A method and apparatus for checking an equipment digital model

Abstract

The application discloses a kind of equipment digital model verification method and device, the device includes: resource support subsystem, service support subsystem and verification evaluation subsystem;The resource support subsystem is connected with the service support subsystem, for unified management to data, model, resource, system etc.;The service support subsystem is connected with the verification evaluation subsystem, for the equipment digital model according to the specification interface is integrated to multi-source heterogeneous, according to the simulation demand received, equipment digital model is carried out joint simulation operation, and simulation operation result is obtained.The verification evaluation subsystem is used for according to specific test task, by selecting diversified test elements, generating test scheme information, according to test scheme generating simulation demand, the simulation demand is sent to the service support subsystem, the simulation operation result obtained by the service support subsystem is verified and evaluated, and the verification result of equipment digital model is obtained.

Description

Technical Field

[0001] This invention relates to the fields of model simulation, dynamic modeling of industrial mechanisms, and digital evaluation of equipment, and specifically to a method and apparatus for verifying digital models of equipment. Background Technology

[0002] With the deep application of digital engineering throughout the entire equipment lifecycle, traditional on-site testing methods are no longer sufficient to meet the efficient verification needs of complex equipment systems. There is an urgent need to address key issues such as inconsistent model standards, the lack of a data-physical integration mechanism, and insufficient test design methods through digital model verification platforms. The core issue that urgently needs to be addressed is how to build a multi-level digital model verification system covering the system, platform, and unit levels, based on the current urgent needs for digital equipment testing and evaluation, and relying on the core requirements of relevant standards and specifications regarding the confidence level of digital models. This system enables the synchronous verification and iterative optimization of equipment digital models and physical equipment. Summary of the Invention

[0003] This invention primarily addresses the urgent need for digital testing and evaluation of current equipment, relying on the core requirements of relevant standards and specifications for the confidence level of digital models. By constructing a multi-level digital model verification system covering the system level, platform level, and unit level, this invention achieves the synchronous verification and iterative optimization of equipment digital models and physical equipment. This invention discloses a verification method and apparatus for equipment digital models.

[0004] In a first aspect, the present invention discloses a verification device for an equipment digital model, comprising: a resource support subsystem, a service support subsystem, and a verification and evaluation subsystem;

[0005] The resource support subsystem is connected to the service support subsystem and is used for unified management of data, models, resources, systems, etc.

[0006] The service support subsystem is connected to the verification and evaluation subsystem. It is used to integrate multi-source heterogeneous equipment digital models according to the standard interface, and to perform joint simulation operation on the equipment digital models according to the received simulation requirements to obtain simulation results.

[0007] The verification and evaluation subsystem is used to generate test plan information by selecting diverse test elements according to specific test tasks, generate simulation requirements according to the test plan, send the simulation requirements to the service support subsystem, verify and evaluate the simulation operation results obtained by the service support subsystem, and obtain the verification results of the equipment digital model; the test plan information includes the expected results of the equipment digital model.

[0008] The resource support subsystem includes a data management module, a model management module, a resource management module, and a system management module.

[0009] The data management module is used to collect, manage and process the test data required for the evaluation, and to store the collected test data in the database.

[0010] The model management module is used to manage platform-level models, unit-level models, and related test models;

[0011] The resource management module is used to achieve unified management of various resources for the experiment; the various resources for the experiment include indicator system, indicator model, visualization template, report template, model operator, data file library, evaluation model library and microservice program;

[0012] The system management module is used to manage system logs, roles, users, and permissions for the verification device of the equipment digital model.

[0013] The service support subsystem integrates multi-source heterogeneous equipment digital models according to standardized interfaces by connecting the multi-source heterogeneous equipment digital models to the interconnection middleware. The interconnection middleware connects the heterogeneous models to the simulation operation support module of the service support subsystem, and builds a transmission mechanism based on physical networks and IP protocols, using publish-subscribe and remote transmission methods as the basic means.

[0014] The verification and evaluation subsystem verifies and evaluates the simulation results obtained by the service support subsystem to obtain the verification results of the equipment digital model, including:

[0015] Static consistency analysis is performed on the simulation results obtained from the service support subsystem and the static results of the expected results of the equipment digital model in the test plan information to obtain the first verification result value; the simulation results are a data sequence of the equipment model state; the static and dynamic results of the expected results of the equipment digital model are both represented as the expected data sequence form of the equipment model state.

[0016] Dynamic consistency analysis is performed on the simulation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain a second verification result value.

[0017] The first and second verification result values ​​are fused and estimated to obtain the fused verification result value.

[0018] Determine whether the fusion verification result value is greater than a preset discrimination threshold value to obtain a first discrimination result;

[0019] If the first discrimination result is greater than, the verification result of the equipment digital model is determined to be passed; if the first discrimination result is not greater than, the verification result of the equipment digital model is determined to be failed.

[0020] The static consistency analysis includes:

[0021] The data sequence of each parameter of the equipment model state in the simulation results is represented as the corresponding state row vector;

[0022] The expected data sequence of each parameter of the equipment model state in the static results is represented as the corresponding standard row vector;

[0023] Subtract the state row vector from the standard row vector for each parameter to obtain the corresponding difference row vector;

[0024] The difference matrix is ​​constructed by using the difference row vectors of all parameters as row vectors;

[0025] The cross-correlation value is calculated on the difference matrix to obtain the cross-correlation matrix;

[0026] The cross-correlation matrix is ​​subjected to eigenvalue decomposition to obtain an eigenvector; the eigenvector is a vector composed of all eigenvalues.

[0027] For each row vector of the difference matrix, statistical features are calculated to obtain the corresponding statistical feature values;

[0028] The expression for calculating the statistical features is:

[0029] ,

[0030] in, Let be the statistical characteristic value of the i-th row. and These are the elements in the i-th row and i-th column of the difference matrix, and the elements in the i-th row and j-th column, respectively, where N is the column dimension of the difference matrix. The maximum value of the vector in the i-th row of the difference matrix;

[0031] Multiply the statistical eigenvalues ​​corresponding to each row vector of the difference matrix to obtain the multiplication result of each row vector; then sum the multiplication results of all row vectors to obtain the first verification result value.

[0032] A second aspect of this invention discloses a method for verifying a digital model of equipment, implemented using the aforementioned verification device for the digital model of equipment, comprising:

[0033] S1. Using the verification and evaluation subsystem, based on the specific test task, test plan information is generated by selecting diverse test elements, simulation requirements are generated based on the test plan, and the simulation requirements are sent to the service support subsystem.

[0034] S2, using the service support subsystem, perform joint simulation of the equipment digital model according to the received simulation requirements, and obtain the simulation results;

[0035] S3. Using the verification and evaluation subsystem, the simulation operation results obtained by the service support subsystem are verified and evaluated to obtain the verification results of the equipment digital model.

[0036] The process of using the verification and evaluation subsystem to verify and evaluate the simulation results obtained from the service support subsystem to obtain the verification results of the equipment digital model includes:

[0037] Static consistency analysis is performed on the static results of the simulation operation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain the first verification result value;

[0038] Dynamic consistency analysis is performed on the simulation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain a second verification result value.

[0039] The first and second verification result values ​​are fused and estimated to obtain the fused verification result value.

[0040] Determine whether the fusion verification result value is greater than a preset discrimination threshold value to obtain a first discrimination result;

[0041] If the first discrimination result is greater than, the verification result of the equipment digital model is determined to be passed; if the first discrimination result is not greater than, the verification result of the equipment digital model is determined to be failed.

[0042] A third aspect of the present invention discloses a verification device for an equipment digital model, the device comprising:

[0043] Memory containing executable program code;

[0044] A processor coupled to the memory;

[0045] The processor calls the executable program code stored in the memory to execute the verification method of the equipment digital model.

[0046] In a fourth aspect of this invention, a computer-readable storage medium is disclosed, the computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the verification method for the equipment digital model.

[0047] In a fifth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the verification method for the equipment digital model.

[0048] The beneficial effects of this invention are as follows:

[0049] This invention integrates the operational and computational digital models within the equipment's digital model to construct a verification platform that combines model calibration and validation, supporting multi-dimensional assessment of digital equipment in performance testing, operational testing, and system testing. Relying on a high-performance simulation engine and a distributed computing environment, this platform enables dynamic comparison and analysis between the digital model and actual equipment test data, improving the confidence level of the model in terms of tactical and technical indicators, operational effectiveness, and applicability, and providing authoritative data support for the synchronized finalization and deployment of equipment.

[0050] This invention addresses the problems of poor model interoperability and fragmented verification processes in existing testing and evaluation methods. Through standardized interfaces and unified data specifications, it achieves an integrated verification environment for cross-level models, improving the efficiency of collaborative verification of multi-source heterogeneous models. Simultaneously, by combining data processing and big data technologies, it constructs an automated test design and analysis evaluation toolchain, supporting extreme model testing under complex boundary conditions, significantly shortening the testing cycle and reducing operational risks. Furthermore, by establishing a model iterative optimization mechanism, a closed loop of "test-verification-feedback" is formed, providing continuous technical support for equipment digital upgrades and combat capability generation.

[0051] The implementation method proposed in this invention fills the technical gap in the full-process verification of equipment digital models. By constructing a standardized and automated verification platform, it solves the core pain points in current digital testing, such as insufficient model credibility and low verification efficiency, and ultimately provides fundamental support for digital collaboration and efficient testing and evaluation of equipment throughout its entire life cycle. Attached Figure Description

[0052] Figure 1 This is a basic structural diagram of the device of the present invention. Detailed Implementation

[0053] To better understand the content of this invention, an embodiment is provided here.

[0054] Figure 1 This is a basic structural diagram of the device of the present invention.

[0055] In a first aspect, the present invention discloses a verification device for an equipment digital model, comprising: a resource support subsystem, a service support subsystem, and a verification and evaluation subsystem;

[0056] The resource support subsystem is connected to the service support subsystem and is used for unified management of data, models, resources, systems, etc.

[0057] The service support subsystem is connected to the verification and evaluation subsystem. It is used to integrate multi-source heterogeneous equipment digital models according to the standard interface, and to perform joint simulation operation on the equipment digital models according to the received simulation requirements to obtain simulation results.

[0058] The verification and evaluation subsystem is used to generate test plan information by selecting diverse test elements according to specific test tasks, generate simulation requirements according to the test plan, send the simulation requirements to the service support subsystem, verify and evaluate the simulation operation results obtained by the service support subsystem, and obtain the verification results of the equipment digital model; the test plan information includes the expected results of the equipment digital model.

[0059] The resource support subsystem, service support subsystem, and verification and evaluation subsystem also include corresponding running servers, and all of them run on their respective running servers.

[0060] The data includes data collected by the data management module;

[0061] The models include various multi-source heterogeneous equipment digital models, as well as platform-level models, unit-level models, and related test models.

[0062] The system refers to a verification device equipped with a digital model.

[0063] The resource support subsystem includes a data management module, a model management module, a resource management module, and a system management module.

[0064] The data management module is used to collect, manage and process the test data required for the evaluation, and to store the collected test data in the database.

[0065] The process of generating test plan information by selecting diverse test elements based on specific test tasks involves selecting corresponding test elements that meet the requirements from a pre-set test element information database based on the specific test task's needs. Test plan information is then generated based on the test elements and the requirements. The requirements include the required technical indicators for each test element, test site requirements, and test environment setup requirements. The test elements include measuring radar equipment, communication equipment, data processing equipment, and a test mobile platform. The test environment setup requirements include electromagnetic interference environment settings, photoelectric interference environment settings, and meteorological environment requirements.

[0066] The process of generating simulation requirements based on the test plan involves taking the technical indicators of each test requirement in the test plan, as well as the test site requirements and test environment setup requirements, as simulation requirements.

[0067] The model management module is used to manage platform-level models, unit-level models, and related test models;

[0068] The resource management module is used to achieve unified management of various resources for the experiment; the various resources for the experiment include indicator system, indicator model, visualization template, report template, model operator, data file library, evaluation model library and microservice program;

[0069] The system management module is used to manage system logs, roles, users, and permissions for the verification device of the equipment digital model.

[0070] Each module of the resource support subsystem receives a data query request sent by the service support subsystem, retrieves the corresponding data based on the data query request, and sends it to the service support subsystem for joint simulation.

[0071] The service support subsystem includes a model packaging service module, a model assembly service module, a scenario editing service module, an evaluation index construction service module, an evaluation method service module, a node operation service module, a simulation operation support module, and a data acquisition service module.

[0072] The service support subsystem integrates multi-source heterogeneous equipment digital models according to standardized interfaces by connecting the multi-source heterogeneous equipment digital models to the interconnection middleware; the interconnection middleware connects the heterogeneous models to the simulation operation support module of the service support subsystem, and builds a transmission mechanism based on physical networks and IP protocols, using publish-subscribe and remote transmission methods as the basic means.

[0073] The interconnect middleware provides quality policies for different transmission requirements, including Best-Effort and reliable published subscription services, controls service quality through QoS policies, and supports dynamic QoS policies.

[0074] Besides its application in distributed parallel high-performance simulation, network interconnection is also the foundation for distributed interconnection of heterogeneous systems. Just as DIS focuses on interconnecting multiple simulators of different types located in different locations through a network to form a networked training capability, HLA focuses on distributed interactive simulation in a broader sense (including real-time and non-real-time, training and non-training), and TENA focuses on the experimental training field (primarily experimental in practical applications). This project, in applying heterogeneous model interconnection, inevitably involves interconnecting heterogeneous model simulation with constructed simulation, that is, forming an integrated cyber-physical system through interaction and collaboration between the simulation engine and the constructed model.

[0075] Heterogeneous interconnection generally consists of the following parts: heterogeneous model and engine, communication protocol and middleware (software), interaction protocol (data standard), and engine adaptation.

[0076] In distributed parallel high-performance simulation engine solutions, interconnection for homogeneous simulations also involves some of these components, including some networks (mainly wired and LANs), communication protocols and middleware, and interaction protocols. Therefore, these components can be reused. This platform focuses on designing a middleware for connecting heterogeneous models.

[0077] Interconnect middleware is an information superhighway connecting heterogeneous models with constructed simulations. It relies on physical networks and the IP protocol to construct a transmission mechanism based on publish-subscribe and remote methods. Interconnect middleware has the following main functions:

[0078] (1) Dynamic subscription based on topic

[0079] To improve communication quality and reduce network overhead, interconnect middleware should provide a topic-based dynamic publishing and subscription mechanism. Publishers can declare one or more topics, and subscribers can adjust subscription rules at any time during runtime to change the topics they are interested in. Topic data can be coarse-grained or fine-grained, subscribing to all data of a certain type of object model, data of a specific instance, or even down to a single attribute of an object. Furthermore, interconnect middleware should provide a comprehensive set of conditional statement parsing rules, supporting value-based filtering of object attributes, further refining the subscription rules.

[0080] (2) Dynamic QoS strategy

[0081] Interconnect middleware should provide quality policies tailored to different transmission needs, including best-effort and reliable subscription services. Service quality should be controlled through QoS policies, and dynamic QoS should be supported. Dynamic QoS allows subscribers to meet varying data quality requirements from the publisher at different times, controlling aspects such as transmission frequency, reliability, and latency. Dynamic QoS significantly reduces network overhead and is crucial for the overall availability of distributed systems.

[0082] (3) Real-time communication

[0083] The Real-Time Publish / Order (RTPS) protocol originated in industrial automation and is actually part of the IEC-PAS-62030 unit, an IEC-approved Real-Time Industrial Ethernet standard. This is a very mature technology currently used on tens of thousands of industrial devices worldwide. The Interconnect Middleware Line Protocol conforms to the latest RTPS-V2.25 standard and can seamlessly interface with other OMG-compliant DDS products.

[0084] (4) Flexible switching of transmission methods

[0085] The interconnect middleware uses a layered design, separating the upper-layer application from the lower-layer data transmission protocol. The application layer only needs to be concerned with the data content, not the protocol used for transmission. This allows for the use of different data transmission channels (such as RTPS, multicast, TCP, UDP, reflective memory networks, etc.) and flexible switching between them without modifying existing code, based on actual needs.

[0086] (5) Supports cross-local area network interconnection.

[0087] Interconnect middleware is suitable for wide area network (WAN) interconnection. In extreme cases, when the communicating parties are located in different local area networks (LANs), the interconnect middleware uses ICE (Interactive Connectivity Establishment) technology to achieve point-to-point interconnection across LANs, or uses the TURN (Traversal Using Relay NAT) alternative method for communication, ensuring that communication can be established 100%.

[0088] The middleware adopts a layered architecture, as shown in the diagram below, consisting of a data optimization layer, an interface layer, and a data management and communication layer. The data optimization layer primarily provides interfaces for convenient data searching and filtering, improving program efficiency and reducing network load. The interface layer mainly provides interfaces for data differentiation by domain, publish / subscribe, and corresponding services. The data management and communication layer primarily provides functions such as data caching, disaster recovery, and data transmission.

[0089] The verification and evaluation subsystem verifies and evaluates the simulation results obtained by the service support subsystem to obtain the verification results of the equipment digital model, including:

[0090] Static consistency analysis is performed on the simulation results obtained from the service support subsystem and the static results of the expected results of the equipment digital model in the test plan information to obtain the first verification result value; the simulation results are a data sequence of the equipment model state; the static and dynamic results of the expected results of the equipment digital model are both represented as the expected data sequence form of the equipment model state.

[0091] Dynamic consistency analysis is performed on the simulation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain a second verification result value.

[0092] The first and second verification result values ​​are fused and estimated to obtain the fused verification result value.

[0093] Determine whether the fusion verification result value is greater than a preset discrimination threshold value to obtain a first discrimination result;

[0094] If the first discrimination result is greater than, the verification result of the equipment digital model is determined to be passed; if the first discrimination result is not greater than, the verification result of the equipment digital model is determined to be failed.

[0095] The static consistency analysis includes:

[0096] The data sequence of each parameter of the equipment model state in the simulation results is represented as the corresponding state row vector;

[0097] The expected data sequence of each parameter of the equipment model state in the static results is represented as the corresponding standard row vector;

[0098] Subtract the state row vector from the standard row vector for each parameter to obtain the corresponding difference row vector;

[0099] By using the difference row vectors of all parameters, a difference matrix is ​​constructed.

[0100] The cross-correlation value is calculated on the difference matrix to obtain the cross-correlation matrix;

[0101] The cross-correlation matrix is ​​subjected to eigenvalue decomposition to obtain an eigenvector; the eigenvector is a vector composed of all eigenvalues.

[0102] For each row vector of the difference matrix, statistical features are calculated to obtain the corresponding statistical feature values;

[0103] The expression for calculating the statistical features is:

[0104] ,

[0105] in, Let be the statistical characteristic value of the i-th row. and These are the elements in the i-th row and i-th column of the difference matrix, and the elements in the i-th row and j-th column, respectively, where N is the column dimension of the difference matrix. The maximum value of the vector in the i-th row of the difference matrix;

[0106] The statistical eigenvalues ​​corresponding to each row vector of the difference matrix are multiplied together, and then the results of multiplication of all row vectors are summed to obtain the first verification result value.

[0107] The expression for the fusion estimation is:

[0108] ,

[0109] in, and The first and second verification result values ​​obtained from the i-th verification evaluation are respectively: and The preset weighting values, Let M be the mean of all second verification results, and M be the number of verification evaluations. This is the result value of the fusion verification.

[0110] The step of performing cross-correlation calculation on the difference matrix to obtain a cross-correlation matrix includes: performing cross-correlation calculation on the row vectors of the difference matrix to obtain cross-correlation values, and constructing a cross-correlation matrix using the cross-correlation values; the elements of the i-th row and j-th column of the cross-correlation matrix are obtained by performing cross-correlation operation on the i-th row vector and j-th row vector of the difference matrix.

[0111] Static data consistency analysis methods primarily focus on evaluating the accuracy of static model design and source code. These methods do not require generating an executable application of the model; instead, they infer its execution results through imagination. Static data consistency analysis methods have found widespread application with the help of automated tools that can assist the V&V process. Using static data consistency analysis methods, a range of information can be obtained regarding the model framework, modeling techniques and implementation, and the model's data flow and control flow. Static data consistency analysis methods mainly include direct statistical methods, parametric hypothesis testing (classical hypothesis testing, Bayesian hypothesis testing, etc.), and nonparametric testing methods (rank-sum test, runs test, etc.).

[0112] Dynamic consistency analysis is a method used to evaluate whether different components, variables, or behaviors in a system or model can maintain consistency over time. It is widely used in economics, control theory, computer science, systems engineering, and other fields, especially when dealing with time series, decision processes, or dynamic systems. In this system's multi-level model validation application, Pearson correlation coefficient method and grey relational analysis were primarily used.

[0113] The dynamic consistency analysis involves calculating the second verification result value by using the Pearson correlation coefficient method or grey relational analysis method on the data sequence of the simulation results and the data sequence of the dynamic results.

[0114] Grey relational analysis treats the factor values ​​of the research object and influencing factors as points on a line, compares them with the curves plotted by the factor values ​​of the object to be identified and its influencing factors, compares the degree of closeness between them and quantifies them separately, calculates the degree of correlation between the research object and each influencing factor of the object to be identified, and judges the degree of influence of the object to be identified on the research object by comparing the magnitude of each correlation. The operation steps are as follows;

[0115] Step 1: Determine the reference sequence as the actual experimental data, denoted as . Compare the simulation experimental data of the sequence bit, denoted as .

[0116] Step 2: Normalize the sequence, such as by centering or standardizing, and select appropriate preprocessing methods based on the actual situation.

[0117] Step 3: Calculate the grey relational coefficient

[0118]

[0119] The value is in (0,1), and the smaller the value, the greater the discrimination. It is usually taken as 0.5.

[0120] The basic idea of ​​grey relational analysis is to judge the closeness of the relationship between sequences based on the similarity of their geometric shapes. The closer the curves are, the greater the correlation between the corresponding sequences, and vice versa.

[0121] The service support subsystem includes a model encapsulation service module, a model assembly service module, a scenario editing service module, an evaluation index construction service module, an evaluation method service module, a node operation service module, a simulation operation support module, and a data acquisition service module. The composition and implementation of each module are described in detail below.

[0122] The model encapsulation service module is responsible for standardizing and encapsulating equipment digital models from different manufacturers, formats, and precision levels, thus solving the problem of "inconsistent model standards." It primarily implements model interface standardization, metadata standardization, model verification, and error diagnosis functions.

[0123] The model encapsulation service module adopts an encapsulation framework based on the FMI (Functional Mock-up Interface) 2.0 standard, supporting import and export of FMU (Functional Mock-up Unit) format. It features a multi-layered adapter architecture: the bottom layer is a protocol conversion layer (supporting protocols such as HLA, DIS, and TENA), the middle layer is a model parsing layer (supporting tool models such as Simulink, AMESim, and Modelica), and the top layer is a standardized interface layer. It integrates a lightweight model algorithm to automatically simplify complex models and supports multi-level LOD (Level of Detail) precision management.

[0124] The model assembly service module enables flexible assembly and integration of multi-level (system-level, platform-level, unit-level) equipment digital models, addressing the lack of a "data-physical integration mechanism." It primarily provides functions for defining model connection relationships, parameter mapping, coupling coordination, and assembly verification. Semantic matching technology (based on an equipment domain ontology library) facilitates automatic parameter association and unit conversion between different models. A loosely coupled architecture is adopted: asynchronous communication between models is achieved through message middleware (such as RabbitMQ), supporting distributed deployment. Petri net theory is used for assembly logic verification to ensure the correctness and completeness of the assembly structure.

[0125] The scenario editing service module provides a visual simulation scenario editing environment, supporting the construction and management of complex test scenarios and addressing the issue of insufficient test design methods. It primarily implements functions for test scenario definition, environmental parameter configuration, event sequence arrangement, and scenario version management. It employs a WebGL-based 3D visual editor, supporting drag-and-drop construction of test scenarios; it includes pre-set templates for typical equipment test scenarios (such as combat environments, failure modes, and extreme conditions), and supports parameterized customization of these templates. Event sequences are arranged using Gantt graphs, supporting three modes: condition-triggered, timed-triggered, and external event-triggered.

[0126] The evaluation index construction service module constructs a multi-dimensional evaluation index system based on relevant standards, enabling dynamic definition, weight allocation, and correlation analysis of indicators. It supports the construction of multi-level index trees, automatically calculates index weights, and performs consistency checks; it pre-sets evaluation index templates for dimensions such as equipment performance, behavioral characteristics, and response characteristics; it uses OWL (WebOntology Language) to construct an equipment evaluation ontology model, achieving semantic description of indicators; it automatically optimizes index weights based on historical evaluation data and uses fuzzy C-means clustering for index importance analysis; and it calculates the correlation between indicators using Pearson correlation coefficients and mutual information to avoid index redundancy.

[0127] The evaluation method service module provides a variety of model confidence evaluation algorithms, enabling quantitative comparison and analysis of simulation results and measured data, and supporting flexible selection and combination of evaluation methods. It integrates multiple evaluation algorithm libraries, including time-domain analysis: RMSE (Root Mean Square Error), MAE (Mean Absolute Error), NRMSE (Normalized Root Mean Square Error); frequency-domain analysis: FFT (Fast Fourier Transform), coherence analysis; feature extraction: DTW (Dynamic Time Warping), Hausdorff distance; and statistical analysis: KS test, t-test, and confidence interval analysis.

[0128] The node operation service module is responsible for the management and task scheduling of computing nodes in the distributed simulation environment, ensuring the efficient execution of large-scale simulation tasks and addressing the high computational resource requirements of complex equipment system simulations. A Kubernetes-based container orchestration system enables automatic discovery, health monitoring, and elastic scaling of computing nodes. It dynamically adjusts task allocation by monitoring CPU, memory, and network metrics in real time; and employs checkpoint technology to save task state and support fault recovery.

[0129] The simulation operation support module provides basic service support required for simulation operation, including core functions such as time management, data synchronization, and event handling, to ensure the coordination and consistency of multi-model joint simulation.

[0130] The data acquisition service module is responsible for the real-time acquisition, preprocessing, and storage of multi-source heterogeneous data during simulation, providing a data foundation for subsequent verification and evaluation and addressing the lack of a data-real integration mechanism. A layered data acquisition architecture is adopted, comprising an acquisition layer, a transmission layer, and a processing layer. The acquisition layer supports access from various data sources (model output, sensor data, databases, etc.). The transmission layer uses the MQTT protocol for efficient data transmission and supports QoS level configuration. The processing layer performs real-time data cleaning, format conversion, and anomaly detection.

[0131] A second aspect of this invention discloses a method for verifying a digital model of equipment, implemented using the aforementioned verification device for the digital model of equipment, comprising:

[0132] Using the aforementioned verification and evaluation subsystem, based on the specific experimental task, diverse experimental elements are selected to generate experimental plan information. Simulation requirements are then generated based on the experimental plan and sent to the service support subsystem.

[0133] Using the service support subsystem, the equipment digital model is jointly simulated based on the received simulation requirements to obtain simulation results.

[0134] Using the aforementioned verification and evaluation subsystem, the simulation operation results obtained from the service support subsystem are verified and evaluated to obtain the verification results of the equipment digital model;

[0135] The process of using the verification and evaluation subsystem to verify and evaluate the simulation results obtained from the service support subsystem to obtain the verification results of the equipment digital model includes:

[0136] Static consistency analysis is performed on the static results of the simulation operation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain the first verification result value;

[0137] Dynamic consistency analysis is performed on the simulation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain a second verification result value.

[0138] The first and second verification result values ​​are fused and estimated to obtain the fused verification result value.

[0139] Determine whether the fusion verification result value is greater than a preset discrimination threshold value to obtain a first discrimination result;

[0140] If the first discrimination result is greater than, the verification result of the equipment digital model is determined to be passed; if the first discrimination result is not greater than, the verification result of the equipment digital model is determined to be failed.

[0141] The static consistency analysis includes:

[0142] The data sequence of each parameter of the equipment model state in the simulation results is represented as the corresponding state row vector;

[0143] The expected data sequence of each parameter of the equipment model state in the static results is represented as the corresponding standard row vector;

[0144] Subtract the state row vector from the standard row vector for each parameter to obtain the corresponding difference row vector;

[0145] By using the difference row vectors of all parameters, a difference matrix is ​​constructed.

[0146] The cross-correlation value is calculated on the difference matrix to obtain the cross-correlation matrix;

[0147] The cross-correlation matrix is ​​subjected to eigenvalue decomposition to obtain an eigenvector; the eigenvector is a vector composed of all eigenvalues.

[0148] For each row vector of the difference matrix, statistical features are calculated to obtain the corresponding statistical feature values;

[0149] The expression for calculating the statistical features is:

[0150] ,

[0151] in, Let be the statistical characteristic value of the i-th row. and These are the elements in the i-th row and i-th column of the difference matrix, and the elements in the i-th row and j-th column, respectively, where N is the column dimension of the difference matrix. The maximum value of the vector in the i-th row of the difference matrix;

[0152] The statistical eigenvalues ​​corresponding to each row vector of the difference matrix are multiplied together, and then the results of multiplication of all row vectors are summed to obtain the first verification result value.

[0153] The expression for the fusion estimation is:

[0154] ,

[0155] in, and The first and second verification result values ​​obtained from the i-th verification evaluation are respectively: and These are the preset weighting values. Let M be the mean of all second verification results, and M be the number of verification evaluations. This is the result value of the fusion verification.

[0156] The step of performing cross-correlation calculation on the difference matrix to obtain a cross-correlation matrix includes: performing cross-correlation calculation on the row vectors of the difference matrix to obtain cross-correlation values, and constructing a cross-correlation matrix using the cross-correlation values; the elements of the i-th row and j-th column of the cross-correlation matrix are obtained by performing cross-correlation operation on the i-th row vector and j-th row vector of the difference matrix.

[0157] In all embodiments of the present invention, the variables involved in all computational expressions or mathematical functions have been dimensionlessized before computation.

[0158] In all embodiments of the present invention, the values ​​of the independent variables in the input of all computational expressions or mathematical functions meet the reasonable requirements of the input range of the computational expressions or mathematical functions, and can ensure that the computational expressions or mathematical functions can be calculated smoothly without violating physical laws or mathematical rules.

[0159] A third aspect of the present invention discloses a verification device for an equipment digital model, the device comprising:

[0160] Memory containing executable program code;

[0161] A processor coupled to the memory;

[0162] The processor calls the executable program code stored in the memory to execute the verification method of the equipment digital model.

[0163] In a fourth aspect of this invention, a computer-readable storage medium is disclosed, the computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the verification method for the equipment digital model.

[0164] In a fifth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the verification method for the equipment digital model.

[0165] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A verification device equipped with a digital model, characterized in that, include: Resource support subsystem, service support subsystem, and verification and evaluation subsystem; The resource support subsystem is connected to the service support subsystem and is used for unified management of data, models, resources, and systems. The service support subsystem is connected to the verification and evaluation subsystem and is used to integrate multi-source heterogeneous equipment digital models according to the standard interface, perform joint simulation operation on the equipment digital models according to the received simulation requirements, and obtain simulation operation results. The verification and evaluation subsystem is used to generate test plan information by selecting diverse test elements according to specific test tasks, generate simulation requirements based on the test plans, send the simulation requirements to the service support subsystem, and verify and evaluate the simulation results obtained by the service support subsystem to obtain the verification results of the equipment digital model, including: Static consistency analysis is performed on the simulation results obtained from the service support subsystem and the static results of the expected results of the equipment digital model in the test plan information to obtain the first verification result value; the simulation results are a data sequence of the equipment model state; the static and dynamic results of the expected results of the equipment digital model are both represented as the expected data sequence form of the equipment model state. Dynamic consistency analysis is performed on the simulation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain a second verification result value. The first verification result value and the second verification result value are fused and estimated to obtain a fused verification result value; the expression for the fused estimation is: ， in, and The first and second verification result values ​​obtained from the i-th verification evaluation are respectively: and The preset weighting values, Let M be the mean of all second verification results, and M be the number of verification evaluations. This is the value of the fusion verification result; Determine whether the fusion verification result value is greater than a preset discrimination threshold value to obtain a first discrimination result; If the first discrimination result is greater than, the verification result of the equipment digital model is determined to be passed; if the first discrimination result is not greater than, the verification result of the equipment digital model is determined to be failed. The test plan information includes the expected results of the equipment digital model.

2. The verification device for the equipment digital model as described in claim 1, characterized in that, The resource support subsystem includes a data management module, a model management module, a resource management module, and a system management module. The data management module is used to collect, manage and process the test data required for the evaluation, and to store the collected test data in the database. The model management module is used to manage platform-level models, unit-level models, and related test models; The resource management module is used to achieve unified management of various resources for the experiment; the various resources for the experiment include indicator system, indicator model, visualization template, report template, model operator, data file library, evaluation model library and microservice program; The system management module is used to manage system logs, roles, users, and permissions for the verification device of the equipment digital model.

3. The verification device for the equipment digital model as described in claim 1, characterized in that, The service support subsystem integrates multi-source heterogeneous equipment digital models according to standardized interfaces by connecting the multi-source heterogeneous equipment digital models to the interconnection middleware. The interconnection middleware connects the heterogeneous models to the simulation operation support module of the service support subsystem, and builds a transmission mechanism based on physical networks and IP protocols, using publish-subscribe and remote transmission methods as the basic means.

4. The verification device for the equipment digital model as described in claim 1, characterized in that, The static consistency analysis includes: The data sequence of each parameter of the equipment model state in the simulation results is represented as the corresponding state row vector; The expected data sequence of each parameter of the equipment model state in the static results is represented as the corresponding standard row vector; Subtract the state row vector from the standard row vector for each parameter to obtain the corresponding difference row vector; The difference matrix is ​​constructed by using the difference row vectors of all parameters as row vectors; The cross-correlation value is calculated on the difference matrix to obtain the cross-correlation matrix; The cross-correlation matrix is ​​subjected to eigenvalue decomposition to obtain an eigenvector; the eigenvector is a vector composed of all eigenvalues. For each row vector of the difference matrix, statistical features are calculated to obtain the corresponding statistical feature values; The expression for calculating the statistical features is: ， in, Let be the statistical characteristic value of the i-th row. and These are the elements in the i-th row and i-th column of the difference matrix, and the elements in the i-th row and j-th column, respectively, where N is the column dimension of the difference matrix. The maximum value of the vector in the i-th row of the difference matrix; Multiply the statistical eigenvalues ​​corresponding to each row vector of the difference matrix to obtain the multiplication result of each row vector; then sum the multiplication results of all row vectors to obtain the first verification result value.

5. A method for verifying a digital model of equipment, characterized in that, This is achieved using the verification device for the equipment digital model as described in any one of claims 1 to 4, comprising: S1. Using the verification and evaluation subsystem, based on the specific test task, test plan information is generated by selecting diverse test elements, simulation requirements are generated based on the test plan, and the simulation requirements are sent to the service support subsystem. S2, using the service support subsystem, perform joint simulation of the equipment digital model according to the received simulation requirements, and obtain the simulation results; S3. Using the verification and evaluation subsystem, the simulation operation results obtained by the service support subsystem are verified and evaluated to obtain the verification results of the equipment digital model.

6. The verification method for the equipment digital model as described in claim 5, characterized in that, The process of using the verification and evaluation subsystem to verify and evaluate the simulation results obtained from the service support subsystem to obtain the verification results of the equipment digital model includes: Static consistency analysis is performed on the static results of the simulation operation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain the first verification result value; Dynamic consistency analysis is performed on the simulation results obtained from the service support subsystem and the expected results of the equipment digital model in the test plan information to obtain a second verification result value. The first and second verification result values ​​are fused and estimated to obtain the fused verification result value. Determine whether the fusion verification result value is greater than a preset discrimination threshold value to obtain a first discrimination result; If the first discrimination result is greater than, the verification result of the equipment digital model is determined to be passed; if the first discrimination result is not greater than, the verification result of the equipment digital model is determined to be failed.

7. A verification device equipped with a digital model, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the verification method of the equipment digital model as described in any one of claims 5 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the verification method for the equipment digital model as described in any one of claims 5 to 6.

9. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the verification method for the equipment digital model as described in any one of claims 5 to 6.

Images