A multi-physics field calculation method and system for online simulation of digital twins
Through simplifying and downgrading multi-physics coupled simulation calculation model, combining basic data-driven model and tolerance factor correction, the problem of high cost and low efficiency of modeling multi-physics simulation technology is solved, and efficient and low error calculation is achieved The result output.
Patent Information
- Application Number
- CN202411134601.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-19
AI Technical Summary
Multiphysics simulation technology has high time cost and low efficiency. Traditional agent modeling methods require a large number of accurate multiphysics simulation calculation results as input, and the modeling cost is high and there may be large errors.
By establishing a multi-physics field coupled simulation calculation model of the simulated object, simplifying it, the lowering temperature field simulation model is performed, and analyzing it based on the preset sample space is constructed, and the temperature field is quickly outputted using low-precision data sets. The distribution results further reduce the calculation error by modifying the model through the tolerance factor.
It significantly reduces the calculation amount, improves the calculation efficiency, reduces modeling costs, and reduces calculation errors, achieving fast multi-physics distribution result output.
Smart Images

Figure CN118657092B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of physical field simulation technology, and in particular to a multi-physical field calculation method and system for online simulation of digital twins. Background Art
[0002] The digital twin technology of power transmission and transformation equipment is one of the key points of the current digital transformation of the power grid system. It adopts digital and information-based technical means to build a digital model in the virtual space, and interacts through sensor data to achieve real-time update output of equipment status, thereby achieving the purpose of "virtual and real symbiosis" between the equipment entity and the twin, and supporting the operation and maintenance of the equipment and the full life cycle status assessment.
[0003] Due to insulation safety issues with power equipment, it is difficult for sensors to directly penetrate into the equipment to measure physical quantities. Therefore, one of the current technical routes for real-time and accurate characterization of equipment operating characteristics and comprehensive utilization of perception data in equipment digital twins is to combine it with multi-physics field simulation computing technology. When the sensor obtains specific monitoring data from the equipment entity, multi-physics field simulation is applied, and the finite signal is used as the boundary condition to solve the black box inverse problem of the multi-physics field, so as to obtain the overall multi-physics field distribution of the equipment in this case and realize the visual performance evaluation of the internal state of the equipment.
[0004] Although multi-physics field simulation technology has been widely used in the design and manufacturing stages, it takes a lot of time to simulate due to the complex structure of the equipment and the characteristics of finite element and finite volume method meshing, and its real-time or quasi-real-time performance is weak. Developing real-time simulation calculation methods is an urgent problem to be solved.
[0005] The use of data-driven proxy models can quickly obtain the multi-physics field distribution of equipment, but currently this technology requires enumerating different operating conditions that may occur in a limited set of equipment and performing high-precision finite element calculations. There are the following problems: the more operating conditions there are, the more accurate the model will be after training and construction, but the more operating conditions need to be calculated, the higher the calculation cost; using less data has low calculation cost but large error; under a limited set of operating conditions, calculating multiple sets of conditions, finite element simulation itself will take a lot of time and have high calculation cost. When the situation exceeds the enumeration boundary, the model needs to be rebuilt, and the time required to rebuild the additional data and configuration is relatively long.
[0006] In summary, the current multi-physics simulation technology modeling has a high time cost and low efficiency; the traditional proxy model method requires a large number of accurate multi-physics simulation calculation results as input, which has high modeling costs and may have large errors. Summary of the invention
[0007] The present invention provides a multi-physics field calculation method and system for online simulation of digital twins, which solves the technical problems that the current multi-physics field simulation technology has high time cost and low efficiency in modeling, and the traditional proxy model method requires a large number of accurate multi-physics field simulation calculation results as input, has high modeling cost, and may have large errors.
[0008] In view of this, the first aspect of the present invention provides a multi-physics field calculation method for online simulation of digital twins, comprising:
[0009] Establishing a multi-physics field coupling simulation calculation model of a simulated object, and simplifying the multi-physics field coupling simulation calculation model;
[0010] Determine a temperature field simulation model according to the simplified multi-physics field coupling simulation calculation model, and reduce the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model;
[0011] Based on a preset sample space, performing temperature field analysis according to the temperature field simulation reduced-order model to obtain a low-precision data set;
[0012] Constructing a basic data-driven model according to the low-precision data set and the sample space corresponding to the low-precision data set, wherein the basic data-driven model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data-driven model;
[0013] The current sample space is input into the basic data driven model, and a low-precision data set corresponding to the current sample space is output.
[0014] Preferably, the step of establishing a multi-physics field coupling simulation calculation model of the simulated object and simplifying the multi-physics field coupling simulation calculation model specifically includes:
[0015] Constructing a multi-physics field coupling simulation calculation model according to the geometric structure of the simulated object, wherein the physical fields include mutually coupled temperature field, stress field and fluid field;
[0016] The coupled grid division between the temperature field and the stress field is simplified, and the convection heat transfer process of the fluid field is equivalent to a heat conduction process.
[0017] Preferably, the method further comprises:
[0018] The temperature field simulation model is reduced in order by using an intrinsic orthogonal decomposition method to obtain a temperature field simulation reduced-order model.
[0019] Preferably, the step of constructing a basic data-driven model according to the low-precision data set and the sample space corresponding to the low-precision data set, wherein the basic data-driven model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data-driven model, specifically includes:
[0020] Constructing a first training data set according to the low-precision data set and a sample space corresponding to the low-precision data set;
[0021] An initial basic data-driven model is constructed, and the initial basic data-driven model is trained using the first training data set, and after the training converges, a basic data-driven model is obtained.
[0022] Preferably, the method further comprises:
[0023] The basic data driven model is modified.
[0024] Preferably, the step of modifying the basic data driven model specifically includes:
[0025] Determine multiple groups of sample spaces and tolerance factors corresponding to the multiple groups of sample spaces according to the basic data-driven model and the gap between the high-precision data set and the low-precision data set;
[0026] Constructing a second training data set according to a plurality of groups of sample spaces and tolerance factors corresponding to the plurality of groups of sample spaces;
[0027] Training the second training data set based on a deep neural network to obtain a tolerance factor prediction regression model;
[0028] The basic data driven model is modified according to the tolerance factor prediction regression model to obtain a multi-precision proxy model.
[0029] Preferably, the method further comprises:
[0030] Performing an error test on the output result of the multi-precision proxy model;
[0031] If the error test result of the output result of the multi-precision proxy model is qualified, outputting the multi-precision proxy model;
[0032] If the error test result of the output result of the multi-precision proxy model fails to meet the requirements, the step of switching to the preset sample space, performing temperature field analysis according to the temperature field simulation reduced-order model, and iterating the step of obtaining a low-precision data set is performed until the error test result of the output result of the multi-precision proxy model meets the requirements, stopping the iteration and outputting the multi-precision proxy model.
[0033] In a second aspect, the present invention further provides a multi-physics field computing system for online simulation of digital twins, comprising:
[0034] A simulation modeling module is used to establish a multi-physics field coupling simulation calculation model of a simulated object and simplify the multi-physics field coupling simulation calculation model;
[0035] A model reduction module is used to determine a temperature field simulation model according to the simplified multi-physics field coupling simulation calculation model, and reduce the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model;
[0036] A temperature field analysis module, used to perform temperature field analysis based on a preset sample space and the temperature field simulation reduced-order model to obtain a low-precision data set;
[0037] A driving model construction module, used to construct a basic data driving model according to the low-precision data set and the sample space corresponding to the low-precision data set, wherein the basic data driving model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data driving model;
[0038] The data driving module is used to input the current sample space into the basic data driving model and output the low-precision data set corresponding to the current sample space.
[0039] In a third aspect, the present invention further provides an electronic device, the electronic device comprising a memory and a processor;
[0040] The memory is used to store programs;
[0041] The processor executes the program to implement the steps of the above-mentioned multi-physics field calculation method for digital twin online simulation.
[0042] In a fourth aspect, the present invention also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the steps of the multi-physics field calculation method for online simulation of digital twins.
[0043] It can be seen from the above technical solutions that the present invention has the following advantages:
[0044] The present invention establishes a multi-physical field coupling simulation calculation model of the simulated object, simplifies the multi-physical field coupling simulation calculation model, and reduces the order of the temperature field simulation model, thereby greatly reducing the amount of calculation and improving the calculation efficiency. The low-precision data set obtained through temperature field analysis uses the low-precision data to reduce the amount of data required for the model and reduce the cost of modeling. The basic data-driven model is constructed through the low-precision data set and its corresponding sample space, so that the temperature field distribution results can be quickly output, which further improves the calculation efficiency and reduces the calculation error. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 A flowchart of a multi-physics field calculation method for online simulation of a digital twin provided by an embodiment of the present invention;
[0046] Figure 2 A flowchart of step S1 provided in an embodiment of the present invention;
[0047] Figure 3 A flowchart for modifying a basic data driven model provided by an embodiment of the present invention;
[0048] Figure 4 A schematic diagram of the structure of a multi-physics field computing system for online simulation of a digital twin provided by an embodiment of the present invention;
[0049] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0050] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0052] The current multi-physics simulation technology requires enumerating different possible working conditions of a finite set of equipment and performing high-precision finite element calculations. At the same time, there are the following problems: the more working conditions there are, the more accurate the model will be after training and construction, but the more working conditions need to be calculated, the higher the calculation cost; using less data means lower calculation cost, but large errors; under a finite set of working conditions, calculating multiple sets of conditions will cost a lot of time and high calculation cost for finite element simulation itself. When the situation exceeds the enumeration boundary, the model needs to be rebuilt, and the time required to rebuild the additional data and configuration is relatively long.
[0053] In summary, the current multi-physics field simulation technology has high modeling time cost, low efficiency and large calculation errors.
[0054] In view of this, an embodiment of the present invention provides a multi-physics field calculation method for online simulation of digital twins. The embodiment of the present invention can be applicable to the case of multi-physics field calculation. The method can be executed by a multi-physics field calculation device. The multi-physics field calculation device can be implemented in the form of hardware and / or software. The multi-physics field calculation device can be configured in a computer device.
[0055] See also Figure 1 , Figure 1 The flowchart of a multi-physics field calculation method for online simulation of digital twins provided by an embodiment of the present invention is illustrated.
[0056] An embodiment of the present invention provides a multi-physics field calculation method for online simulation of a digital twin, including:
[0057] Step S1: establishing a multi-physics field coupling simulation calculation model of a simulated object, and simplifying the multi-physics field coupling simulation calculation model.
[0058] In the field of power applications, the simulated object may be power equipment, such as a UHV converter transformer valve-side bushing.
[0059] Specifically, if Figure 2As shown, step S1 specifically includes:
[0060] Step S101 : constructing a multi-physical field coupling simulation calculation model according to the geometric structure of the simulated object, wherein the physical field includes mutually coupled temperature field, stress field and fluid field.
[0061] Step S102, simplifying the coupled grid division between the temperature field and the stress field, and equating the convection heat transfer process of the fluid field to the heat conduction process.
[0062] It should be noted that, affected by the thermal expansion of the material, the stress characteristics of the elastic electrical connection structure will change after the temperature changes. Considering this problem, the difficulty of meshing is increased, which will lead to an unnecessary increase in the number of meshes and increase the amount of calculation. For this reason, this embodiment simplifies the coupling simulation process between the temperature field and the stress field by dividing the coupled meshes between the temperature field and the stress field, so as not to consider the influence of the temperature field distribution on the thermal expansion of the material, thereby reducing the difficulty of meshing and reducing the amount of calculation.
[0063] Among them, the convective heat transfer process is not easy to obtain. In this embodiment, the convective heat transfer process of the fluid field is equivalent to a heat conduction process, thereby reducing the amount of calculation.
[0064] Step S2: determining a temperature field simulation model according to the simplified multi-physics field coupling simulation calculation model, and reducing the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model.
[0065] In some specific embodiments, an intrinsic orthogonal decomposition method is used to reduce the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model.
[0066] Specifically, the temperature field simulation model can be characterized by the temperature field finite element discrete equation. The temperature field simulation reduced-order model can map the temperature field calculation nodes to a set of orthogonal basis eigenvectors to form a low-dimensional space to further improve the calculation efficiency of the model. Through research, it is found that the single simulation calculation time of the temperature field simulation reduced-order model can be controlled within seconds.
[0067] Step S3: Based on the preset sample space, the temperature field is analyzed according to the temperature field simulation reduced-order model to obtain a low-precision data set.
[0068] Here, the sample space refers to the boundary conditions and loads in multiphysics simulations.
[0069] The sample space is obtained by optimizing the nested Latin hypercube. That is, firstly, the low-dimensional sample space value U is obtained by sampling. L , and then select the high-precision sample space U from the low-dimensional sample space H , U H U L.
[0070] After setting the sample space, the temperature field analysis distribution results, i.e., the grid node data, are obtained through temperature field analysis, and a low-precision data set is formed. Among them, a set of sample spaces corresponds to a set of grid node data.
[0071] Step S4: construct a basic data-driven model based on the low-precision data set and the sample space corresponding to the low-precision data set. The basic data-driven model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data-driven model.
[0072] Among them, the basic data-driven model can adopt one of the methods such as neural network, response surface method and Kriging method. By using the above methods to establish the basic data-driven model respectively, and performing error tests on the basic data-driven model, the method with the smallest error is selected to construct the basic data-driven model.
[0073] In a specific embodiment, step S4 includes:
[0074] Step S401: construct a first training data set according to a low-precision data set and a sample space corresponding to the low-precision data set.
[0075] Step S402: construct an initial basic data driven model, train the initial basic data driven model using a first training data set, and obtain a basic data driven model after the training converges.
[0076] Among them, the basic data-driven model is constructed with the sample space as input and the low-precision data set corresponding to the sample space as output. That is, the basic data-driven model can quickly output the grid node data when any simulation boundary conditions are input.
[0077] Step S5: input the current sample space into the basic data driven model, and output the low-precision data set corresponding to the current sample space.
[0078] It should be noted that the embodiments of the present invention establish a multi-physics field coupling simulation calculation model of the simulated object, simplify the multi-physics field coupling simulation calculation model, and reduce the order of the temperature field simulation model, thereby greatly reducing the amount of calculation and improving the calculation efficiency. The low-precision data set obtained through temperature field analysis uses low-precision data to reduce the amount of data required for the model and reduce the cost of modeling. The basic data-driven model is constructed through the low-precision data set and its corresponding sample space, so that the temperature field distribution results can be quickly output, further improving the calculation efficiency and reducing the calculation error.
[0079] In some specific embodiments, in order to further reduce the error of the basic data-driven model, the embodiments of the present invention also modify the basic data-driven model.
[0080] Specifically, if Figure 3 As shown in FIG. 1 , the steps for modifying the basic data-driven model include:
[0081] Step S601: Determine multiple groups of sample spaces and tolerance factors corresponding to the multiple groups of sample spaces according to the basic data-driven model and the gap between the high-precision data set and the low-precision data set.
[0082] The process of obtaining high-precision data sets is as follows:
[0083] By refining the multi-physical field coupling simulation calculation model, taking the multi-physical field coupling simulation calculation model of the UHV converter transformer valve side bushing as an example, the elastic electrical connection structure is finely modeled according to the actual design to consider the influence of the current-carrying connection structure on the compression and contact stress of the elastic electrical connection during assembly, and the relationship between contact stress and contact resistance is obtained according to the Holm electrical contact theory.
[0084] The contact resistance value is taken as a parameter condition, and the overall loss distribution of the bushing on the converter transformer valve side is obtained through electromagnetic field calculation as a heat source, which is substituted into the thermal-fluid coupling calculation equation to obtain the temperature field distribution.
[0085] Due to the influence of thermal expansion of materials, the stress characteristics of the elastic electrical connection structure will change after the temperature changes. The contact resistance of the electrical connection structure is updated according to the calculated temperature field distribution. At the same time, the characteristics of the change of metal material conductivity with the temperature field are considered in the electromagnetic field calculation. The electromagnetic field equation is re-solved to obtain a new loss distribution, and the thermal-fluid coupling calculation results are updated until the calculation results of the two temperature field distributions have an error less than the set value. It can be judged that the convergence state has been reached. At this time, the high-precision calculation results of the temperature field distribution of the bushing on the converter transformer valve side are obtained, that is, the high-precision data set.
[0086] Since the process of acquiring high-precision data sets is relatively complicated, generally measured in days, it can be concluded that the calculation time of low-precision data is very fast compared to high-precision data sets. The combination of high- and low-precision data can shorten the data set acquisition time, thereby reducing the modeling cost.
[0087] The tolerance factor is expressed as:
[0088]
[0089] In the formula, represents the sample space, represents the tolerance factor, Represents high-precision data, Represents the basic data-driven model, where ∈U H .
[0090] Step S602: construct a second training data set according to the multiple groups of sample spaces and the tolerance factors corresponding to the multiple groups of sample spaces.
[0091] Among them, through step S601, it can be obtained that the sample space and the tolerance factor are in a mapping relationship, and the second training data set is constructed based on the mapping relationship between the sample space and the tolerance factor.
[0092] Step S603: training the second training data set based on a deep neural network to obtain a tolerance factor prediction regression model.
[0093] In the training process, the deep neural network is trained with the sample space as input and the tolerance factor as output to obtain a tolerance factor prediction regression model.
[0094] Step S604: modify the basic data-driven model according to the tolerance factor prediction regression model to obtain a multi-precision proxy model.
[0095] Among them, the multi-precision proxy model is:
[0096]
[0097] In the formula, is the output of the multi-precision proxy model, Predictive regression model for tolerance factor.
[0098] In order to verify the accuracy of the output results of the multi-precision proxy model, this embodiment also includes:
[0099] S701. Perform error testing on the output results of the multi-precision proxy model.
[0100] S702, if the error test result of the output result of the multi-precision proxy model is qualified, then the multi-precision proxy model is output;
[0101] S703. If the error test result of the output result of the multi-precision proxy model fails, go to step S3 to iterate until the error test result of the output result of the multi-precision proxy model passes, stop iterating and output the multi-precision proxy model.
[0102] Exemplarily, a set of data outside the constructed proxy model is assumed to be sample space a, a set of grid node data b is obtained through finite element calculation, and another set of grid node data c is obtained through multi-precision proxy model calculation. The difference between each corresponding grid node data in the grid node data b and the grid node data c is obtained, and it is determined whether the difference is less than a preset error value. If the difference is less than the preset error value, the error test result of the output result of the multi-precision proxy model is determined to be qualified; otherwise, the error test result of the output result of the multi-precision proxy model is determined to be unqualified.
[0103] If the error test result of the output result of the multi-precision proxy model is determined to be unqualified, the network parameters of the tolerance factor prediction regression model can also be optimized. That is, according to the convergence index, the temperature field distribution result obtained by the multi-precision proxy model is differentiated by a differentiator to obtain the error value of the corresponding simulation calculation model convergence index, that is, the physical information error.
[0104] Taking the tolerance factor as the observation data error, combined with the physical information error and sample space, a loss function is constructed. The loss function is used to analyze and adjust the weights and structure of the tolerance factor prediction regression model so that the loss function is minimized or the convergence condition requirements are met. The tolerance factor that considers the physical information constraints is obtained, and a more accurate multi-precision proxy model is obtained.
[0105] The above is a detailed description of an embodiment of a multi-physics field computing method for online simulation of a digital twin provided by the present invention. The following is a detailed description of an embodiment of a multi-physics field computing system for online simulation of a digital twin provided by the present invention.
[0106] like Figure 4 As shown, Figure 4 The structure of a multi-physics field computing system for online simulation of a digital twin provided by an embodiment of the present invention is illustrated.
[0107] Some embodiments of the present invention further provide a multi-physics field computing system for online simulation of digital twins, including:
[0108] The simulation modeling module 100 is used to establish a multi-physics field coupling simulation calculation model of the simulated object and to simplify the multi-physics field coupling simulation calculation model.
[0109] Specifically, a multi-physical field coupling simulation calculation model is constructed according to the geometric structure of the simulated object, wherein the physical field includes a temperature field, a stress field and a fluid field that are coupled to each other.
[0110] The coupled mesh division between the temperature field and the stress field is simplified, and the convection heat transfer process of the fluid field is equivalent to the heat conduction process.
[0111] It should be noted that, affected by the thermal expansion of the material, the stress characteristics of the elastic electrical connection structure will change after the temperature changes. Considering this problem, the difficulty of meshing is increased, which will lead to an unnecessary increase in the number of meshes and increase the amount of calculation. For this reason, this embodiment simplifies the coupling simulation process between the temperature field and the stress field by dividing the coupled meshes between the temperature field and the stress field, so as not to consider the influence of the temperature field distribution on the thermal expansion of the material, thereby reducing the difficulty of meshing and reducing the amount of calculation.
[0112] Among them, the convective heat transfer process is not easy to obtain. In this embodiment, the convective heat transfer process of the fluid field is equivalent to a heat conduction process, thereby reducing the amount of calculation.
[0113] The model order reduction module 200 is used to determine the temperature field simulation model according to the simplified multi-physics field coupling simulation calculation model, and reduce the order of the temperature field simulation model to obtain the temperature field simulation reduced-order model.
[0114] In some specific embodiments, an intrinsic orthogonal decomposition method is used to reduce the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model.
[0115] Specifically, the temperature field simulation model can be characterized by the temperature field finite element discrete equation. The temperature field simulation reduced-order model can map the temperature field calculation nodes to a set of orthogonal basis eigenvectors to form a low-dimensional space to further improve the calculation efficiency of the model. Through research, it is found that the single simulation calculation time of the temperature field simulation reduced-order model can be controlled within seconds.
[0116] The temperature field analysis module 300 is used to perform temperature field analysis based on a preset sample space and a temperature field simulation reduced-order model to obtain a low-precision data set.
[0117] The driving model construction module 400 is used to construct a basic data-driven model based on a low-precision data set and a sample space corresponding to the low-precision data set. The basic data-driven model is used to output a low-precision data set corresponding to the sample space after the sample space is input into the basic data-driven model.
[0118] Among them, the basic data-driven model can adopt one of the methods such as neural network, response surface method and Kriging method. By using the above methods to establish the basic data-driven model respectively, and performing error tests on the basic data-driven model, the method with the smallest error is selected to construct the basic data-driven model.
[0119] The data driving module 500 is used to input the current sample space into the basic data driving model and output the low-precision data set corresponding to the current sample space.
[0120] like Figure 5 As shown, some embodiments of the present invention further provide an electronic device, where the electronic device 10 includes a memory 20 and a processor 30;
[0121] The memory 20 is used to store programs;
[0122] The processor 30 executes the program to implement the steps of the multi-physics field calculation method for online simulation of digital twins in any of the above embodiments.
[0123] Some embodiments of the present invention also provide a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the multi-physics field calculation method for online simulation of digital twins in any of the above embodiments.
[0124] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, electronic device and computer storage medium can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0125] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a part of a module, a program segment or a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.
[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk, etc. Various media that can store program codes.
[0130] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A multi-physics field calculation method for online simulation of digital twins, It is characterized in that include: Establishing a multi-physics field coupling simulation calculation model of the simulated object and simplifying the multi-physics field coupling simulation calculation model specifically includes: Constructing a multi-physics field coupling simulation calculation model according to the geometric structure of the simulated object, wherein the physical fields include mutually coupled temperature field, stress field and fluid field; Simplifying the coupled grid division between the temperature field and the stress field, and equating the convection heat transfer process of the fluid field to a heat conduction process; Determine a temperature field simulation model according to the simplified multi-physics field coupling simulation calculation model, and reduce the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model; Based on a preset sample space, performing temperature field analysis according to the temperature field simulation reduced-order model to obtain a low-precision data set; Constructing a basic data-driven model according to the low-precision data set and the sample space corresponding to the low-precision data set, wherein the basic data-driven model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data-driven model; The current sample space is input into the basic data driven model, and a low-precision data set corresponding to the current sample space is output.
2. The multi-physics field calculation method for online simulation of digital twins according to claim 1, It is characterized in that Also includes: The temperature field simulation model is reduced in order by using an intrinsic orthogonal decomposition method to obtain a temperature field simulation reduced-order model.
3. The multi-physics field calculation method for online simulation of digital twins according to claim 1, It is characterized in that The step of constructing a basic data-driven model according to the low-precision data set and the sample space corresponding to the low-precision data set, wherein the basic data-driven model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data-driven model, specifically includes: Constructing a first training data set according to the low-precision data set and a sample space corresponding to the low-precision data set; An initial basic data-driven model is constructed, and the initial basic data-driven model is trained using the first training data set, and after the training converges, a basic data-driven model is obtained.
4. The multi-physics field calculation method for online simulation of digital twins according to claim 1, It is characterized in that Also includes: The basic data driven model is modified.
5. The multi-physics field calculation method for online simulation of digital twins according to claim 4, It is characterized in that The step of correcting the basic data driven model specifically includes: Determine multiple groups of sample spaces and tolerance factors corresponding to the multiple groups of sample spaces according to the basic data-driven model and the gap between the high-precision data set and the low-precision data set; Constructing a second training data set according to a plurality of groups of sample spaces and tolerance factors corresponding to the plurality of groups of sample spaces; Training the second training data set based on a deep neural network to obtain a tolerance factor prediction regression model; The basic data driven model is modified according to the tolerance factor prediction regression model to obtain a multi-precision proxy model.
6. The multi-physics field calculation method for online simulation of digital twins according to claim 5, It is characterized in that Also includes: Performing an error test on the output result of the multi-precision proxy model; If the error test result of the output result of the multi-precision proxy model is qualified, outputting the multi-precision proxy model; If the error test result of the output result of the multi-precision proxy model fails to meet the requirements, the step of switching to the preset sample space, performing temperature field analysis according to the temperature field simulation reduced-order model, and iterating the step of obtaining a low-precision data set is performed until the error test result of the output result of the multi-precision proxy model meets the requirements, stopping the iteration and outputting the multi-precision proxy model.
7. A multi-physics field computing system for online simulation of digital twins, It is characterized in that include: A simulation modeling module is used to establish a multi-physics field coupling simulation calculation model of a simulated object and simplify the multi-physics field coupling simulation calculation model; Establishing a multi-physics field coupling simulation calculation model of the simulated object and simplifying the multi-physics field coupling simulation calculation model specifically includes: Constructing a multi-physics field coupling simulation calculation model according to the geometric structure of the simulated object, wherein the physical fields include mutually coupled temperature field, stress field and fluid field; Simplifying the coupled grid division between the temperature field and the stress field, and equating the convection heat transfer process of the fluid field to a heat conduction process; A model reduction module is used to determine a temperature field simulation model according to the simplified multi-physics field coupling simulation calculation model, and reduce the order of the temperature field simulation model to obtain a temperature field simulation reduced-order model; A temperature field analysis module, used to perform temperature field analysis based on a preset sample space and the temperature field simulation reduced-order model to obtain a low-precision data set; A driving model construction module, used to construct a basic data driving model according to the low-precision data set and the sample space corresponding to the low-precision data set, wherein the basic data driving model is used to output the low-precision data set corresponding to the sample space after the sample space is input into the basic data driving model; The data driving module is used to input the current sample space into the basic data driving model and output the low-precision data set corresponding to the current sample space.
8. An electronic device, It is characterized in that The electronic device comprises a memory and a processor; The memory is used to store programs; The processor executes the program to implement the steps of the multi-physics field calculation method for digital twin online simulation described in any one of claims 1 to 6.
9. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the steps of the multi-physics field calculation method for digital twin online simulation described in any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Multi-physics field time domain non-intrusive model reduction method and system based on data driving
CN117473826A
Method and device for constructing multi-physics field digital twinborn simulation model of power equipment
CN118468983A
Multi-physics co-simulation method of power semiconductor modules
US20230057941A1
Cited By
Digital twinning-oriented power grid multi-scale-multi-physical field real-time coupling simulation system
CN122527530A
