file and the metamodel interface description file are integrated to generate a metamodel description file, and a metamodel is generated based on the metamodel description file.

4. The method according to claim 1, characterized in that, The analysis of the input-output data logic of the meta-model involves redefining the interface using nonlinear iteration and aggregation optimization methods based on the meta-model interface specification scheduling parameters, and simultaneously implementing numerical fitting of the transmitted logic, including: The underlying physical mechanism logic, intermediate data representation logic, and top-level business application logic of the meta-model are analyzed in their entirety to generate a standardized data logic graph. Based on the meta-model interface specification and the data logic graph, the associated parameters at each level are matched and scheduled to build a logic parameter interface linkage mechanism. A combination of nonlinear iterative algorithm and aggregation optimization method is adopted. With input and output data as the driving force, the meta-model transitive logic numerical fitting model is constructed by alternating between prior input response and random input response, and the transitive logic numerical model description file is generated simultaneously. Based on the aforementioned transmission logic numerical model description file, the underlying logic interface, intermediate representation interface, and top-level application interface of the meta-model are uniformly and standardizedly redefined to clarify the interface data format, transmission protocol, and interaction rules.

5. The method according to claim 1, characterized in that, The process of associating, adapting, and encapsulating the results of fitting the transmitted logic numerical values, and performing numerical verification in conjunction with precision constraints, includes: Load the numerical fitting results of the transmission logic, the full element information of the meta-model and the description file of the accuracy constraints, and sort out the full elements of the meta-model from the horizontal model dimension to obtain the full element analysis results of the meta-model. Based on the full-element analysis results of the meta-model, the numerical fitting results of the transmission logic are precisely matched with the meta-model operation logic and interface logic to form an association mapping relationship between the fitting results and the meta-model logic. Based on the aforementioned correlation mapping relationship, the fitting results are reduced in order to obtain the initial meta-proxy model; Based on the accuracy constraint description file, the initial meta-surrogate model is driven to perform numerical calculations, the output data is obtained and compared with the accuracy constraint standard to obtain the error results; The model parameters are dynamically adjusted based on the error results through a feedback control network until the initial meta-surrogate model meets the accuracy constraint requirements. Based on the validated meta-proxy model, a standardized meta-proxy model description file is generated.

6. The method according to claim 1, characterized in that, The process involves adapting the relationships of the meta-proxy model to generate a business-level numerical proxy model, using scheduling parameters to verify the accuracy of the business-level numerical proxy model, and generating a unified and runnable system model, including: Obtain standardized meta-proxy model description files, instances of each meta-proxy model, and the business flow and information flow specifications of the original digital model; Integrate the data dimension and model dimension proxy model, fuse and match based on interface description protocol, obtain the input, output and time sequence logic between models, and generate a logic adaptation description file between models. Based on the adaptation description file, the meta-proxy models are associated and assembled to generate a business-level numerical proxy model; The system schedules preset verification parameters and datasets to verify the data, computation, timing, and event logic of the proxy model and obtains the verification results. Based on the verification results, optimize the model parameters or solidify the model form to generate a unified, runnable system model.

7. A target-driven digital model orthogonal analytic and dual-adaptation fusion system, used to implement the target-driven digital model orthogonal analytic and dual-adaptation fusion method according to any one of claims 1-6, characterized in that, The system includes: The evaluation target decomposition and adaptation plugin is used to obtain the overall requirements and combine the overall requirements with empirical data and knowledge graphs to generate model requirements, so as to guide model loading and metamodel decomposition. The model decomposition and logic parsing engine is used to obtain the model composition file and the model description file for classifying the model categories based on the model requirements, generate a meta-model description file based on the model composition file and the model description file, and generate a meta-model based on the meta-model description file. The logic fitting and interface adaptation engine is used to analyze the input and output data logic of the meta-model, schedule parameters according to the meta-model interface specification, redefine the interface using nonlinear iteration and aggregation optimization methods, and simultaneously realize the numerical fitting of the transmitted logic. The proxy model encapsulation and verification engine is used to perform correlation adaptation and proxy encapsulation on the results of the numerical fitting of the transmitted logic, and to complete numerical verification in combination with precision constraints. A business-level model reconstruction plugin is used to adapt the association relationship of the meta-proxy model, generate a business-level numerical proxy model, and perform accuracy verification of the business-level numerical proxy model using scheduling parameters to generate a unified and runnable system model.

8. The system according to claim 7, characterized in that, The evaluation target decomposition and adaptation plugin includes: The evaluation target acquisition module is used to obtain the overall requirements for model evaluation and calculation analysis based on the indicator description documents of equipment development and design and the test description documents of test evaluation. The target autonomous analysis module is used to analyze and decompose the mission-level indicators corresponding to the overall requirements by pre-constructing equipment system and digital model knowledge graph, and form a detailed indicator requirement description document that matches the test subjects; The indicator system construction module is used to construct a multi-level evaluation indicator system description file based on the detailed indicator requirement description file, matching the indicator evaluation requirements of the professional field, and combining empirical data and test subject requirements. The model professional adaptation module is used to match the test resource requirements from the professional dimensions involved in the test subjects, based on the constraints of the multi-level evaluation index system, and form a meta-model professional description file. The model category segmentation module is used to perform model clustering analysis based on the meta-model professional description file, segment categories from the evaluation task and application business level, generate meta-model category description files, and integrate the meta-model category description files to generate model requirements.

9. The system according to claim 7, characterized in that, The model decomposition and logic parsing engine includes: The original model element analysis module is used to perform element decomposition and logical analysis on the dispatched equipment system model based on the model composition file and model description file, determine the profession, field and expression form of the meta-model, and form the element analysis result. The metamodel interface parsing module is used to parse the key core parameters and feature factors of various types of models and generate metamodel construction description files. The metamodel tool parsing module is used to parse the encapsulation protocol, data connection and transmission specifications of input and output interfaces based on the tool type and format of the metamodel, complete interface matching and generate metamodel interface description files; The metamodel module generation module is used to integrate the metamodel construction description file and the metamodel interface description file to generate a metamodel description file, and generate a metamodel based on the metamodel description file.

10. The system according to claim 7, characterized in that, The logic fitting and interface adaptation engine includes: The meta-model parameter scheduling module is used to perform full-link analysis of the underlying physical mechanism logic, intermediate data representation logic and top-level business application logic of the meta-model, generate a standardized data logic graph, and match and schedule the associated parameters at each level according to the meta-model interface specification and data logic graph to build a logical parameter interface linkage mechanism. The transit logic numerical fitting module is used to construct a meta-model transit logic numerical fitting model by combining nonlinear iterative algorithms and aggregation optimization methods, guided by input and output data, and by iteratively alternating between prior input response and random input response, and simultaneously generating a transit logic numerical model description file. The metamodel interface redefinition module is used to redefine the underlying logic interface, intermediate representation interface and top-level application interface of the metamodel in a unified and standardized manner based on the transmitted logic numerical model description file, so as to clarify the interface data format, transmission protocol and interaction rules.

Images