Method, device and equipment for obtaining three-dimensional simulation sample of thrust chamber

Abstract

The invention particularly relates to a method, a device and equipment for acquiring a three-dimensional simulation sample of a thrust chamber, relates to the technical field of liquid rocket engines, and aims to solve the problem of low acquisition efficiency of the three-dimensional simulation sample of the thrust chamber in the prior art. Comprising the following steps: for a thrust chamber in a liquid rocket engine, acquiring a first type of three-dimensional simulation results under a first target number of working condition points; obtaining a second type of three-dimensional simulation result under a second target number of working condition points; the first target number of operating points is at least ten times of the second target number of operating points; and performing data fusion on the first type of three-dimensional simulation results under the first target number of working condition points and the second type of three-dimensional simulation results under the second target number of working condition points to obtain thrust chamber three-dimensional simulation samples under the first target number of working condition points. Therefore, the acquisition efficiency of the three-dimensional simulation sample of the thrust chamber can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid rocket engines, and particularly to a method, device, and equipment for obtaining three-dimensional simulation samples of a thrust chamber. Background Art

[0002] In the digital twin of a liquid rocket engine, it is necessary to construct a twin body for the thrust chamber, the core component of the liquid rocket engine. Through the twin body of the thrust chamber, the performance of the thrust chamber can be monitored and analyzed in real time.

[0003] In the process of constructing the twin body of the thrust chamber, first, the three-dimensional simulation results of the thrust chamber need to be obtained through three-dimensional modeling and simulation technology. The simulation results at least include the pressure field, temperature field, and Mach number. The three-dimensional simulation results reflect the performance problems of the thrust chamber. After obtaining the three-dimensional simulation results, a reduction technology is adopted to reduce the three-dimensional simulation results to a three-dimensional feature model.

[0004] In the process of repeatedly using three-dimensional modeling and simulation calculations to obtain a large number of three-dimensional simulation results of the thrust chamber, for example, calculating 180 times, the three-dimensional simulation results at 180 different operating points can be obtained. Calculated according to, for example, 20 million grids, a single simulation calculation takes about 5 hours. Preparing the operating conditions, starting the simulation calculation, and collecting the three-dimensional simulation results once a day, it takes about 200 days to complete the three-dimensional simulation calculations at these 180 different operating points. This process consumes too much time and incurs a lot of labor costs. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, device, and equipment for obtaining three-dimensional simulation samples of a thrust chamber, which is used to improve the acquisition efficiency of three-dimensional simulation samples of the thrust chamber.

[0006] To achieve the above purpose, the present invention provides a method, device, and equipment for obtaining three-dimensional simulation samples of a thrust chamber, as follows:

[0007] In a first aspect, the present invention provides a method for obtaining three-dimensional simulation samples of a thrust chamber, including:

[0008] For the thrust chamber in a liquid rocket engine, obtain the first type of three-dimensional simulation results at the first target number of operating points;

[0009] Obtain the second type of three-dimensional simulation results at the second target number of operating conditions; wherein, the first type of three-dimensional simulation results indicate that the first actual grid number used in the simulation calculation is less than or equal to the first preset grid number, and the second type of three-dimensional simulation results indicate that the second actual grid number used in the simulation calculation is greater than or equal to the second preset grid number; the first preset grid number is a grid number in the hundred thousands level, and the second preset grid number is a grid number in the tens of millions level; the first target number of operating conditions is at least ten times the second target number of operating conditions;

[0010] Perform data fusion on the first type of three-dimensional simulation results at the first target number of operating conditions and the second type of three-dimensional simulation results at the second target number of operating conditions to obtain the three-dimensional simulation samples of the thrust chamber at the first target number of operating conditions.

[0011] Optionally, the obtaining of the first type of three-dimensional simulation results at the first target number of operating conditions includes:

[0012] Obtain historical three-dimensional simulation data of the thrust chamber, and obtain the first type of three-dimensional simulation results at the first number of operating conditions from the historical three-dimensional simulation data of the thrust chamber;

[0013] If the first number is less than the first target number, determine the first difference between the first target number and the first number;

[0014] Based on the first actual grid number, perform three-dimensional simulation of the thrust chamber at the first difference number of operating conditions to obtain the first type of three-dimensional simulation results at the first difference number of operating conditions;

[0015] Sum up the first type of three-dimensional simulation results at the first difference number of operating conditions and the first type of three-dimensional simulation results at the first number of operating conditions to obtain the first type of three-dimensional simulation results at the first target number of operating conditions.

[0016] Optionally, the obtaining of the second type of three-dimensional simulation results at the second target number of operating conditions includes:

[0017] Obtain the second type of three-dimensional simulation results at the second number of operating conditions from the historical three-dimensional simulation data of the thrust chamber;

[0018] If the second number is less than the second target number, determine the second difference between the second target number and the second number;

[0019] Based on the second actual grid number, perform three-dimensional simulation of the thrust chamber at the second difference number of operating conditions to obtain the second type of three-dimensional simulation results at the second difference number of operating conditions;

[0020] Sum up the second type of three-dimensional simulation results at the second difference number of operating points and the second type of three-dimensional simulation results at the second number of operating points to obtain the second type of three-dimensional simulation results at the second target number of operating points.

[0021] Optionally, the data fusion of the first type of three-dimensional simulation results at the first target number of operating points and the second type of three-dimensional simulation results at the second target number of operating points includes:

[0022] Adopt a multi-source data fusion algorithm based on transfer learning to fuse the first type of three-dimensional simulation results at the first target number of operating points and the second type of three-dimensional simulation results at the second target number of operating points.

[0023] Optionally, the method for obtaining the three-dimensional simulation samples of the thrust chamber further includes:

[0024] Reduce the order of the three-dimensional simulation samples of the thrust chamber at the first target number of operating points.

[0025] Compared with the prior art, a method for obtaining three-dimensional simulation samples of a thrust chamber provided by the present invention, in order to obtain the three-dimensional simulation samples of the thrust chamber at the first target number of operating points, fuses the first type of three-dimensional simulation results at the same number of operating points as the first target number with the second type of three-dimensional simulation results at the second target number of operating points less than the first target number, which can reduce a large amount of three-dimensional simulation calculation time and improve the acquisition rate and acquisition efficiency of the three-dimensional simulation samples of the thrust chamber.

[0026] In a second aspect, the present invention provides an apparatus for obtaining three-dimensional simulation samples of a thrust chamber, including:

[0027] A first acquisition module, configured to acquire the first type of three-dimensional simulation results at the first target number of operating points for the thrust chamber in a liquid rocket engine;

[0028] A second acquisition module, configured to acquire the second type of three-dimensional simulation results at the second target number of operating points; wherein, the first type of three-dimensional simulation results indicate that the first actual grid number used in the simulation calculation is less than or equal to the first preset grid number, and the second type of three-dimensional simulation results indicate that the second actual grid number used in the simulation calculation is greater than or equal to the second preset grid number; the first preset grid number is a grid number in the hundred-thousand level, and the second preset grid number is a grid number in the ten-million level; the first target number of operating points is at least ten times the second target number of operating points;

[0029] A fusion module for fusing the first type of three-dimensional simulation results at the first target number of operating points and the second type of three-dimensional simulation results at the second target number of operating points to obtain three-dimensional simulation samples of the thrust chamber at the first target number of operating points.

[0030] Optionally, the first acquisition module is specifically configured to acquire historical three-dimensional simulation data of the thrust chamber, and obtain the first type of three-dimensional simulation results at the first number of operating points from the historical three-dimensional simulation data of the thrust chamber;

[0031] If the first number is less than the first target number, determine a first difference between the first target number and the first number;

[0032] Based on the first actual grid number, perform three-dimensional simulation of the thrust chamber at the first difference number of operating points to obtain the first type of three-dimensional simulation results at the first difference number of operating points;

[0033] Sum the first type of three-dimensional simulation results at the first difference number of operating points and the first type of three-dimensional simulation results at the first number of operating points to obtain the first type of three-dimensional simulation results at the first target number of operating points.

[0034] Optionally, the second acquisition module is specifically configured to obtain the second type of three-dimensional simulation results at the second number of operating points from the historical three-dimensional simulation data of the thrust chamber;

[0035] If the second number is less than the second target number, determine a second difference between the second target number and the second number;

[0036] Based on the second actual grid number, perform three-dimensional simulation of the thrust chamber at the second difference number of operating points to obtain the second type of three-dimensional simulation results at the second difference number of operating points;

[0037] Sum the second type of three-dimensional simulation results at the second difference number of operating points and the second type of three-dimensional simulation results at the second number of operating points to obtain the second type of three-dimensional simulation results at the second target number of operating points.

[0038] In a third aspect, the present invention provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus; wherein, the processor, the communication interface, and the memory communicate through the communication bus; a computer program that can be run by the processor is stored on the memory; when the processor runs the computer program, it executes the method for obtaining three-dimensional simulation samples of the thrust chamber described in any one of the above.

[0039] Fourthly, the present invention provides a computer storage medium storing instructions which, when run, implement the method for obtaining a three-dimensional simulation sample of a thrust chamber as described in any one of the above BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present invention and form a part of the present invention. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0041] Figure 1 is a schematic flowchart of a method for obtaining a three-dimensional simulation sample of a thrust chamber provided by an embodiment of the present invention;

[0042] Figure 2 is a schematic structural diagram of a device for obtaining a three-dimensional simulation sample of a thrust chamber provided by an embodiment of the present invention;

[0043] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] For the convenience of clearly describing the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. For example, the first threshold and the second threshold are only used to distinguish different thresholds and do not limit their sequence. Those skilled in the art can understand that the terms "first" and "second" do not limit the quantity and execution order, and the terms "first" and "second" do not necessarily mean different.

[0045] It should be noted that in the present invention, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific manner.

[0046] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist.

[0047] As Figure 1 shown, an embodiment of the present invention provides a method for obtaining a three-dimensional simulation sample of a thrust chamber, which may include:

[0048] Step 110: For the thrust chamber in a liquid rocket engine, obtain first-class three-dimensional simulation results at a first target number of operating points;

[0049] Before performing the simulation calculation, it is first necessary to construct a three-dimensional geometric model of the thrust chamber, and three simulation models for the thrust chamber are constructed. The model construction can be carried out using any one of multiple simulation software such as ANSYS ICEM, UG, and CATIA. After the model construction is successful, the simulation software is used to perform mesh division on the three-dimensional simulation model of the thrust chamber. One is to use low-quality meshes, that is, to perform mesh division on the three-dimensional simulation model of the thrust chamber using a mesh number in the order of one hundred thousand, and the other is to use high-quality meshes, that is, to perform mesh division on the three-dimensional simulation model of the thrust chamber using a mesh number in the order of ten million. After the mesh division, some simulation conditions and other parameters are set. Finally, simulation calculations are carried out for the thrust chamber under different numbers of working conditions, and three-dimensional simulation results under different numbers of working conditions are obtained. The three-dimensional simulation results under each working condition include a large amount of simulation result data. The simulation results can be the three-dimensional simulation results of the thrust chamber pressure or the three-dimensional simulation results of the thrust chamber Mach number.

[0050] Step 120: Obtain the second type of three-dimensional simulation results under the second target number of working condition points; wherein, the first type of three-dimensional simulation results characterize that the first actual mesh number used during the simulation calculation is less than or equal to the first preset mesh number, and the second type of three-dimensional simulation results characterize that the second actual mesh number used during the simulation calculation is greater than or equal to the second preset mesh number; the first preset mesh number is a mesh number in the order of one hundred thousand, and the second preset mesh number is a mesh number in the order of ten million; the first target number of working condition points is at least ten times the second target number of working condition points;

[0051] For example, the second target number can be 10.

[0052] For example, the first target number is 200.

[0053] It can be understood that the simulation calculation process corresponding to the first type of three-dimensional simulation results uses a mesh number in the order of one hundred thousand, and the simulation calculation process corresponding to the second type of three-dimensional simulation results uses a mesh number in the order of ten million.

[0054] For example, the first preset mesh number is 900,000 meshes, and the first actual mesh number used during the simulation calculation is 200,000 meshes.

[0055] For example, the second preset mesh number is 90 million meshes, and the second actual mesh number used during the simulation calculation is 20 million meshes.

[0056] Step 130: Perform data fusion on the first type of three-dimensional simulation results under the first target number of working condition points and the second type of three-dimensional simulation results under the second target number of working condition points to obtain a three-dimensional simulation sample of the thrust chamber under the first target number of working condition points.

[0057] As can be seen from the above, in order to obtain three-dimensional simulation samples of the thrust chamber at the first target number (e.g., 200) of operating conditions, by fusing the first type of three-dimensional simulation results at the same number of operating conditions as the first target number with the second type of three-dimensional simulation results at the second target number of operating conditions that is less than the first target number, this can reduce a relatively large amount of three-dimensional simulation calculation time and improve the acquisition rate and efficiency of the three-dimensional simulation samples of the thrust chamber. This is because the first target number of operating conditions is at least ten times the second target number of operating conditions. If the number of the first type of three-dimensional simulation results and the second type of three-dimensional simulation results is the same, due to the large number of grids used in the second type of three-dimensional simulation results and the long calculation time, in this embodiment, a small number of the second type of three-dimensional simulation results are obtained, and then the small number of three-dimensional simulation results are fused with the first type of three-dimensional simulation results at the target number, which can greatly reduce the calculation time and improve the acquisition efficiency of the three-dimensional simulation samples of the thrust chamber.

[0058] In an alternative embodiment, step 110 may include:

[0059] 1) Obtain historical three-dimensional simulation data of the thrust chamber, and obtain the first type of three-dimensional simulation results at the first number of operating conditions from the historical three-dimensional simulation data of the thrust chamber;

[0060] In a specific implementation, the database stores the data of previous multiple simulations, that is, the historical three-dimensional simulation data of the thrust chamber. These data contain both the simulation result data corresponding to a certain number of the first type of three-dimensional simulation results and the simulation result data corresponding to a certain number of the second type of three-dimensional simulation results.

[0061] For example, the first number is 80.

[0062] 2) If the first number is less than the first target number, determine the first difference between the first target number and the first number;

[0063] For example, the first target number is 200 and the first number is 80, then the first difference is 120.

[0064] 3) Based on the first actual number of grids, perform three-dimensional simulation of the thrust chamber at the first difference number of operating conditions to obtain the first type of three-dimensional simulation results at the first difference number of operating conditions;

[0065] For example, the first actual number of grids is 200,000. Then, divide the above-mentioned three-dimensional simulation model of the thrust chamber according to 200,000 grids, and then perform 120 simulation calculations to obtain the first type of three-dimensional simulation results at 120 operating conditions.

[0066] 4) Sum up the first type of three-dimensional simulation results at the first difference number of operating points and the first type of three-dimensional simulation results at the first number of operating points to obtain the first type of three-dimensional simulation results at the first target number of operating points.

[0067] For example, sum up the first type of three-dimensional simulation results at 120 operating points and the first type of three-dimensional simulation results at 80 operating points obtained from historical three-dimensional simulation data of the thrust chamber, that is, obtain the first type of three-dimensional simulation results at the required 200 operating points.

[0068] In an alternative implementation, similar to step 110, step 120 may include:

[0069] Obtain the second type of three-dimensional simulation results at the second number of operating points from historical three-dimensional simulation data of the thrust chamber;

[0070] If the second number is less than the second target number, determine the second difference between the second target number and the second number;

[0071] Based on the second actual mesh number, perform three-dimensional simulation of the thrust chamber at the second difference number of operating points to obtain the second type of three-dimensional simulation results at the second difference number of operating points;

[0072] Sum up the second type of three-dimensional simulation results at the second difference number of operating points and the second type of three-dimensional simulation results at the second number of operating points to obtain the second type of three-dimensional simulation results at the second target number of operating points.

[0073] For example, if the second type of three-dimensional simulation results at 10 operating points are required and 4 second type of three-dimensional simulation results are obtained from historical three-dimensional simulation data of the thrust chamber, then the three-dimensional simulation model of the thrust chamber needs to be meshed according to the second actual mesh number (for example, 20 million), and then 6 simulation calculations are performed to obtain the second type of three-dimensional simulation results at 6 operating points; sum up the second type of three-dimensional simulation results at 6 operating points and the 4 second type of three-dimensional simulation results obtained from historical three-dimensional simulation data of the thrust chamber, that is, obtain the second type of three-dimensional simulation results at the required 10 operating points.

[0074] In an alternative implementation, step 130 may include:

[0075] Adopt a multi-source data fusion algorithm based on transfer learning to fuse the first type of three-dimensional simulation results at the first target number of operating points and the second type of three-dimensional simulation results at the second target number of operating points.

[0076] Specifically, the core of transfer learning lies in the mechanism of knowledge transfer.

[0077] In an alternative embodiment, transfer learning can adopt feature transfer. Feature transfer includes feature transformation and feature selection.

[0078] Feature transformation: The feature representation of the source domain is transformed into a form usable in the target domain through feature transformation. This can be achieved by learning a feature transformation function that can map the features of the source domain to the feature space of the target domain. Such a transformation can be linear or non-linear, depending on the complexity between the source domain and the target domain.

[0079] Feature selection is another way of feature transfer. It is based on the assumption that only some features are shared between the source domain and the target domain. Therefore, transfer can be carried out by selecting the features useful for the target task. Feature selection can be achieved through statistical methods, machine learning algorithms or deep learning methods.

[0080] Pre-trained feature extractor: In deep learning, a common way of feature transfer is to use a pre-trained convolutional neural network (CNN) or recurrent neural network (RNN) as a feature extractor. These networks have been trained on large datasets and have learned to extract general features of data such as images, text or audio. These features can be used for classification, regression or generation tasks in the target task.

[0081] In an alternative embodiment, transfer learning can adopt model transfer. Model transfer can further include fine-tuning and multi-task learning.

[0082] Fine-tuning transfers the model parameters trained in the source domain to the target domain and fine-tunes them on the data of the target domain. During the fine-tuning process, usually only the parameters of some layers are updated, while the parameters of the remaining layers are frozen. This method utilizes the knowledge in the source domain model and adapts to the data distribution of the target domain through fine-tuning.

[0083] Multi-task learning is to learn multiple related tasks simultaneously and utilize the correlation between tasks by sharing network layers or parameters. In multi-task learning, the source domain and the target domain are regarded as different tasks and share part of the network structure or parameters. This method improves the generalization ability of the model by optimizing multiple tasks simultaneously.

[0084] To achieve transfer learning, we use a portion of the labeled training data that has the same distribution as the test data to build a classification model. We call this training data the same-distribution training data. The amount of this same-distribution training data is often insufficient to train a good classifier. The distribution of the training data may be different from that of the test data, which may be because they are outdated. These training data are called different-distribution training data. Suppose these data are very abundant, but due to the different data distributions, the classifier learned from these data cannot classify the test data well. More formally, let X s be the same-distribution instance space, X a be the different-distribution instance space, and Y = {0, 1} be the set of class labels. A concept is a Boolean function c that maps from X to Y, where X = X s ∪X d .

[0085] The test data set is denoted by , where here k is the size of the unlabeled test set S. The training data set tc{X×Y} is divided into two labeled sets T a and T s . T a represents the different-distribution training data of , where T s represents the same-distribution training data, where n and m are the sizes of T a and T s respectively. c(x) returns the label of the data instance x. The combined training set T = {(x i , c(x i ))} is defined as follows

[0086]

[0087] Here, T a corresponds to some labeled data from the old domain, and reuse as much of this data as possible; when it is not known which part of T a is available, label a small portion of the data from the new domain and call it t, and then use this data to find the useful part of T a . The problem to be solved is: Given a small amount of labeled same-distribution training data, a large number of different-distribution training data T a and some unlabeled test data S, the goal is to train a classifier e: X → Y that minimizes the prediction error on the unlabeled data set S.

[0088] Transfer the AdaBoost learning framework TrAdaBoost, which extends the transfer learning of AdaBoost. AdaBoost (Freund & Schapire, 1997) is a learning framework that aims to improve the accuracy of weak learners by carefully adjusting the weights of training instances and accordingly learn a classifier. However, similar to most traditional machine learning methods, AdaBoost assumes that the distributions of training data and test data are the same. In the extension of AdaBoost, AdaBoost is still applied to the training data with the same distribution to build the basis of the model. However, for the training instances with different distributions, when they are mispredicted due to the distribution change of the learning model, these instances may be those that are least similar to the instances with the same distribution. Therefore, a mechanism can be added in the extension to reduce the weights of these instances to weaken their influence.

[0089]

[0090]

[0091] Algorithm 1 gives the formal description of this framework. It can be seen from the algorithm that in each round of iteration, if a training instance with different distributions is mispredicted, then this instance is likely to conflict with the training data with the same distribution. Then, by multiplying its weight by to reduce its training weight to reduce its effect, note that Therefore, in the next round, the influence of the misclassified training instances with different distributions that are different from the training instances with the same distribution on the learning process will be less than that in the current round. After multiple iterations, the training weights of the training instances with different distributions that fit well with the training instances with the same distribution are larger, while the training weights of the training instances with different distributions that are not similar to the training instances with the same distribution are smaller. The postures with larger training weights will help the learning algorithm train a better classifier.

[0092] Next, the intensity characteristics will be discussed and it will be explained why the framework can learn knowledge even when the domain distributions are not the same.

[0093] Let be the loss caused by the hypothesis ht to the training instance x. The distribution d t is the training weight vector with respect to T a in the nth iteration, After N iterations, the training loss of TrAdaBoost with respect to T a is For the training loss of the instance x i (i = 1,..., n), after N iterations, the loss suffered by TrAdaBoost is

[0094] The data corresponding to the first type of three-dimensional simulation results can be regarded as low-fidelity data, and the data corresponding to the second type of three-dimensional simulation results can be regarded as high-fidelity data; therefore, when fusing the first type of three-dimensional simulation results and the second type of three-dimensional simulation results, the above-mentioned multi-source data fusion algorithm based on transfer learning can be used to fuse multi-fidelity data.

[0095] In an alternative embodiment, the method for obtaining the three-dimensional simulation samples of the thrust chamber may further include:

[0096] Reduce the order of the three-dimensional simulation samples of the thrust chamber at the first target number of operating points to obtain the required three-dimensional simulation target model of the thrust chamber.

[0097] In specific implementation, if traditional technologies are used, if three-dimensional simulation results at, for example, 200 different operating points are to be obtained, 200 simulation calculations are required. Each simulation calculation is based on a grid quantity of 20 million. Each single simulation calculation takes about 6 hours. With the preparation of operating conditions, starting the simulation calculation, and collecting simulation results once a day, it takes about 200 days to complete the three-dimensional simulation calculations at 200 different operating points. This process requires too much time and manpower. However, by using the solution in the embodiment of the present invention, the simulation calculation workload that originally took 200 days can be shortened to within 30 days, and a large number of three-dimensional simulation calculation results can be obtained quickly, saving a large amount of manpower and physical costs.

[0098] As Figure 2 shown, the embodiment of the present invention further provides a device for obtaining three-dimensional simulation samples of a thrust chamber, which may include:

[0099] A first acquisition module 210, configured to acquire the first type of three-dimensional simulation results at the first target number of operating points for the thrust chamber in a liquid rocket engine;

[0100] A second acquisition module 220, configured to acquire the second type of three-dimensional simulation results at the second target number of operating points; wherein, the first type of three-dimensional simulation results indicate that the first actual grid number used in the simulation calculation is less than or equal to the first preset grid number, and the second type of three-dimensional simulation results indicate that the second actual grid number used in the simulation calculation is greater than or equal to the second preset grid number; the first preset grid number is a grid number in the hundred-thousand level, and the second preset grid number is a grid number in the ten-million level; the first target number of operating points is at least ten times the second target number of operating points;

[0101] A fusion module 230 is configured to perform data fusion on the first type of three-dimensional simulation results at the first target number of operating points and the second type of three-dimensional simulation results at the second target number of operating points to obtain a three-dimensional simulation sample of the thrust chamber at the first target number of operating points.

[0102] In an alternative embodiment, the first acquisition module 210 is specifically configured to:

[0103] Acquire historical three-dimensional simulation data of the thrust chamber, and obtain the first type of three-dimensional simulation results at the first number of operating points from the historical three-dimensional simulation data of the thrust chamber;

[0104] If the first number is less than the first target number, determine a first difference between the first target number and the first number;

[0105] Based on the first actual grid number, perform three-dimensional simulation of the thrust chamber at the first difference number of operating points to obtain the first type of three-dimensional simulation results at the first difference number of operating points;

[0106] Sum up the first type of three-dimensional simulation results at the first difference number of operating points and the first type of three-dimensional simulation results at the first number of operating points to obtain the first type of three-dimensional simulation results at the first target number of operating points.

[0107] In an alternative embodiment, the second acquisition module 220 is specifically configured to:

[0108] Obtain the second type of three-dimensional simulation results at the second number of operating points from the historical three-dimensional simulation data of the thrust chamber; if the second number is less than the second target number, determine a second difference between the second target number and the second number; based on the second actual grid number, perform three-dimensional simulation of the thrust chamber at the second difference number of operating points to obtain the second type of three-dimensional simulation results at the second difference number of operating points; sum up the second type of three-dimensional simulation results at the second difference number of operating points and the second type of three-dimensional simulation results at the second number of operating points to obtain the second type of three-dimensional simulation results at the second target number of operating points.

[0109] In an alternative embodiment, the fusion module 230 is specifically configured to:

[0110] Obtain a pre-configured co-kriging model; based on the co-kriging model, fuse the first type of three-dimensional simulation results at the first target number of operating points and the second type of three-dimensional simulation results at the second target number of operating points.

[0111] In an alternative embodiment, the device for obtaining the three-dimensional simulation sample of the thrust chamber further includes a reduction module, and the reduction module is configured to reduce the order of the three-dimensional simulation sample of the thrust chamber at the first target number of operating points.

[0112] As shown Figure 3 in the figure, an embodiment of the present invention further provides an electronic device, which may include: a processor 310, a communication interface 320, a memory 330, and a communication bus. Among them, the processor 310, the communication interface 320, and the memory 330 complete mutual communication through the communication bus. A computer program that can be run by the processor 310 is stored on the memory 330; when the processor 310 runs the computer program, it can execute the method for obtaining the three-dimensional simulation sample of the thrust chamber in any of the above embodiments. In addition, when the logical instructions in the above-mentioned memory 330 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0113] On the other hand, the present invention also provides a non-transitory computer-readable storage medium. Instructions are stored in the computer storage medium, and when the instructions are run, the method for obtaining the three-dimensional simulation sample of the thrust chamber in any of the above embodiments is implemented.

[0114] Although the present invention has been described in connection with various embodiments, those skilled in the art will recognize other variations of the disclosed embodiments while practicing the claimed invention by viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may implement several functions recited in the claims. Certain measures are recited in mutually different dependent claims, but this does not indicate that these measures cannot be combined to produce favorable results. Although the present invention has been described in connection with specific features and embodiments, it will be apparent that various modifications and combinations can be made without departing from the spirit and scope of the invention. Accordingly, the specification and drawings are merely exemplary illustrations of the invention defined by the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for obtaining a three-dimensional simulation sample of a thrust chamber, characterized in that: include: For the thrust chamber in the liquid rocket engine, obtain the first type of three-dimensional simulation results under the first target number of operating points; Obtain a second type of three-dimensional simulation results under a second target number of operating points; wherein, the first type of three-dimensional simulation results indicates that the first actual number of grids used in the simulation calculation is less than or equal to the first preset number of grids, and the second type of three-dimensional simulation results indicates that the second actual number of grids used in the simulation calculation is greater than or equal to the second preset number of grids; the first preset number of grids is a number of grids in the order of 100,000, and the second preset number of grids is a number of grids in the order of 10 million; the first target number of operating points is at least ten times the second target number of operating points; The first type of three-dimensional simulation results under the first target number of operating points and the second type of three-dimensional simulation results under the second target number of operating points are data fused to obtain three-dimensional simulation samples of the thrust chamber under the first target number of operating points.

2. The method for obtaining a three-dimensional simulation sample of a thrust chamber according to claim 1, characterized in that: The obtaining of the first type of three-dimensional simulation results under the first target number of operating points includes: Acquire historical thrust chamber three-dimensional simulation data, and obtain a first type of three-dimensional simulation results under a first number of operating points from the historical thrust chamber three-dimensional simulation data; If the first quantity is less than the first target quantity, determining a first difference between the first target quantity and the first quantity; Based on the first actual number of grids, a three-dimensional simulation is performed on the thrust chamber at the first difference number of operating points to obtain a first type of three-dimensional simulation results at the first difference number of operating points; The first type of three-dimensional simulation results under the first difference number of operating points are added to the first type of three-dimensional simulation results under the first number of operating points to obtain the first type of three-dimensional simulation results under the first target number of operating points.

3. The method for obtaining a three-dimensional simulation sample of a thrust chamber according to claim 2, characterized in that: The obtaining of the second type of three-dimensional simulation results under the second target number of operating points includes: Obtaining a second type of three-dimensional simulation results at a second number of operating points from the historical thrust chamber three-dimensional simulation data; If the second amount is less than the second target amount, determining a second difference between the second target amount and the second amount; Based on the second actual number of grids, performing a three-dimensional simulation on the thrust chamber at the second difference number of operating points, and obtaining a second type of three-dimensional simulation results at the second difference number of operating points; The second type of three-dimensional simulation results at the second difference number of operating points are added to the second type of three-dimensional simulation results at the second number of operating points to obtain the second type of three-dimensional simulation results at the second target number of operating points.

4. The method for obtaining a three-dimensional simulation sample of a thrust chamber according to claim 1, characterized in that: The data fusion of the first type of three-dimensional simulation results under the first target number of operating points and the second type of three-dimensional simulation results under the second target number of operating points includes: A multi-source data fusion algorithm based on transfer learning is adopted to fuse the first type of three-dimensional simulation results under the first target number of operating points and the second type of three-dimensional simulation results under the second target number of operating points.

5. The method for obtaining a three-dimensional simulation sample of a thrust chamber according to claim 1, characterized in that: Also includes: The three-dimensional simulation samples of the thrust chamber under the first target number of operating points are reduced in order.

6. A device for obtaining a three-dimensional simulation sample of a thrust chamber, characterized in that: include: A first acquisition module is used to acquire a first type of three-dimensional simulation results under a first target number of operating points for a thrust chamber in a liquid rocket engine; A second acquisition module is used to obtain a second type of three-dimensional simulation results under a second target number of operating points; wherein, the first type of three-dimensional simulation results represents that the first actual number of grids used in the simulation calculation is less than or equal to the first preset number of grids, and the second type of three-dimensional simulation results represents that the second actual number of grids used in the simulation calculation is greater than or equal to the second preset number of grids; the first preset number of grids is a number of grids in the order of 100,000, and the second preset number of grids is a number of grids in the order of 10 million; the first target number of operating points is at least ten times the second target number of operating points; A fusion module is used to fuse the first type of three-dimensional simulation results under the first target number of operating points with the second type of three-dimensional simulation results under the second target number of operating points to obtain the three-dimensional simulation samples of the thrust chamber under the first target number of operating points.

7. The device for obtaining a three-dimensional simulation sample of a thrust chamber according to claim 6, characterized in that: The first acquisition module is specifically used to acquire historical thrust chamber three-dimensional simulation data, and obtain a first type of three-dimensional simulation results under a first number of operating points from the historical thrust chamber three-dimensional simulation data; If the first quantity is less than the first target quantity, determining a first difference between the first target quantity and the first quantity; Based on the first actual number of grids, a three-dimensional simulation is performed on the thrust chamber at the first difference number of operating points to obtain a first type of three-dimensional simulation results at the first difference number of operating points; The first type of three-dimensional simulation results under the first difference number of operating points are added to the first type of three-dimensional simulation results under the first number of operating points to obtain the first type of three-dimensional simulation results under the first target number of operating points.

8. The device for obtaining a three-dimensional simulation sample of a thrust chamber according to claim 7, characterized in that: The second acquisition module is specifically used to obtain a second type of three-dimensional simulation results under a second number of operating points from the historical thrust chamber three-dimensional simulation data; If the second amount is less than the second target amount, determining a second difference between the second target amount and the second amount; Based on the second actual number of grids, performing a three-dimensional simulation on the thrust chamber at the second difference number of operating points, and obtaining a second type of three-dimensional simulation results at the second difference number of operating points; The second type of three-dimensional simulation results at the second difference number of operating points are added to the second type of three-dimensional simulation results at the second number of operating points to obtain the second type of three-dimensional simulation results at the second target number of operating points.

9. An electronic device, characterized in that: include: A processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate via the communication bus; The memory stores a computer program that can be executed by the processor; when the processor executes the computer program, the method for obtaining a three-dimensional simulation sample of a thrust chamber as described in any one of claims 1 to 5 is executed.

10. A computer storage medium, characterized in that: The computer storage medium stores instructions, and when the instructions are executed, the method for obtaining a three-dimensional simulation sample of a thrust chamber according to any one of claims 1 to 5 is implemented.

Images