A high-performance virtual simulation method and system driven by digital twin data model
Through the high-performance virtual simulation method driven by the digital twin data model, the virtual simulation model is constructed using sensing and mechanism simulation data, which solves the problem of obtaining virtual experimental data in the product design stage, realizes rapid performance verification and iterative design, and reduces simulation Cycles and costs.
Patent Information
- Application Number
- CN202211272262.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-10-18
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2042-10-18
AI Technical Summary
It is difficult for the prior art to quickly obtain virtual experimental data close to digital or physical simulation experiments during the product design stage, resulting in long product performance verification time, especially in product design under complex configurations and operating conditions, where the simulation cycle is difficult to control.
Using the high-performance virtual simulation method driven by digital twin data model, by obtaining sensing data, mechanism simulation data and fusion data, we build a high-performance virtual simulation model driven by data model, reversely analyze the operating mechanism, obtain virtual simulation results, and replace the construction of Simulation or physical experiments to conduct design product performance analysis and prediction.
It shortens the product performance verification time, accelerates the forward design and iteration of products, solves the problems of high physical experiment costs and long computer simulation solutions, and realizes the performance prediction and analysis of quickly obtaining virtual experimental data.
Smart Images

Figure CN115659791B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of intelligent and digital industrial equipment, and in particular to a high-performance virtual simulation method and system driven by a digital twin data model. Background Art
[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
[0003] As we all know, in order to verify the performance of new products during the design, molding, and optimization and improvement stages, it is necessary to simulate the product operating environment and conduct product verification experiments. However, the cost and time of physical experimental verification are huge, especially for product designs with complex configurations and complex working conditions. Sometimes, physical experimental testing of samples to accurately simulate the actual product operating conditions and conduct performance evaluation is limited by economic and manual experimental costs, and has poor feasibility. In recent years, with the development of numerical simulation and computer capabilities, computer simulation technology has become an effective tool to replace sample physical experiments for experimental verification.
[0004] At present, the method adopted by computer simulation technology is to use simulation software to simulate the physical experimental conditions of the product, and to obtain virtual simulation results that characterize the product performance by conducting digital simulation, so as to evaluate / verify the product performance. However, the traditional mechanism simulation method has high requirements on computer performance throughout the whole process, and the simulation solution cycle is long. Especially when conducting iterative optimization design of products and analyzing the performance of the designed products under continuous operation conditions, due to the complexity of simulation calculations, the duration of the simulation cycle is often very long and difficult to estimate, even in high-performance computer clusters. At present, the solution to the above problems is to construct a reduced-order model to shorten the simulation solution time by reducing the dimension of the state space model. However, the shortened solution time of this method is limited, the loss of accuracy is difficult to control, and it is not applicable to all simulation software. And when conducting iterative optimization design of products and analyzing the performance of the designed products under continuous operation conditions, it still cannot fundamentally solve the problem of long simulation cycle time for performance verification evaluation.
[0005] With the development of new generation information technology, the birth of the concept of digital twins provides an effective method for product virtual performance verification. Digital twins are considered to be an effective means of enabling information-physical integration. They are simulation technologies that integrate disciplines, multiple physical quantities, multiple scales, and multiple probabilities. The system implementation elements include three parts: mechanism model, data model, and algorithm model. Through digital twin technology, reliable data information can be obtained from the product digital twin model or physical space, and based on the application of algorithm models, it is possible to carry out product design performance verification.
[0006] To sum up, in the product design stage, how to use digital twin technology to quickly obtain virtual experimental data that is close to digital simulation or physical simulation experiments / operation experiments, realize product performance prediction and analysis, and accelerate product forward design and iteration is a challenge in the current virtual experimental verification. Summary of the invention
[0007] In order to overcome the shortcomings of the above-mentioned existing technologies, a high-performance virtual simulation method and system driven by a digital twin data model is proposed, which is suitable for the iterative optimization design of products and the analysis of the performance of designed products under continuous operation conditions. The digital twin technology is used to construct a mechanism model of related products / designed products, and obtain one or more of the mechanism simulation data, operation monitoring data or physical experiment simulation data and fusion data that characterize the product performance. Then, based on the simulation needs, a high-performance virtual simulation model driven by a single data or fusion data is constructed. The product mechanism model is reversely parsed according to the data-driven algorithm to obtain the physical space operation mechanism simulation and performance evaluation required for the same type of designed products or designs under different working conditions, so as to replace modeling simulation or physical experiments, analyze and predict the performance of the designed products, and shorten the performance verification time of the designed products.
[0008] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:
[0009] A first aspect of the present invention provides a high-performance virtual simulation method driven by a digital twin data model.
[0010] A high-performance virtual simulation method driven by a digital twin data model comprises the following steps:
[0011] Based on the product design process, pay attention to the changing working conditions or changing product configurations, and obtain three types of data related to the designed product, related products and working conditions: sensor data, mechanism simulation data, and fusion data obtained by correcting sensor data with mechanism simulation data;
[0012] Process the acquired data, rely on data-driven algorithms, reversely analyze the operating mechanism, and build a high-performance virtual simulation model driven by data models;
[0013] Based on the new working conditions of the designed product or the improved new product, the high-performance virtual simulation model driven by the constructed data model is called to obtain the virtual simulation results;
[0014] Analyze and design product performance under new working conditions or new products based on virtual simulation results, and provide feedback to guide product design.
[0015] A second aspect of the present invention provides a high-performance virtual simulation system driven by a digital twin data model.
[0016] A high-performance virtual simulation system driven by a digital twin data model, comprising:
[0017] The data acquisition module is configured to: based on the product design process, pay attention to the changing working conditions or changing product configurations, and acquire three types of data related to the designed product, related products and working conditions: sensor data, mechanism simulation data, and fusion data obtained by correcting the sensor data with the mechanism simulation data;
[0018] The model building module is configured to: process the acquired data, rely on the data-driven algorithm, reversely analyze the operation mechanism, and build a high-performance virtual simulation model driven by the data model;
[0019] The virtual simulation result acquisition module is configured to: based on the new working condition of the designed product or the improved new product, call the constructed data model driven high-performance virtual simulation model to obtain the virtual simulation result;
[0020] The feedback guidance module is configured to: analyze the performance of the designed product under new working conditions or new products according to the virtual simulation results, and provide feedback to guide product design.
[0021] The third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the high-performance virtual simulation method driven by a digital twin data model as described in the first aspect of the present invention.
[0022] The fourth aspect of the present invention provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the high-performance virtual simulation method driven by the digital twin data model as described in the first aspect of the present invention are implemented.
[0023] One or more of the above technical solutions have the following beneficial effects:
[0024] The present invention utilizes digital twin technology to construct a mechanism model of the design or related products and a high-performance virtual simulation model driven by a data model, acquires simulation data, physical data, and fusion data, and performs digital simulation solutions for new products or new working conditions from a data-driven perspective to obtain their operating mechanisms and conduct performance evaluations, thereby solving the problems of high physical experiment costs and long computer simulation solution times, and is conducive to accelerating product positive performance verification.
[0025] The present invention utilizes digital twin technology to target discrete operating conditions of designed products, obtains usage process sensor data, physical experiment sensor data, or multi-domain digital twin mechanism model simulation data of designed products under discrete operating conditions, or fusion data of the above data, to drive a high-performance virtual simulation model to analyze the digital simulation or physical space operation mechanism of the designed product under discrete operating conditions, and analyze the performance of the designed product under continuous operating conditions.
[0026] The present invention utilizes digital twin technology to target variable product configurations under the same operating conditions, obtains usage process sensor data, physical experiment sensor data, or multi-domain digital twin mechanism model simulation data of related products under the same operating conditions, or fusion data of the above data, to drive a high-performance virtual simulation model to analyze the digital simulation or physical space operation mechanism of variable related products under the same operating conditions, and promote the optimized iterative design of designed products.
[0027] The present invention utilizes one or more of the three types of data, namely simulation data, sensor data and fusion data, to construct a data model-driven high-performance virtual simulation model, which can replace modeling simulation or physical experiments to analyze and predict the performance of designed products, shorten product performance verification time, and accelerate product forward design and iteration.
[0028] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0030] Figure 1 This is a flow chart of a method according to Embodiment 1 of the present invention;
[0031] Figure 2 This is a flowchart of the multi-domain digital twin model construction process of the first embodiment of the present invention;
[0032] Figure 3 This is a flow chart of a correction process of sensor data to simulation data according to a first embodiment of the present invention;
[0033] Figure 4 This is a flowchart of a process for building a high-performance virtual simulation system driven by a digital twin data model according to Embodiment 1 of the present invention;
[0034] Figure 5 A flowchart of product virtual simulation (performance verification prediction and analysis) designed for embodiment 1 of the present invention;
[0035] Figure 6This is a flowchart of product virtual simulation (performance prediction and analysis) based on constant working conditions in Embodiment 1 of the present invention;
[0036] Figure 7 This is a flow chart of product virtual simulation (performance prediction and analysis) based on a constant design product in Embodiment 1 of the present invention;
[0037] Figure 8 It is a structural diagram of embodiment 2 of the present invention. DETAILED DESCRIPTION
[0038] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.
[0039] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.
[0040] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0041] Embodiment 1
[0042] This embodiment discloses a high-performance virtual simulation method driven by a digital twin data model.
[0043] like Figure 1 As shown, a high-performance virtual simulation method driven by a digital twin data model includes the following steps:
[0044] Based on the product design process, pay attention to the changing working conditions or changing product configurations, and obtain three types of data related to the designed product, related products and working conditions: sensor data, mechanism simulation data, and fusion data obtained by correcting sensor data with mechanism simulation data;
[0045] Process the acquired data, rely on data-driven algorithms, reversely analyze the operating mechanism, and build a high-performance virtual simulation model driven by data models;
[0046] Based on the new working conditions of the designed product or the improved new product, the high-performance virtual simulation model driven by the constructed data model is called to obtain the virtual simulation results;
[0047] Analyze and design product performance under new working conditions or new products based on virtual simulation results, and provide feedback to guide product design.
[0048] In data acquisition, based on the product design process, attention is paid to changing operating conditions or changing product configurations, and three types of data related to the designed products, related products and operating conditions are obtained, namely: sensor data obtained from the product usage process or physical experiments related to product design; mechanism simulation data obtained by constructing multi-domain digital twin models of related or designed products, and performing mechanism simulation on related or designed products and operating conditions; and fusion data formed by correcting simulation data with sensor data through data correction algorithms.
[0049] In data-driven modeling, one or more of the above three types of data are processed and further summarized into the data repository. Then, relying on the data-driven algorithm, one or more of the three types of data are used to reversely analyze the operating mechanism according to the working condition set or product model set to build a high-performance virtual simulation model driven by the data model.
[0050] In virtual simulation, according to the needs of designers, based on the new operating conditions of the designed product or improved new products, the constructed data model-driven high-performance virtual simulation model is called to obtain virtual simulation results that characterize product performance that are close to digital simulation or physical experiments. The performance of the designed product under new operating conditions or new products is further analyzed based on the simulation results, and feedback is used to guide product design.
[0051] Among them, sensor data is obtained from existing physical experiments or product operation processes related to the designed product, and it has the function of replacing simulation data or correcting simulation data. Considering the limitations of actual application scenarios, not all designed products have the conditions for sensing physical data, so sensor data cannot be used in all situations.
[0052] The reason for adopting the data fusion method is that the number of sensor data representing performance obtained from the actual process / physical experiment is small, and a large amount of performance characterization data is required to build a high-performance virtual simulation model driven by a data model. Therefore, according to the data correction algorithm, the performance characterization data obtained through the mechanism simulation is corrected by the obtained sensor data representing performance, so that the performance characterization data is more accurate, thereby obtaining fused data. Data correction algorithms include particle swarm optimization, genetic algorithm, and ant colony optimization.
[0053] Among them, in the process of high-performance virtual simulation of products, the data samples representing product performance obtained by focusing on physical experiments / actual product operation / digital twin mechanism model simulation for two or more similar products must be large enough. Only in this way can the three types of data obtained meet the needs of building high-performance virtual simulation models in two situations.
[0054] like Figure 2 As shown in the figure, the multi-domain digital twin model of related or designed products is constructed as follows:
[0055] Based on the product design process, we focus on changing working conditions or changing product configurations, and map the twin parameters related to physical space and product design, which are the same working conditions of related and designed products and the configuration attributes of related products; the designed product itself and the changing working conditions of the designed product.
[0056] Based on modeling tools, model correction algorithms and mapped twin parameters, digital twin models of multi-domain related products under constant working conditions including different systems such as mechanical, control, and electromagnetic, or digital twin models of multi-domain designed products under discrete working conditions are constructed.
[0057] Among them, the correction of the model is mainly the correction of the twin parameters of the model, including the use of Bayesian, genetic algorithm and other model correction algorithms to carry out global optimization, local optimization and combined optimization of the model.
[0058] like Figure 3 As shown in the figure, the sensor data representing the product performance corrects the simulation data. The specific process is as follows:
[0059] According to the changing working conditions or changing product configurations related to product design, a small amount of sensor data of the product usage process / physical experiment and the simulation results of the multi-domain digital twin model mechanism of the design or related products are obtained, and the digital twin model construction theory is converted into a state space model.
[0060] Select a suitable model correction algorithm and set the correction threshold according to application requirements. At the same time, set the sensor data as the observation equation of the data model correction process, integrate the state model into the state equation of the correction algorithm, and obtain the mechanism simulation results as the internal value of the correction algorithm.
[0061] The acquired mechanism simulation data is input into the model correction algorithm to output the result data, evaluate the performance of the designed product or related products, and compare the distribution trends of the sensor data, simulation data and result data. It is analyzed whether the result data meets the set threshold conditions. If the set threshold is met, the fusion / corrected data is output. If not, the model correction algorithm is reselected and iterated.
[0062] like Figure 4 As shown in the figure, the high-performance virtual simulation system driven by the digital twin data model has the following specific construction process:
[0063] Obtain the above-mentioned sensor data based on constant operating conditions or physical experiments / product usage processes of designed products, the mechanism simulation result data of the multi-domain digital twin model of the above-mentioned related products or designed products, and the above-mentioned fusion data obtained through the data correction algorithm.
[0064] One or more of the above three types of data are processed, and the processed data are further summarized into a data repository for processing for subsequent data mining.
[0065] From the perspective of big data, the processed data is stored in the database to manage and organize the relevant data in an orderly manner.
[0066] Set product design conditions with variable operating conditions or variable product configurations, select and train regression algorithms based on multi-category, multi-indicator, and multi-dimensional product data, and build a high-performance virtual simulation model driven by a data model.
[0067] Among them, the data processing may include data preprocessing, data expansion, feature extraction and feature selection. Feature extraction and feature selection are selectively determined based on the data-driven algorithm and are not necessary. Data expansion is to expand the magnitude of simulation data through the algorithm.
[0068] The high-performance virtual simulation model driven by the data model based on the data-driven algorithm is a data regression black box model that uses the product design conditions as input and the performance characterization data as output through the regression algorithm. The regression algorithm includes CNN, ANN, SVM, etc.
[0069] The data model driven high-performance virtual simulation model is multifunctional, based on application requirements and facing constant working conditions or design products. For example, virtual simulation analysis of stress and strain of variable product configurations under the same working condition only requires building a data model driven high-performance virtual simulation model around stress and strain to support improved product performance analysis.
[0070] like Figure 5 As shown, the virtual simulation (performance prediction and analysis) of the designed product is implemented as follows:
[0071] Based on the new working conditions of the designed product or the new improved product, the regression model of the constant working conditions or the constant designed product driven by the data model is called respectively to output the virtual simulation results close to the digital simulation / physical experiment / operational experiment.
[0072] Statistical analysis, data visualization, interpolation processing and other data methods are further used to analyze and predict the performance of the designed products to guide product design.
[0073] Specifically, virtual simulation (product performance verification) based on constant operating conditions, such as Figure 6As shown in the figure, first, based on the existing physical experiments / operation experiments, the sensor data of related products under constant working conditions are obtained, and the mechanism simulation of the related products is carried out to obtain the mechanism simulation data, and the fusion data is obtained based on the data correction algorithm. Then, a high-performance virtual simulation model is constructed based on one or more of the above three data characterizing product performance. Finally, the virtual simulation data of the new product that is progressively improved is used to analyze the performance of the improved new product through statistical analysis, data visualization, etc.
[0074] Specifically, virtual simulation (product performance verification) based on constant design products, such as Figure 7 As shown, first, based on the existing physical experiments / operation experiments, the sensor data of the designed product under variable working conditions is obtained, and the mechanism simulation of the variable working conditions is carried out to obtain the mechanism simulation data, and the fusion data is obtained based on the data correction algorithm. Then, a high-performance virtual simulation model is constructed based on one or more of the above three data representing product performance. Finally, the virtual simulation data under the new working conditions is derived, and the data representing product performance is incorporated through statistical analysis, interpolation processing, etc., to analyze the performance of the designed product under continuous working conditions.
[0075] Embodiment 2
[0076] This embodiment discloses a high-performance virtual simulation system driven by a digital twin data model.
[0077] like Figure 8 As shown, a high-performance virtual simulation system driven by a digital twin data model includes:
[0078] The data acquisition module is configured to: based on the product design process, pay attention to the changing working conditions or changing product configurations, and acquire three types of data related to the designed product, related products and working conditions: sensor data, mechanism simulation data, and fusion data obtained by correcting the sensor data with the mechanism simulation data;
[0079] The model building module is configured to: process the acquired data, rely on the data-driven algorithm, reversely analyze the operation mechanism, and build a high-performance virtual simulation model driven by the data model;
[0080] The virtual simulation result acquisition module is configured to: based on the new working condition of the designed product or the improved new product, call the constructed data model driven high-performance virtual simulation model to obtain the virtual simulation result;
[0081] The feedback guidance module is configured to: analyze the performance of the designed product under new working conditions or new products according to the virtual simulation results, and provide feedback to guide product design.
[0082] Embodiment 3
[0083] The purpose of this embodiment is to provide a computer-readable storage medium.
[0084] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the high-performance virtual simulation method driven by a digital twin data model as described in Example 1 of the present disclosure.
[0085] Embodiment 4
[0086] The purpose of this embodiment is to provide an electronic device.
[0087] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the high-performance virtual simulation method driven by the digital twin data model as described in Example 1 of the present disclosure are implemented.
[0088] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.
[0089] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.
[0090] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.
Claims
1. A high-performance virtual simulation method driven by a digital twin data model, characterized in that: The following steps are involved: Based on the product design process, pay attention to the changing working conditions or changing product configurations, and obtain three types of data related to the designed product, related products and working conditions: sensor data, mechanism simulation data, and fusion data obtained by correcting the mechanism simulation data with sensor data; The sensor data that characterizes product performance corrects the simulation data. The specific process is as follows: According to the variable working conditions or variable product configurations related to product design, obtain a small amount of sensor data from the product use process / physical experiment and the simulation results of the multi-domain digital twin model mechanism of the construction design or related products, and convert the digital twin model construction theory into a state space model; Select a suitable model correction algorithm, set the correction threshold according to the application requirements, set the sensor data as the observation equation of the data model correction process, integrate the state model into the state equation of the correction algorithm, and obtain the mechanism simulation results as the internal value of the correction algorithm; Input the acquired mechanism simulation data into the model correction algorithm to output the result data, evaluate the performance of the designed product or related products, and compare the distribution trends of the sensor data, simulation data and result data, and analyze whether the result data meets the set threshold conditions. If the set threshold is met, the fusion / corrected data is output. If not, the model correction algorithm is reselected and iterated; Process the acquired data, rely on data-driven algorithms, reversely analyze the operating mechanism, and build a high-performance virtual simulation model driven by data models; Based on the new working conditions of the designed product or the improved new product, the high-performance virtual simulation model driven by the constructed data model is called to obtain the virtual simulation results; Analyze and design product performance under new working conditions or new products based on virtual simulation results, and provide feedback to guide product design; The acquired data is processed, and the operation mechanism is reversely analyzed based on the data-driven algorithm to build a high-performance virtual simulation model driven by the data model. Specifically: Process one or more of the sensor data, mechanism simulation data, and fusion data, obtain the data processing results, select a regression algorithm and train the regression algorithm based on the data processing results, and build a high-performance virtual simulation model driven by a data model; The high-performance virtual simulation model driven by the data model is a data regression black box model built through regression algorithms, which takes product design conditions of variable operating conditions or variable product configurations as input and performance characterization data as output. The regression algorithms include CNN, ANN, and SVM. Sensor data, mechanism simulation data, fusion data and virtual simulation results are all multi-category, multi-indicator and multi-dimensional data. The high-performance virtual simulation model driven by data models built for constant working conditions or design products is multifunctional.
2. The high-performance virtual simulation method driven by the digital twin data model according to claim 1, characterized in that: Acquire sensor data through product usage processes or physical experiments related to product design. Acquire mechanism simulation data by constructing multi-domain digital twin models of related products or designed products and operating conditions to perform mechanism simulation. Correct simulation data through sensor data to acquire fused data.
3. The high-performance virtual simulation method driven by the digital twin data model according to claim 2, characterized in that: Build multi-domain digital twin models of related products or design products, specifically: Based on the product design process, pay attention to changing working conditions or changing product configurations, and map the twin parameters related to physical space and product design; based on modeling tools, model correction algorithms and mapped twin parameters, build digital twin models of multi-domain related products under constant working conditions or digital twin models of multi-domain designed products under discrete working conditions.
4. The high-performance virtual simulation method driven by the digital twin data model according to claim 2, characterized in that: The simulation data is corrected through the sensor data to obtain the fusion data, specifically: Based on the data correction algorithm, the simulation data representing the performance obtained through the mechanism simulation is corrected by obtaining a small amount of sensor data representing the performance. The data correction algorithm includes a particle swarm algorithm, a genetic algorithm, and an ant colony algorithm.
5. A high-performance virtual simulation system driven by a digital twin data model, characterized in that: include: The data acquisition module is configured to: based on the product design process, pay attention to the changing working conditions or changing product configurations, and acquire three types of data related to the designed product, related products and working conditions: sensor data, mechanism simulation data, and fusion data obtained by correcting the mechanism simulation data with the sensor data; The sensor data that characterizes product performance corrects the simulation data. The specific process is as follows: According to the variable working conditions or variable product configurations related to product design, obtain a small amount of sensor data from the product use process / physical experiment and the simulation results of the multi-domain digital twin model mechanism of the construction design or related products, and convert the digital twin model construction theory into a state space model; Select a suitable model correction algorithm, set the correction threshold according to the application requirements, set the sensor data as the observation equation of the data model correction process, integrate the state model into the state equation of the correction algorithm, and obtain the mechanism simulation results as the internal value of the correction algorithm; Input the acquired mechanism simulation data into the model correction algorithm to output the result data, evaluate the performance of the designed product or related products, and compare the distribution trends of the sensor data, simulation data and result data, and analyze whether the result data meets the set threshold conditions. If the set threshold is met, the fusion / corrected data is output. If not, the model correction algorithm is reselected and iterated; The model building module is configured to: process the acquired data, rely on the data-driven algorithm, reversely analyze the operation mechanism, and build a high-performance virtual simulation model driven by the data model; The virtual simulation result acquisition module is configured to: based on the new working condition of the designed product or the improved new product, call the constructed data model driven high-performance virtual simulation model to obtain the virtual simulation result; The feedback guidance module is configured to: analyze the performance of the designed product under new working conditions or new products according to the virtual simulation results, and provide feedback guidance for product design; The acquired data is processed, and the operation mechanism is reversely analyzed based on the data-driven algorithm to build a high-performance virtual simulation model driven by the data model. Specifically: Process one or more of the sensor data, mechanism simulation data, and fusion data, obtain the data processing results, select a regression algorithm and train the regression algorithm based on the data processing results, and build a high-performance virtual simulation model driven by a data model; The high-performance virtual simulation model driven by the data model is a data regression black box model built through regression algorithms, which takes product design conditions of variable operating conditions or variable product configurations as input and performance characterization data as output. The regression algorithms include CNN, ANN, and SVM. Sensor data, mechanism simulation data, fusion data and virtual simulation results are all multi-category, multi-indicator and multi-dimensional data. The high-performance virtual simulation model driven by data models built for constant working conditions or design products is multifunctional.
6. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps in the high-performance virtual simulation method driven by a digital twin data model as described in any one of claims 1 to 4 are implemented.
7. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the high-performance virtual simulation method driven by the digital twin data model as described in any one of claims 1 to 4 are implemented.
Citation Information
Patent Citations
Method and system for correcting digital twinborn model data of evaporation area of thermal power plant
CN112818595A
Complex product digital twinning construction and application method based on model fusion
CN114611313A
