Digital twin system for liquid rocket engine

Abstract

The invention discloses a digital twin system for a liquid-propellant rocket engine, relates to the digital technical field of liquid-propellant rocket engines, and aims to solve the problem that a digital twin system in the prior art cannot meet data driving interaction between a test model of the liquid-propellant rocket engine and a physical space and real-time synchronous feedback of the state of each test part. The system at least comprises a physical space module, a digital space module, a training module, a model construction module, a service platform and a core support module, based on this, the system can establish a physical field of a plurality of target test parts according to test data of the plurality of target test parts of the liquid rocket engine, perform real-time data driving interaction between the physical field and a physical space, and perform real-time synchronous feedback on the state of each test part; data driving interaction between a test model and a physical space and real-time synchronous feedback of the state of each test part in a liquid rocket engine test are met.

Description

[0001] This application claims priority to Chinese Patent Application No. 2025102431842, filed on March 3, 2025, entitled "A Digital Twin System for a Liquid Rocket Engine", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This invention relates to the field of liquid rocket engine technology, and in particular to a digital twin system for liquid rocket engines. Background Technology

[0003] In the engineering applications of liquid rocket engines, structural field (physical field) analysis of the test object is an important means of evaluating its safety, stability, and performance. Traditional structural field simulations based on digital twin technology often employ high-precision but computationally intensive full-order models. The high computational complexity of structural field simulation models makes real-time prediction difficult, and data fusion issues still exist in real-time prediction and visualization of structural fields. This fails to meet the requirements of data-driven interaction between the test model and the physical space of liquid rocket engines, as well as the real-time synchronous feedback of the status of various test components.

[0004] Therefore, there is an urgent need to establish a digital twin system for liquid rocket engines to solve the problem that existing digital twin systems cannot meet the requirements of data-driven interaction between the test model and the physical space of liquid rocket engines, as well as the real-time synchronous feedback of the status of each test component. Summary of the Invention

[0005] The purpose of this invention is to disclose a digital twin system for liquid rocket engines, which solves the problem that existing digital twin systems cannot meet the requirements of data-driven interaction between the test model and the physical space of liquid rocket engines, as well as the real-time synchronous feedback of the status of each test component.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a digital twin system for liquid rocket engines, the system comprising at least: The module consists of a physical space module, a digital space module, a training module, a model building module, a service platform, and core support modules. The physical space module is used to acquire test data from multiple target test components and drive the digital space module to perform visualization; the multiple target test components can be any component of the liquid rocket engine; The digital space module is used to establish and visualize the physical fields of the multiple target test components based on the test data of the multiple target test components, and to feed back the state perception information to the physical space module. The training module is used to extract data features and train models based on the test data and / or simulation data of multiple target test components; The model building module is used to build multiple target models for the digital twin system based on the sample data generated by the training module; The service platform is used to provide platform services for the digital twin system; the platform services include at least status monitoring, 3D field rendering, performance monitoring, virtual-real interaction, data display, data processing, model building, remote control, and large-screen display; The core support module is used to provide core relevant information during the operation of the digital twin system, so as to support the normal operation of the digital twin system.

[0007] Preferably, the physical space module may include at least: a test bench and multiple sensor components; the multiple sensor components are used to acquire test data based on rotational performance testing and fiber optic array sensor acquisition on the test bench, the test data including at least: pressure sensor data, fiber optic grating sensor data, rotational signal acquisition data, vibration acquisition card data, temperature acquisition card data, and strain signal acquisition card data; the test bench is used to conduct physical tests on multiple target test components.

[0008] Preferably, the digital space module may include at least: A 3D physics field visualization submodule and a 1D data visualization submodule; The three-dimensional physical field visualization submodule is used to perform real-time three-dimensional physical field visualization of multiple target test components based on experimental data and / or simulation data. The one-dimensional data visualization submodule is used to perform real-time one-dimensional graphical visualization of multiple target test components based on experimental data and / or simulation data.

[0009] Preferably, the training module may include at least: a data fusion submodule, a data dimensionality reduction submodule, a data feature extraction submodule, a model training submodule, a feature model output submodule, a data feature prediction submodule, and a data restoration submodule.

[0010] Preferably, the plurality of target models may include at least a 3D model, a simulation model, a reduced-order model, a reconstructed model, and a prediction model.

[0011] Preferably, the service platform may include at least a status monitoring submodule, a 3D field rendering submodule, a performance monitoring submodule, a virtual-real interaction submodule, a data display submodule, a data processing submodule, a remote control submodule, a model building submodule, and a large screen display submodule.

[0012] Preferably, the core support module may include multiple target algorithms for supporting the digital twin system; the multiple target algorithms include one or more of the following: 3D modeling algorithm, simulation calculation algorithm, order reduction algorithm, neural network model, machine learning algorithm, 3D rendering algorithm, and data reconstruction algorithm.

[0013] Preferably, the core support module may include multiple target technologies for supporting the digital twin system; the multiple target technologies include at least preprocessing technology, postprocessing technology and model lightweighting technology.

[0014] Preferably, the digital space module can be a high-precision digital twin space module constructed using a 3D engine and combined with the actual structures of multiple target test components; the 3D engine includes at least the Unity3D engine; the high-precision digital twin space module is used to realize real-time visualization of the digital twin of the target test components based on the received multi-source data; the multi-source data includes at least real-time prediction data from the physical field reduction model and detection data from multiple sensors.

[0015] Preferably, the plurality of target test components may include test bench components with similar structures. For the test bench components with similar structures, the interface of the digital space module supports structural field modeling of three-dimensional models that vary within a preset size range, thereby constructing a digital space model of the target test component.

[0016] Compared with existing technologies, this invention provides a digital twin system for liquid rocket engines. This system comprises at least a physical space module, a digital space module, a training module, a model building module, a service platform, and a core support module. The physical space module acquires test data from multiple target test components and drives the digital space module to visualize this data. These target test components can be any parts of the liquid rocket engine. The digital space module establishes and visualizes the physical fields of the multiple target test components based on their test data and feeds back state awareness information to the physical space module. The training module extracts data features and trains a model based on the test data and / or simulation data of the multiple target test components. The model building module constructs a model based on the sample data generated by the training module. This invention relates to a digital twin system comprising multiple target models; a service platform providing platform services for the digital twin system; wherein the platform services include at least status monitoring, 3D field rendering, performance monitoring, virtual-real interaction, data display, data processing, model building, remote control, and large-screen display; and a core support module providing core relevant information during the operation of the digital twin system to support its normal operation. Based on this, a physical field for multiple target test components can be established based on test data of multiple target test components of a liquid rocket engine, enabling real-time data-driven interaction between the physical field and physical space, and real-time synchronous feedback of the status of each test component. This satisfies the requirements for data-driven interaction between the test model and physical space and real-time synchronous feedback of the status of each test component in liquid rocket engine testing. This solves the problem that existing digital twin systems cannot meet the requirements for data-driven interaction between the test model and physical space and real-time synchronous feedback of the status of each test component in liquid rocket engine testing. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 A schematic diagram of the system framework for a digital twin system of a liquid rocket engine provided by the present invention; Figure 2 A schematic diagram of the interactive processing of a digital twin system for a liquid rocket engine provided by the present invention; Figure 3 A schematic diagram of the display window of multiple sensor data terminals in a system operation example of a digital twin system for a liquid rocket engine provided by the present invention; Figure 4A schematic diagram of the display window of the liquid rocket engine parameter setting terminal in a system operation example of a digital twin system for a liquid rocket engine provided by the present invention; Figure 5 This is a schematic diagram of the display window of the dynamic physics field end in a system operation example of a digital twin system for a liquid rocket engine provided by the present invention.

[0018] Figure reference numerals: 1100 - Physical Space Module, 1200 - Digital Space Module, 1300 - Training Module, 1400 - Model Building Module, 1500 - Service Platform, 1600 - Core Support Module, 1110 - Test Bench, 1120 - Sensor Components, 1210 - 3D Physical Field Visualization Submodule, 1220 - 1D Data Visualization Submodule, 1310 - Data Fusion Submodule, 1320 - Data Dimensionality Reduction Submodule, 1330 - Data Feature Extraction Submodule, 1340 - Model Training Submodule, 1350 - Feature Model Output Submodule, 1360 - Data Feature Prediction Submodule, 1370 - Data Restoration Submodule, 1410 - 3D Model, 1420 - Simulation Model, 1430 - Reduced Dimensionality Model, 1440 - Reconstructed Model, 14 50 - Predictive Model, 1510 - Status Monitoring Submodule, 1520 - 3D Field Rendering Submodule, 1530 - Performance Monitoring Submodule, 1540 - Virtual-Real Interaction Submodule, 1550 - Data Display Submodule, 1560 - Data Processing Submodule, 1570 - Remote Control Submodule, 1580 - Model Building Submodule, 1590 - Large Screen Display Submodule, 1601 - 3D Modeling Algorithm, 1602 - Simulation Calculation Algorithm, 1603 - Order Reduction Algorithm, 1604 - Machine Learning Algorithm, 1605 - 3D Rendering Algorithm, 1606 - Data Reconstruction Algorithm, 1607 - Preprocessing Technology, 1608 - Post-processing Technology, 1609 - Model Lightweighting Technology, 1610 - Distributed Technology, 1611 - Data Acquisition Technology, 1612 - Neural Network Model. Detailed Implementation

[0019] To facilitate a clear description of the technical solutions in the embodiments of the present invention, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. For example, the first threshold and the second threshold are merely used to distinguish different thresholds and do not limit their order. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0020] It should be noted that in this invention, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0021] In this invention, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding related objects have an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, a combination of a and b, a combination of a and c, a combination of b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0022] In existing technologies, digital twin technology, as an emerging technology, has been applied in multiple fields. However, its structural field (physical field) simulation typically employs high-precision but computationally intensive full-order models, leading to issues such as insufficient data fusion and real-time performance in real-time prediction and visualization during simulation testing. Furthermore, the real-time data interaction between the existing digital twin model and the physical field also faces technical bottlenecks, making it difficult to achieve the fusion of multi-source data and real-time feedback, thus failing to meet the high real-time requirements of applications such as liquid rocket engines. Therefore, digital twin systems built using existing digital twin technologies cannot meet the requirements of data-driven interaction between the test model and physical space of liquid rocket engines, as well as the real-time synchronous feedback of the status of various test components.

[0023] In view of this, the present invention provides a digital twin system for liquid rocket engines, which can establish physical fields of multiple target test components based on test data of multiple target test components of liquid rocket engines, realize real-time data-driven interaction between the physical field and physical space, and real-time synchronous feedback of the status of each test component; it solves the problem that the existing digital twin system cannot meet the requirements of data-driven interaction between the test model and physical space of liquid rocket engines and real-time synchronous feedback of the status of each test component.

[0024] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings: Please see Figure 1 , Figure 1This invention provides a schematic diagram of the system framework for a digital twin system of a liquid rocket engine.

[0025] exist Figure 1 In the context of a digital twin system for a liquid rocket engine, the system framework may include at least: a physical space module 1100, a digital space module 1200, a training module 1300, a model building module 1400, a service platform 1500, and a core support module 1600.

[0026] The physical space module 1100 can be used to acquire test data from at least multiple target test components and drive the digital space module to perform visualization; wherein, the multiple target test components are any components of the liquid rocket engine. Specifically, the physical space module 1100 can collect various physical parameters of the test object in real time, such as stress, strain, and temperature, and transmit them to the data processing center through a wireless network access point device to ensure the real-time performance and accuracy of the data.

[0027] Specifically, the physical space module is a physical object in the digital twin system. The target test component can be a liquid rocket engine, a part of the system of the liquid rocket engine, or other physical objects. Test data of multiple target test components are acquired through sensors and / or other data acquisition methods, and the digital space module is used to visualize the data. The multiple target test components can be any part of the liquid rocket engine.

[0028] The digital space module 1200 can be used to establish and visualize the physical fields of multiple target test components based on test data of multiple target test components, and to feed back state perception information to the physical space module. That is, the digital space module can be used to construct digital models (digital twins) of multiple target test components of physical objects, corresponding to the physical space module, and can reflect the performance information of the physical module in real time.

[0029] Specifically, the Digital Space Module 1200 utilizes 3D engines such as Unity3D to construct a high-precision digital twin model based on the actual structure of the test object. By receiving real-time prediction data from the structural field reduced-order model and limited measurement data from multiple sources, it achieves real-time visualization of the digital twin of the test object; for example, it intuitively reflects the stress distribution and deformation of the test object through various methods such as color, shape, and animation.

[0030] The training module 1300 can be used at least to extract data features and train models based on test data and / or simulation data of multiple target test components.

[0031] The model building module 1400 can be used to construct at least several target models for the digital twin system based on the sample data generated by the training module. For example, the dimensionality reduction technique of POD (Proper Orthogonal Decomposition) can be selected to reduce the order of the full-order model. The input and output parameters of the reduced-order model are determined to ensure that the model can accurately reflect the structural field characteristics of the experimental object.

[0032] The service platform 1500 can at least be used to provide platform services for the digital twin system; the platform services can at least include status monitoring, 3D field rendering, performance monitoring, virtual-real interaction, data display, data processing, model building, remote control, and large-screen display. For example, the service platform can monitor and predict the operating status of the test object, and adjust the working conditions of the test object to ensure the safety and stability of the test object.

[0033] The core support module 1600 can at least provide core-related information during the operation of the digital twin system to support its normal operation. This core-related information may include technologies, algorithms, network models, and historical data files that support the operation of the digital twin system, thereby ensuring its normal operation.

[0034] Based on this, the present invention provides a digital twin system for liquid rocket engines, which includes a physical space module, a digital space module, a training module, a model building module, a service platform, and a core support module. The physical space module acquires test data from multiple target test components of the liquid rocket engine and drives the digital space module to visualize this data. The digital space module establishes and visualizes the physical fields of the multiple target test components based on their test data, while simultaneously feeding back state awareness information to the physical space module. Furthermore, the training module extracts data features and trains the model based on the test data and / or simulation data of the multiple target test components. The model construction module constructs multiple target models for the digital twin system based on sample data generated by the training module, and provides platform services for the digital twin system using the service platform; and provides core relevant information during the operation of the digital twin system using the core support module to support the normal operation of the digital twin system; thereby realizing the establishment of physical fields for multiple target test components based on test data of multiple target test components of liquid rocket engines, enabling real-time data-driven interaction between the physical field and physical space, and real-time synchronous feedback of the status of each test component; thus satisfying the requirements of data-driven interaction between the test model and physical space and real-time synchronous feedback of the status of each test component in liquid rocket engine testing.

[0035] Preferably, the physical space module 1100 includes at least: a test bench 1110 and a plurality of sensor components 1120.

[0036] The test bench 1110 is used to conduct physical tests on multiple target test components. It is understood that multiple target test components are placed on the test bench, and multiple sensor components are arranged on the test components. Test data of the target test components can be collected through the sensor components. Of course, other methods can also be used to assist in data collection at the same time, such as non-contact temperature guns or infrared detection devices, to collect test data of the target test components.

[0037] The multiple sensor components 1120 can be used to acquire test data based on rotational performance testing and fiber optic array sensor acquisition on the test bench. The test data includes at least: pressure sensor data, fiber optic grating sensor data, rotational signal acquisition data, vibration acquisition card data, temperature acquisition card data, and strain signal acquisition card data.

[0038] Specifically, a liquid rocket engine can be installed on a test bench, and various sensors can be installed on the liquid rocket engine to collect physical parameters in real time; the sensor data can be transmitted to a data processing center through a wireless network access point device; the sensors can include pressure sensors, fiber optic grating sensors, rotation signal acquisition devices, vibration acquisition cards, temperature acquisition cards, and strain signal acquisition cards, etc.

[0039] For example, sensor technology can be combined to collect various physical parameters of the test object in real time, such as stress, strain, and temperature, as input data for the structural field reduced-order model; and the sensor data can be transmitted to the data processing center through wireless network access point equipment to ensure the real-time performance and accuracy of the data.

[0040] Preferably, the digital space module 1200 may include at least: a three-dimensional physical field visualization submodule 1210 and a one-dimensional data visualization submodule 1220.

[0041] The 3D physics visualization submodule 1210 can be used to visualize the 3D physics field of multiple target test components in real time based on experimental data and / or simulation data.

[0042] The one-dimensional data visualization submodule 1220 can be used to perform real-time one-dimensional graphical visualization of multiple target test components based on experimental data and / or simulation data.

[0043] Specifically, a high-precision digital twin model can be built in Unity3D based on the actual structure of the test object. The digital twin model can then interact with the data processing center in real time, and the real-time prediction results of the structural field reduction model can be transmitted to the Unity3D digital twin model. Through various methods such as color, shape, and animation, the stress distribution and deformation of the test object can be intuitively reflected. This enables real-time visualization of the structural field data (three-dimensional and one-dimensional data) of the test object in Unity3D, providing strong support for safety assessment.

[0044] For example, the digital space module can be a high-precision digital twin space module constructed using a 3D engine and combining the actual structures of multiple target test components; the 3D engine can at least include the Unity3D engine; the high-precision digital twin space module is used to realize real-time visualization of the digital twin of the target test components based on received multi-source data. The multi-source data includes at least real-time prediction data from a reduced-order physics model and detection data from multiple sensors.

[0045] Preferably, the plurality of target test components include test bench components with similar structures. For the test bench components with similar structures, the interface of the digital space module supports structural field modeling of three-dimensional models that vary within a preset size range, thereby constructing a digital space model of the target test component.

[0046] Preferably, the training module 1300 may include at least: a data fusion submodule 1310, a data dimensionality reduction submodule 1320, a data feature extraction submodule 1330, a model training submodule 1340, a feature model output submodule 1350, a data feature prediction submodule 1360, and a data restoration submodule 1370.

[0047] The specific data fusion submodule 1310 can be used to fuse multi-source sensor data, wherein the multi-source sensor data includes at least data collected in real time by multiple sensors and reconstructed data after data reconstruction according to preset rules.

[0048] The data dimensionality reduction submodule 1320 can be used to reduce the dimensionality of fused data. For example, the data obtained by the data fusion submodule 1310 after fusing multi-source sensor data is usually high-dimensional data, which poses a huge challenge to data storage, analysis and processing. The data dimensionality reduction submodule 1320 converts the fused high-dimensional data into low-dimensional data while retaining the key information of the original data (such as stress, strain, temperature or location information), compressing the data into a lower-dimensional space; for example, using the POD intrinsic orthogonal decomposition dimensionality reduction technology.

[0049] The data feature extraction submodule 1330 can be used to extract key feature data for model training from the dimensionality-reduced data, obtain training sample data, and input the training sample data into the model training submodule for model training.

[0050] The model training submodule 1340 can be used to train a model on key feature data based on neural network models or machine learning methods, such as self-learning or semi-supervised learning, to obtain the target feature model.

[0051] The feature model output submodule 1350 can be used to output the feature model.

[0052] The data feature prediction submodule 1360 can be used to predict data based on the output feature model in order to evaluate the training effect of the model training process.

[0053] The data restoration submodule 1370 can be used to restore the training sample data.

[0054] Preferably, the plurality of target models may include at least a 3D model 1410, a simulation model 1420, a reduced-order model 1430, a reconstructed model 1440, and a prediction model 1450.

[0055] Specifically, the 3D model 1410 can be used to train the model using the target 3D data through the training module 1300, thereby obtaining the target 3D model.

[0056] The simulation model 1420 can be used to train the model using the target simulation data through the training module 1300, thereby obtaining the target simulation model.

[0057] The reduced-order model 1430 can be used to train the model using target sensor data through the training module 1300, thereby obtaining the target reduced-order model; for example, constructing a structural field reduced-order model for the test object. This model uses the reduced-order algorithm provided by the core support module, such as intrinsic orthogonal decomposition (POD), to reasonably simplify the full-order model, retain key features, and thus obtain the target reduced-order model.

[0058] The reconstruction model 1440 can be used to train the model using the target reconstruction data through the training module 1300, thereby obtaining the target reconstruction model.

[0059] The prediction model 1450 can be used to train the model using target sample data (training sample data and validation sample data) through the training module 1300, thereby obtaining the target prediction model.

[0060] Preferably, the service platform 1500 may include at least the following sub-modules: status monitoring sub-module 1510, 3D field rendering sub-module 1520, performance monitoring sub-module 1530, virtual-real interaction sub-module 1540, data display sub-module 1550, data processing sub-module 1560, remote control sub-module 1570, model building sub-module 1580, and large screen display sub-module 1590.

[0061] Specifically, the status monitoring submodule 1510 can be used to monitor the operating status of each module of the digital twin system; such as whether the system acquires and parses the data collected by the sensors of the test object in real time, and whether data-driven interaction and status synchronization feedback are realized between the digital twin model and the control system, etc.

[0062] The 3D field rendering submodule 1520 can be used in a digital twin system to render 3D physical field images by calling 3D rendering technology and rendering the 3D physical field according to the target rendering requirements.

[0063] The performance monitoring submodule 1530 can be used to monitor the performance data of various modules in a digital twin system.

[0064] The virtual-real interaction submodule 1540 can be used to perform virtual-real data interaction between the simulation platform and the physical space.

[0065] The data display submodule 1550 can be used to display target data (key data) in a digital twin system.

[0066] The data processing submodule 1560 can be used to process target data in a digital twin system.

[0067] The remote control submodule 1570 can be used to remotely control various target modules in a digital twin system; for example, it can remotely control the test status of the test bench, start or stop it, etc.

[0068] The model building submodule 1580 can be used to call corresponding technologies and algorithms, such as reconstruction algorithms or order reduction algorithms, during the model building process in a digital twin system.

[0069] The large screen display submodule 1590 can be used to send target display data from the digital twin system to the large screen for display.

[0070] Preferably, the core support module 1600 may include multiple target algorithms for supporting the digital twin system; these multiple target algorithms include a 3D modeling algorithm 1601, a simulation calculation algorithm 1602, a reduction algorithm 1603, a machine learning algorithm 1604, a 3D rendering algorithm 1605, a data reconstruction algorithm 1606, and a neural network model 1612, etc. The core support module may also include multiple target technologies for supporting the digital twin system; these multiple target technologies include at least pre-processing technology 1607, post-processing technology 1608, model lightweighting technology 1609, distributed technology 1610, and data acquisition technology 1611, etc. Based on the above core information, a core support is established for the digital twin system provided by this invention, thereby ensuring the normal operation of the digital twin system. For example, the above key technologies can be called according to actual needs to build corresponding service modules, such as physical space modules, digital space modules, etc.; thereby enabling the construction of structural field reduction models, sensor data acquisition and input, Unity3D digital twin model development, real-time prediction and visualization, and data-driven interaction and state synchronization feedback in the digital twin system.

[0071] In summary, the digital twin system for liquid rocket engines provided by this invention has the following beneficial effects: (i) Realize data-driven interaction and state synchronization feedback between the digital twin model and the physical space (control system). For example, the control system can adjust the operating parameters of the test object or take corresponding measures based on the real-time prediction results of the digital twin model to ensure the safety and stability of the test object; and at the same time, transmit the feedback data of the control system to the digital twin model in real time to realize state synchronization update.

[0072] (ii) By using a structural field reduction model, the structural field of the test object can be predicted in real time, which greatly reduces the computational complexity.

[0073] (iii) Multi-source data fusion: Combining limited measurement data, real-time prediction data and other multi-source data to improve the accuracy and reliability of digital twin models.

[0074] (iv) Real-time visualization display: Using 3D engines such as Unity3D, the structural field of the test object is visualized in real time, which intuitively reflects the safety of the test object.

[0075] (v) Data-driven interaction and feedback: Realize data-driven interaction and state synchronization feedback between the digital twin model and the control system to improve the control accuracy and response speed of the test object.

[0076] Furthermore, as a more specific example, please refer to Figure 2 , Figure 2This invention provides a schematic diagram of the interactive processing of a digital twin system for a liquid rocket engine. It should be noted that... Figure 2 yes Figure 1 The corresponding interactive processing procedures and effects between the various modules are illustrated in the diagram, along with the system framework. Figure 1 The system framework is the same. Figure 2 Compared to Figure 1 , Figure 2 It provides a more intuitive display of the test bench, the three-dimensional physical field, and the curves generated based on one-dimensional data.

[0077] exist Figure 2 The physical space module 1100 may include a test bench and multiple sensor components. The multiple sensor components are arranged at different positions of the liquid rocket engine, and the test data of the test object is acquired in real time through the multiple sensor components. The data collected by the physical space module 1100 can drive the realization of three-dimensional and one-dimensional display in the digital space module 1200. The digital space module 1200 can feed back the displayed state structure to the physical space module 1100, and determine whether the state in the digital space and the physical space is synchronized through state perception.

[0078] Furthermore, the digital space module 1200 can be used to perform three-dimensional physical field visualization and one-dimensional data visualization of the data in the physical space module. The three-dimensional and one-dimensional models used in the three-dimensional physical field and one-dimensional data field can be constructed based on the model building module 1400. Because the model building module 1400 establishes multiple target models based on dimensionality-reduced data, the digital space module 1200 only needs to extract a small amount of data from the physical space module 1100 to drive the three-dimensional and one-dimensional models for real-time visualization, thus ensuring the real-time nature of the data display. In practical applications, it is possible to use a small amount of feature data from the physical space module 1100 to drive the digital space module 1200 for real-time visualization.

[0079] Of course, it can also be a 3D model and a 1D model established using existing technologies, and then the 3D model and the 1D model are optimized based on the feature data obtained from the training module 1300; this specification does not make any specific limitations.

[0080] Furthermore, the training module 1300 acquires and parses the experimental data of the test objects detected by the physical space module 1100, including data reconstruction and preprocessing. The parsed data is then fused and its order reduced to extract key component data features, resulting in sample data. Based on this sample data, data training is performed to obtain the corresponding feature model. The model building module 1400 then constructs various models within the digital twin system. For example, an order reduction algorithm can be used to apply the parsed data as boundary conditions to construct a reduced-order model that can drive the three-dimensional physical field. This three-dimensional physical field model is then placed within the digital space module. Thus, the reduced-order model can be used to reduce the order of the experimental data, thereby driving the three-dimensional model, improving data transmission efficiency, and ensuring real-time performance.

[0081] Furthermore, the service platform 1500 can at least be used for status monitoring, 3D field rendering, performance monitoring, virtual-real interaction, data display, data processing, remote control, model building, and large-screen display of the modules it interacts with; real-time monitoring and prediction of the operating status of the test object; adjustment of the working conditions of the test object to ensure its safety and stability; and, based on the changes in the working conditions of the test object, influence the data collected by the sensors, and in turn change the boundary conditions of the reduced-order model to form a closed loop and achieve synchronous status updates.

[0082] Furthermore, during the operation of the physical space module 1100, digital space module 1200, training module 1300, model building module 1400, and service platform 1500, when computational algorithms and data processing technologies are needed, they can be invoked from the core support module 1600. The core support module 1600 can include at least: 3D modeling, simulation technology, pre / post-processing technology, model lightweighting technology, order reduction algorithms, neural network models, machine learning, reconstruction algorithms, 3D rendering, big data, data acquisition, distributed systems, microservices, data storage services, and file storage technologies; thus, these technologies can be used to support the normal operation and data interaction of various modules in the digital twin system. It should be noted that the technologies in the core support module 1600 can be updated periodically according to user needs to meet the calling requirements of various modules in the digital twin system.

[0083] For further information, please refer to [link / reference]. Figures 3 to 5 , Figure 3 A schematic diagram of the display window of multiple sensor data terminals in a system operation example of a digital twin system for a liquid rocket engine provided by the present invention; Figure 4 A schematic diagram of the display window of the liquid rocket engine parameter setting terminal in a system operation example of a digital twin system for a liquid rocket engine provided by the present invention; Figure 5This is a schematic diagram of the display window for the dynamic physics field in a system operation example of a digital twin system for a liquid rocket engine, provided by the present invention. It should be noted that in practical applications, Figures 3 to 5 The corresponding display windows can be displayed in the same display window or in separate display windows; no specific limitation is made in this invention.

[0084] exist Figure 3 In this process, test data can be collected from the liquid rocket engine in physical space using pressure sensors, temperature sensors, and velocity sensors. Examples include: main turbopump speed data, oxygen pre-compression turbopump speed data, oxygen pump outlet pressure data, first-stage fuel pump outlet pressure data, second-stage fuel pump outlet pressure data, gas generator fuel pre-injection pressure data, gas generator oxidizer pre-injection pressure data, flow regulator inlet pressure data, etc. This data is then uploaded to a digital twin system for data visualization, thereby enabling the acquisition of data such as… Figure 3 The distribution curves of different sensor data are shown.

[0085] Furthermore, based on the setting parameters of the liquid rocket engine in physical space, corresponding parameter settings can be made for the liquid rocket engine model in the digital twin system, such as... Figure 4 The parameter settings shown indicate that experimental data collected from the liquid rocket engine can be used to drive the operation of the liquid rocket engine core component twin through the digital twin system provided by this invention, enabling real-time monitoring and prediction of the test vehicle's status. For example, the thrust chamber twin reflects the velocity and temperature fields in the physical thrust chamber in real time in a three-dimensional field form; for example: Figure 5 The velocity or temperature fields corresponding to the core components shown may include the oxygen pump pressure field, oxygen pump velocity field, regulator pressure field, regulator velocity field, thrust chamber pressure field, and thrust chamber temperature field, etc.

[0086] Based on this, the present invention provides a digital twin system for liquid rocket engines. By combining a reduced-order structural field model with a 3D engine such as Unity3D, it enables real-time prediction of the structural field of the test object and visualization of the digital twin. This system allows for real-time data-driven interaction between the physical field and physical space, as well as real-time synchronous feedback of the status of each test component. This satisfies the requirements for data-driven interaction between the test model and physical space, and real-time synchronous feedback of the status of each test component in liquid rocket engine testing. It solves the problems of high computational complexity and difficulty in real-time prediction of structural field simulation models in existing technologies, as well as insufficient real-time data fusion and visualization of digital twin models. In particular, it addresses the problem that existing digital twin systems cannot meet the requirements for data-driven interaction between the test model and physical space, and real-time synchronous feedback of the status of each test component in liquid rocket engines.

[0087] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, the disclosure, and the appended claims in carrying out the claimed invention. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0088] Although the invention has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made therein without departing from the spirit and scope of the invention. Accordingly, this specification and drawings are merely exemplary descriptions of the invention as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if such modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include such modifications and modifications.

Claims

1. A digital twin system for a liquid rocket engine, characterized in that, The system includes at least: The module consists of a physical space module, a digital space module, a training module, a model building module, a service platform, and core support modules. The physical space module is used to acquire test data from multiple target test components and drive the digital space module to perform visualization; the multiple target test components can be any component of the liquid rocket engine; The digital space module is used to establish and visualize the physical fields of the multiple target test components based on the test data of the multiple target test components, and to feed back the state perception information to the physical space module. The training module is used to extract data features and train models based on the test data and / or simulation data of multiple target test components; The model building module is used to build multiple target models for the digital twin system based on the sample data generated by the training module; The service platform is used to provide platform services for the digital twin system; the platform services include at least status monitoring, 3D field rendering, performance monitoring, virtual-real interaction, data display, data processing, model building, remote control, and large-screen display; The core support module is used to provide core relevant information during the operation of the digital twin system, so as to support the normal operation of the digital twin system.

2. The system as described in claim 1, characterized in that, The physical space module includes at least: The test bench and multiple sensor components; Multiple sensor components are used to acquire test data based on rotational performance testing and fiber optic array sensor acquisition on the test bench. The test data includes at least: pressure sensor data, fiber optic grating sensor data, rotational signal acquisition data, vibration acquisition card data, temperature acquisition card data, and strain signal acquisition card data. The test bench is used to conduct physical tests on multiple target test components.

3. The system as described in claim 1, characterized in that, The digital space module includes at least: A 3D physics field visualization submodule and a 1D data visualization submodule; The three-dimensional physical field visualization submodule is used to perform real-time three-dimensional physical field visualization of multiple target test components based on experimental data and / or simulation data. The one-dimensional data visualization submodule is used to perform real-time one-dimensional graphical visualization of multiple target test components based on experimental data and / or simulation data.

4. The system as described in claim 1, characterized in that, The training module includes at least: a data fusion submodule, a data dimensionality reduction submodule, a data feature extraction submodule, a model training submodule, a feature model output submodule, a data feature prediction submodule, and a data restoration submodule.

5. The system as described in claim 1, characterized in that, The multiple target models include at least a 3D model, a simulation model, a reduced-order model, a reconstructed model, and a prediction model.

6. The system as described in claim 1, characterized in that, The service platform includes at least a status monitoring submodule, a 3D field rendering submodule, a performance monitoring submodule, a virtual-real interaction submodule, a data display submodule, a data processing submodule, a remote control submodule, a model building submodule, and a large screen display submodule.

7. The system as described in claim 1, characterized in that, The core support module includes multiple target algorithms for supporting the digital twin system; the multiple target algorithms include one or more of the following: 3D modeling algorithm, simulation calculation algorithm, order reduction algorithm, neural network model, machine learning algorithm, 3D rendering algorithm, and data reconstruction algorithm.

8. The system as described in claim 1, characterized in that, The core support module includes multiple target technologies for supporting the digital twin system; the multiple target technologies include at least pre-processing technology, post-processing technology, and model lightweighting technology.

9. The system as described in claim 1, characterized in that, The digital space module is a high-precision digital twin space module constructed using a 3D engine and combined with the actual structure of multiple target test components; The 3D engine includes at least the Unity3D engine; the high-precision digital twin space module is used to realize real-time visualization of the digital twin of the target test component based on the received multi-source data; the multi-source data includes at least real-time prediction data from the physical field reduction model and detection data from multiple sensors.

10. The system as claimed in claim 1, characterized in that, The multiple target test components include test bench components with similar structures. For the test bench components with similar structures, the interface of the digital space module supports structural field modeling of three-dimensional models that vary within a preset size range, thereby constructing a digital space model of the target test component.

Images