Digital-real fusion test evaluation method and device based on digital test verification reference architecture

Abstract

The invention discloses a digital-real fusion test evaluation method and device based on a digital test verification reference architecture, and the method comprises the steps: carrying out the preliminary digital deduction in a digital domain based on physical domain data; a digital domain evaluation method is constructed, and iterative optimization of the digital model is completed; physical domain verification is carried out, and a comparison evaluation method of digital domain and physical domain results is established; and combining physical domain data and digital domain evaluation indexes, continuously iteratively optimizing until actual requirements are met, and then applying to real installation. According to the method, dynamic calibration and fusion verification between a physical domain and a digital domain are realized by constructing a digital-real fusion iterative closed-loop evaluation process, and the adaptability and credibility of a digital test system under complex working conditions are remarkably improved. The method has the advantages of high efficiency, low cost and good expansibility, provides a feasible number-real fusion test evaluation means for performance evaluation and optimization design of a complex system, and has a wide engineering application prospect.

Description

Technical Field

[0001] This invention belongs to the fields of electronic engineering and computer science, and specifically relates to a digital-real fusion test evaluation method and device based on a digital test and verification reference architecture. Background Technology

[0002] In the design and verification of complex systems, traditional physical experiments are costly and time-consuming; for example, wind tunnel testing in the aerospace field often requires significant financial and time investment. Digital simulation suffers from large model errors and difficulty in dynamic correction. For instance, in finite element analysis of stress and strain on aircraft wings, the model assumptions and parameters have significant uncertainties, resulting in a clear deviation between simulation results and the real system. Current data-real fusion methods often employ a linear calibration approach of "simulation first, experiment later," correcting model parameters or establishing empirical correction coefficients using experimental data. However, this method suffers from feedback lag and low data utilization, leading to a lack of effective collaboration between data and experiment, making it difficult to support high-reliability performance evaluation. With the development of digital twins and next-generation information technologies, there is an urgent need to establish a data-real fusion evaluation mechanism with dynamic feedback capabilities to achieve iterative optimization and verification closed-loop of digital experiments. Summary of the Invention

[0003] This invention provides a digital-physical fusion test evaluation method and apparatus based on a digital test and verification reference architecture. This method achieves dynamic calibration and fusion verification between the physical and digital domains by constructing an iterative closed-loop evaluation process for digital-physical fusion, thereby improving the adaptability and reliability of the digital test system under complex operating conditions. It provides an efficient, low-cost, and highly scalable digital-physical fusion test evaluation method, which is suitable for the performance evaluation and optimization design of complex systems and has broad engineering application prospects.

[0004] Technical solution of the present invention:

[0005] A digital-real fusion test evaluation method based on a digital test verification reference architecture includes the following steps:

[0006] Step S110: Based on the physical domain data, conduct preliminary digital simulation in the digital domain;

[0007] Step S120: Construct a digital domain evaluation method and complete the iterative optimization of the digital model;

[0008] Step S130: Conduct physical domain verification and construct a comparative evaluation method for digital domain and physical domain results;

[0009] Step S140: Combine physical domain data with digital domain evaluation functions, continuously iterate and optimize until the actual requirements are met, and then apply it to real-world installations.

[0010] A digital-real fusion test evaluation device based on a digital test verification reference architecture includes:

[0011] The digital simulation module is used to conduct preliminary digital simulations in the digital domain based on physical domain data.

[0012] The digital model iterative optimization module is used to construct digital domain evaluation methods and complete the iterative optimization of digital models.

[0013] The comparison and evaluation module is used to conduct physical domain verification and construct a comparison and evaluation method for digital domain and physical domain results;

[0014] The iterative optimization module combines physical domain data with digital domain evaluation functions to continuously iterate and optimize until it meets actual requirements before being applied to real-world implementations.

[0015] A computing device includes: at least one processor and a memory storing program instructions; when the program instructions are read and executed by the processor, the computing device performs the method.

[0016] A readable storage medium storing program instructions that, when read and executed by a computing device, cause the computing device to perform the method.

[0017] A computer program product includes a computer program that, when executed by a processor, implements the method.

[0018] Beneficial effects:

[0019] Compared with existing technologies, this invention constructs an iterative closed-loop evaluation process that integrates physical and digital domains, achieving dynamic calibration and fusion verification between the physical and digital domains. This significantly improves the adaptability and reliability of digital test systems under complex operating conditions. The method is highly efficient, low-cost, and has good scalability, providing a practical and feasible method for performance evaluation and optimization design of complex systems, and has broad engineering application prospects. Attached Figure Description

[0020] Figure 1 This is a flowchart of a data-real fusion test evaluation method based on a digital test verification reference architecture according to the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above objectives, this invention adopts the following technical solution.

[0022] This invention provides a data-real domain fusion test evaluation method based on a digital experimental testing and verification reference architecture. The method comprises four parts: preliminary digital deduction, digital domain deduction evaluation, physical domain verification evaluation, and continuous iterative optimization. By constructing an iterative closed-loop evaluation process for data-real domain fusion, dynamic calibration and fusion verification between the physical and digital domains are achieved, significantly improving the adaptability and reliability of the digital test system under complex operating conditions. This method possesses high efficiency, low cost, and good scalability, providing a practical data-real domain fusion test evaluation tool for the performance evaluation and optimization design of complex systems, and has broad engineering application prospects.

[0023] Figure 1 This is a flowchart illustrating a data-real domain fusion experimental evaluation method based on a digital experimental test and verification reference architecture according to the present invention. In the diagram, reference numeral 1 represents preliminary digital simulation, reference numeral 2 represents digital domain simulation evaluation, reference numeral 3 represents physical domain verification evaluation, and reference numeral 4 represents continuous iterative optimization. Figure 1 As shown, the method includes:

[0024] Step S110: Based on the physical domain data, conduct preliminary digital simulation in the digital domain. The specific steps are as follows:

[0025] S110-1, collects relevant physical domain data based on physical objects, physical scenes, and physical tasks in the physical domain.

[0026] S110-2: Based on the physical domain data collected in step S110-1, complete the construction of a high-fidelity digital model in the digital domain, and based on the experimental type, provide a digital experimental plan, conduct digital experimental testing and verification, and complete the preliminary digital simulation.

[0027] Step S120: Construct a digital domain evaluation method and complete the iterative optimization of the digital model. The specific steps are as follows:

[0028] S120-1, Construct a digital domain evaluation function based on the digital domain derivation results.

[0029] For the geometric layer of the digital model, two factors can be identified: dimensions and assembly relationships. Each factor has a corresponding function to evaluate the accuracy of the digital domain derivation results.

[0030] The function constructed for the size factor is as follows:

[0031] ,

[0032] in, Let be the size factor evaluation function for the i-th experiment, and N be the number of modules in the digital model. Let n be the weight of a certain module n in the overall digital model in the i-th trial. Let n be the size of module n in the digital domain during the i-th trial. Let be the actual size of the physical domain of module n in the i-th experiment.

[0033] The function for the assembly relation factor is constructed as follows:

[0034] ,

[0035] in, Let R be the evaluation function for the assembly relationship factors in the i-th experiment, and R be the total number of assembly relationships. Let be the weight of a certain assembly relationship t in the total assembly relationships in the i-th trial. Let t be the error value generated by the digital domain assembly relationship t in the i-th experiment. Let t be the actual error value of the physical domain assembly relationship t in the i-th experiment.

[0036] For the simulation layer of the digital model, it can be divided into two factors: physical information and data information. Each factor has a corresponding function set to evaluate the accuracy of the digital domain inference results.

[0037] The function for the physical information factor is constructed as follows:

[0038] ,

[0039] in, Let M be the evaluation function for the physical information factor in the i-th experiment, and M be the number of relevant physical information categories. Let be the weight of a certain physical information 'e' in the total physical information during the i-th experiment. Let be the error value generated by the physical information e in the digital domain during the i-th trial. Let be the actual error value of the physical information e in the physical domain during the i-th experiment.

[0040] The function for the data information factor is constructed as follows:

[0041] ,

[0042] in, Let T be the evaluation function for the information factors of the data from the i-th experiment, and let T be the total number of relevant data information categories. Let d be the weight of a certain data information in the total data information in the i-th experiment. Let be the predicted value generated by the data information d in the digital domain in the i-th trial. Let d be the true value of the data information d within the physical domain in the i-th experiment.

[0043] Summing all the factor weights mentioned above, we can obtain the numerical domain evaluation function for the i-th trial:

[0044] ,

[0045] in, Let i be the digital domain evaluation function for the i-th trial. These are the weight values ​​for the size factor. The assembly relationship factor weight value. The weight value of the physical information factor. This represents the weight value of the data information factor.

[0046] S120-2, In the (i+1)th round of trials in the digital domain, the digital domain evaluation functions obtained from the i-th and (i+1)th trials are calculated using the following formula:

[0047] ,

[0048] in, It is a digital domain experimental iterative evaluation function.

[0049] Compare the obtained result with the threshold value of the digital field. Make a judgment, among which The range of values ​​for is (0, 1), if the following conditions are met:

[0050] ,

[0051] The physical domain is then verified through digital domain testing. If the threshold requirements are not met, iterative optimization in the digital domain continues until the requirements are met.

[0052] Step S130: Conduct physical domain verification and construct a method for comparing and evaluating digital and physical domain results. The specific steps are as follows:

[0053] S130-1 uses methods such as linear sampling, nonlinear sampling, hierarchical random sampling, and importance sampling to sample data in digital experiments during the experiment. Based on the sampled data, physical verification is carried out in different experimental environments and schemes to obtain physical domain verification data.

[0054] S130-2, Construct an evaluation function, and use the evaluation function to compare and evaluate the data obtained in step S130-1 and the results of the numerical domain derivation in step S120, as shown below:

[0055] ,

[0056] in, It is an evaluation function under different experimental environments, where d represents the experimental scheme and j represents the experimental environment. For the physical domain sampled value of scheme d under environment j, Let be the predicted value of scheme d in the digital domain under environment j.

[0057] The obtained results are compared with the physical domain's pass threshold. Make a judgment, among which The value range of is (0, 1), and if the following is satisfied for each test environment and scheme:

[0058] ,

[0059] Then, a physical domain test is performed to apply the technology to the real domain. If the threshold requirement is not met, proceed to step S140.

[0060] Step S140 involves combining physical domain data with digital domain evaluation functions, continuously iterating and optimizing until the actual requirements are met, and then applying it to real-world installations. The specific steps are as follows:

[0061] S140-1, add the physical domain verification data from step S130 to the data collected in step S110-1 to complete the data supplementation, and restart the iterative optimization from step S110.

[0062] S140-2, when proceeding to step S120, the digital domain evaluation function is updated to complete the digital domain iterative optimization. The specific function update is as follows:

[0063] ,

[0064] in, Let i be the digital domain evaluation function for the i-th trial. These are the weight values ​​for the size factor. The assembly relationship factor weight value. The weight value of the physical information factor. This represents the weight value of the data information factor. This represents the weighting factor for the difference between the physical and digital domains. P represents the total number of schemes under different experimental environments in the physical domain validation.

[0065] In summary, this invention discloses a data-real domain fusion test evaluation method based on a digital experimental testing and verification reference architecture. This method comprises four parts: preliminary digital deduction, digital domain deduction evaluation, physical domain verification evaluation, and continuous iterative optimization. By constructing an iterative closed-loop evaluation process for data-real domain fusion, it achieves dynamic calibration and fusion verification between the physical and digital domains, significantly improving the adaptability and reliability of the digital test system under complex operating conditions. This method possesses high efficiency, low cost, and good scalability, providing a practical data-real domain fusion test evaluation tool for the performance evaluation and optimization design of complex systems, and has broad engineering application prospects.

[0066] The present invention also provides a digital-real fusion test evaluation device based on a digital test verification reference architecture, for performing the above methods, the device comprising:

[0067] The digital simulation module is used to conduct preliminary digital simulations in the digital domain based on physical domain data.

[0068] The digital model iterative optimization module is used to construct digital domain evaluation methods and complete the iterative optimization of digital models.

[0069] The comparison and evaluation module is used to conduct physical domain verification and construct a comparison and evaluation method for digital domain and physical domain results;

[0070] The iterative optimization module combines physical domain data with digital domain evaluation functions to continuously iterate and optimize until it meets actual requirements before being applied to real-world implementations.

[0071] A computing device includes: at least one processor and a memory storing program instructions; when the program instructions are read and executed by the processor, the computing device performs the method.

[0072] A readable storage medium storing program instructions that, when read and executed by a computing device, cause the computing device to perform the method.

[0073] A computer program product includes a computer program that, when executed by a processor, implements the method.

[0074] Contents not described in detail in this specification are prior art known to those skilled in the art. The above descriptions are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A digital-real fusion test evaluation method for verifying a reference architecture based on digital test testing, characterized in that, Comprising the following steps: Step S110, based on physical domain data, preliminary digital deduction in digital domain is carried out; Step S120, a digital domain evaluation method is constructed, and iterative optimization of the digital model is completed; Step S130, physical domain verification is carried out, and a comparison evaluation method of the results of the digital domain and the physical domain is constructed; Step S140, combined with the physical domain data and the digital domain evaluation function, iterative optimization is continuously carried out until the actual requirements are met, and then it is applied to the real implementation.

2. The digital-real fusion test evaluation method based on the digital test verification reference architecture according to claim 1, characterized in that, Step S110 includes: Step S110-1, based on the physical objects, physical scenes and physical tasks of the physical domain, the relevant physical domain data is collected; Step S110-2, according to the physical data collected in step S110-1, high-fidelity digital model construction is completed in the digital domain, and a digital test scheme is given based on the test type, digital test verification is carried out, and preliminary digital deduction is completed.

3. The digital-real fusion test evaluation method based on the digital test verification reference architecture according to claim 2, characterized in that, Step S120 includes: Step S120-1, a digital domain evaluation function is constructed for the digital domain deduction result; Step S120-2, the i th and i + 1 th test digital domain evaluation functions obtained are calculated through the following formula: ， wherein, is a digital domain trial iteration evaluation function; The obtained result is compared with a threshold value in the digital domain wherein the value range of (0, 1) is satisfied: ， Then, through the digital domain test, the physical domain verification is carried out; if the threshold requirement is not met, the digital domain iterative optimization is continued until the requirement is met.

4. The digital-real fusion test evaluation method based on the digital test verification reference architecture according to claim 3, characterized in that, S120-1 includes: For the geometric layer of the digital model, it is divided into two factors of size and assembly relationship; For the size factor, the function is constructed as follows: ， wherein, is the i-th trial size factor evaluation function, N is the number of modules of the digital model, is the proportion weight of a module n in the i-th trial in the total digital model, is the size of the module n in the digital domain in the i-th trial, is the real size of the module n in the physical domain in the i-th trial; For the assembly relationship factor, the function is constructed as follows: ， wherein, R is the total number of assembly relations, is the proportion weight of the assembly relation t in the total assembly relations in the i-th test, is the error value of the digital domain assembly relation t in the i-th test, is the real error value of the physical domain assembly relation t in the i-th test. For the simulation layer of the digital model, it is divided into two factors of physical information and data information; for the physical information factor, the function is constructed as follows: ， wherein, is the i-th test physical information factor evaluation function, M is the number of related physical information categories, is the proportion weight of a certain physical information e in the total physical information in the i-th test, is the error value of the physical information e generated in the digital domain in the i-th test, is the true error value of the physical information e in the physical domain in the i-th test; For the data information factor, the function is constructed as follows: ， wherein, is the i-th test data information factor evaluation function, T is the total number of relevant data information categories, is the proportion weight of a certain data information d in the total data information in the i-th test, is the predicted value of the data information d in the digital domain in the i-th test, is the true value of the data information d in the physical domain in the i-th test; The weights of all the above factors are added to obtain the digital domain evaluation function of the i th test: ， wherein, is a numerical domain evaluation function for the i-th trial, is a weight value for the size factor, is a weight value for the fit relationship factor, is a weight value for the physical information factor, is a weight value for the data information factor.

5. The digital-real fusion test evaluation method based on the digital test validation reference architecture according to claim 3, characterized in that, Step S130 includes: Step S130-1, in the test process, linear sampling, nonlinear sampling, hierarchical random sampling and importance sampling are carried out in the digital test, physical verification is carried out based on the sampling samples in different test environments and schemes, and physical domain verification data is obtained; Step S130-2, an evaluation function is constructed, and the data obtained in step S130-1 and the results of the digital domain deduction in step S120 are compared and evaluated as follows: ， wherein, is an evaluation function under different test environments, d represents a test scheme, j represents a test environment, is a physical domain actual sampling value of scheme d under j environment, is a digital domain internal prediction value of scheme d under j environment; The resulting results are compared with the pass threshold of the physical domain A decision is made, wherein The value range of is (0, 1), and if the following is satisfied for each test environment and scheme: ， Then, through the physical domain test, the real domain application is carried out; if the threshold requirement is not met, step S140 is carried out.

6. The digital-real fusion test evaluation method based on the digital test verification reference architecture according to claim 5, characterized in that, Step S140 includes: Step S140-1, the physical domain verification data in step S130 is added to the data collected in step S110-1, the data is supplemented, and the iterative optimization is restarted from step S110; Step S140-2, when step S120 is reached, the digital domain evaluation function is updated, and the iterative optimization of the digital domain is completed, and the specific function update is as follows: ， wherein, is the digital domain evaluation function for the i-th trial, is the weight value of the size factor, is the weight value of the assembly relationship factor, is the weight value of the physical information factor, is the weight value of the data information factor, is the weight value of the difference factor between the physical domain and the digital domain, and P is the total number of schemes under different trial environments verified in the physical domain.

7. A digital-real fusion test evaluation device for verifying a reference architecture based on digital test tests, characterized by, Comprise: A digital deduction module for carrying out preliminary digital deduction in a digital domain based on physical domain data; A digital model iterative optimization module for constructing a digital domain evaluation method and completing iterative optimization of the digital model; A comparison evaluation module for carrying out physical domain verification and constructing a comparison evaluation method of the results of the digital domain and the physical domain; An iterative optimization module for combining the physical domain data with the digital domain evaluation function, continuously iterating until the real requirements are met, and then applied to the real implementation.

8. A computing device, comprising: Comprising: at least one processor and memory storing program instructions; when the program instructions are read and executed by the processor, causing the computing device to perform the method of any of claims 1-6.

9. A readable storage medium storing program instructions, characterized in that, when the program instructions are read and executed by the computing device, causing the computing device to perform the method of any of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, the computer program, when executed by the processor, implements the method of any of claims 1-6.

Images