Virtual test analysis method, system, medium and electronic device for structural mechanical properties

Abstract

This application relates to a virtual experimental analysis method, system, medium, and electronic device for structural mechanical properties, belonging to the field of simulation testing technology. The method includes: determining an initial digital twin model corresponding to the target physical test object based on the target frequency response function; constructing a mapping relationship between each physical sensor set on the target physical test object and the corresponding simulation model nodes in the initial digital twin model, and mapping the boundary conditions in the target test parameters to the model boundary conditions of the initial digital twin model to obtain the target digital twin model; driving the target digital twin model to perform synchronous virtual loading based on actual excitation signals to obtain virtual response data; and correcting the target digital twin model based on the real response data and virtual response data of each physical sensor to finally determine the test results corresponding to the target physical test object. This application has the effect of improving the accuracy of simulation test results.

Description

Technical Field

[0001] This application relates to the field of simulation testing technology, specifically to a virtual test analysis method, system, medium, and electronic equipment for structural mechanical properties. Background Technology

[0002] Structural mechanical properties refer to the inherent mechanical performance of a structure under dynamic excitation, including natural frequencies, mode shapes, stiffness, mass, stress distribution, strain response, and damage-sensitive areas, reflecting the structure's vibration resistance and failure characteristics. Virtual experimental analysis of structural mechanical properties can be understood as establishing a finite element model of the structure, applying virtual excitation, conducting virtual vibration tests, and calculating the structure's natural frequencies, stiffness, mass, stress distribution, weak areas, and other structural mechanical properties, providing a basis for damage risk assessment, sampling frequency setting, and mesh refinement.

[0003] Currently, the common methods for conducting virtual experimental analysis of structural mechanical properties are traditional numerical analysis methods such as finite element simulation. These methods are usually based on simplified ideal models and preset parameters, which deviate from real physical products, making it difficult to guarantee the confidence level of the model. Moreover, they are mostly in "offline" mode, which cannot access physical sensing data in real time during the test and make dynamic corrections, resulting in poor accuracy of simulation test results. Summary of the Invention

[0004] To improve the accuracy of simulation test results, this application provides a method, system, medium, and electronic equipment for virtual test analysis of structural mechanical properties.

[0005] The first aspect of this application provides a virtual experimental analysis method for structural mechanical properties, specifically including: Obtain the CAE model corresponding to the target physical test object, and obtain the target test parameters of the virtual test of the target physical test object; Based on the CAE model, the natural frequencies and mode shapes of the target physical test object are determined through modal analysis. Based on the natural frequencies and mode shapes, the target frequency response function of the target physical test object is determined. Based on the target frequency response function, the initial digital twin model of the target physical test object is determined. A mapping relationship is constructed between each physical sensor set on the target physical test object and the corresponding simulation model node in the initial digital twin model, and the boundary conditions in the target test parameters are mapped to the model boundary conditions of the initial digital twin model to obtain the target digital twin model; The actual excitation signal of the physical test bench where the target physical test object is located is obtained, and the target digital twin model is driven to perform synchronous virtual loading based on the actual excitation signal to obtain virtual response data; Based on the real response data and virtual response data of each physical sensor, the target digital twin model is modified to obtain a modified twin model, and based on the modified twin model, the test results corresponding to the target physical test object are obtained.

[0006] By adopting the above technical solution, based on the CAE model of the target physical test object, natural frequencies and mode shapes are extracted through modal analysis, and an initial digital twin model is constructed to provide a benchmark for subsequent dynamic correction. On this basis, a mapping relationship is established between each physical sensor deployed on the physical test bench and the corresponding simulation nodes in the digital twin model. Simultaneously, the target boundary conditions of the virtual test are converted into model boundary conditions, resulting in the target digital twin model, ensuring the consistency between the target digital twin model and the physical entity in terms of geometry and mechanical topology. When an excitation signal is applied to the actual physical test bench, the system acquires the actual excitation signal in real time and synchronously drives the target digital twin model to perform virtual loading, ensuring that the simulation model and the physical test are under the same excitation input, achieving synchronous "physical-virtual" operation. Furthermore, by comparing the actual response data of the physical sensors with the virtual response data of the digital twin model, the target digital twin model is dynamically corrected, enabling it to continuously absorb measured information during the physical test process, automatically calibrate model parameters and boundary conditions, and thus continuously approximate the actual mechanical behavior of the physical entity. Ultimately, based on the corrected twin model, the experimental results were output. By constructing a closed-loop architecture of "modal analysis - frequency response function - digital twin modeling - physical sensor mapping - real-time excitation drive - dynamic correction", the deep coupling and synchronous evolution of the simulation model and the physical test object were realized, which significantly improved the accuracy of the simulation test results and enabled the digital twin model to accurately reflect the structural mechanical response of the target physical test object under real working conditions.

[0007] In one implementation, determining the natural frequencies and mode shapes of the target physical test object through modal analysis based on the CAE model specifically includes: The CAE model is discretized using the finite element method to obtain the structural dynamics equations corresponding to the target physical test object; Based on the structural dynamics equations, determine the mass matrix and stiffness matrix corresponding to the CAE model; Substituting the mass matrix and the stiffness matrix into the preset characteristic equation of undamped free vibration, the natural frequencies and mode shapes corresponding to the target physical test object are obtained.

[0008] In one implementation, determining the initial digital twin model corresponding to the target physical test object based on the target frequency response function specifically includes: Obtain key parameters characterizing the materials and structure of the target physical test object; Based on the key parameters and the target frequency response function, an initial digital twin model corresponding to the target physical test object is determined. The mathematical expression of the initial digital twin model is as follows: ; In the formula, Represents the frequency response function matrix. Indicates the excitation frequency. Represents the i-th natural frequency. Denotes the vector of the i-th mode shape. Let represent the damping ratio of the i-th mode, θ represent the vector corresponding to the key parameter, and j represent the imaginary unit.

[0009] In one implementation, after obtaining virtual response data by synchronously virtual loading of the target digital twin model based on the actual excitation signal, the method further includes: Obtain the stress-strain gradient of each region in the target digital twin model; When the stress-strain gradient exceeds a preset gradient threshold, the corresponding region is identified as the key region of the target digital twin model. The key area is subdivided into local grids using a grid encryption algorithm to obtain a subdivided twin model. The process of correcting the target digital twin model based on the real and virtual response data from each physical sensor to obtain the corrected twin model specifically includes: Based on the real and virtual response data of each physical sensor, the subdivided twin model is corrected to obtain the corrected twin model.

[0010] In one embodiment, the step of obtaining the target test parameters of the virtual test of the target physical test object specifically includes: When the target test parameters are sampling setting information, multiple historical locations where damage has occurred on historical objects are obtained, and at least one location requiring attention is determined from each of the historical locations. The historical objects are of the same type as the target physical test object. When the area of ​​interest has been damaged, the historical usage scenarios of the historical object are obtained, and at least one usage scenario of interest is determined from each of the historical usage scenarios; Assess a first risk value for damage to the areas of concern, and assess a second risk value for damage to each of the usage scenarios of concern. Obtain the actual usage scenario of the target physical test object, and based on the actual usage scenario, the first risk value, and each of the second risk values, determine the sampling setting information of the virtual test of the target physical test object.

[0011] In one implementation, determining the sampling settings information for the virtual test of the target physical test object based on the actual usage scenario, the first risk value, and each of the second risk values ​​specifically includes: If the actual usage scenario mentioned above exists in any of the aforementioned usage scenarios that require attention, then the part that requires attention will be determined as the reference part. Based on the first risk value of the reference part and the second risk value of the actual use scenario, a risk index for damage to the reference part during actual application is obtained. The risk index of each reference part is compared with a preset index threshold. When the risk index exceeds the index threshold, the corresponding reference part is determined as a sampling part. Based on the risk index of the sampling part, the sampling frequency of the sampling part is determined, and the sampling part and the sampling frequency are determined as sampling setting information.

[0012] In one embodiment, obtaining the target test parameters of the virtual test of the target physical test object specifically includes: When the target test parameter is the excitation location information, multiple historical excitation points that cover the part of interest in the historical virtual test are obtained, and at least one excitation point of interest is determined from the multiple historical excitation points. Evaluate the weight of each of the aforementioned stimulus points, whereby the weight represents the likelihood that the stimulus point will cover the area of ​​interest. Multiply the first risk value of a single sampling location by the weights of the corresponding incentive points that need attention to obtain multiple multiplication results corresponding to a single sampling location; Summing the multiplication results corresponding to all the sampling locations, the multiplication results involving the same stimulus point of interest are obtained to get the coverage index of the corresponding stimulus point of interest. The coverage index represents the overall probability of covering the sampling location. If the coverage index exceeds a preset threshold, the corresponding incentive point to be focused on will be determined as incentive location information.

[0013] A second aspect of this application provides a virtual test analysis system for structural mechanical properties, specifically including: The data acquisition module is used to acquire the CAE model corresponding to the target physical test object and to acquire the target test parameters of the virtual test of the target physical test object; The model building module is used to determine the natural frequencies and mode shapes of the target physical test object based on the CAE model through modal analysis, determine the target frequency response function of the target physical test object based on the natural frequencies and mode shapes, and determine the initial digital twin model of the target physical test object based on the target frequency response function. The model optimization module is used to construct the mapping relationship between each physical sensor set on the target physical test object and the corresponding simulation model node in the initial digital twin model, and to map the boundary conditions in the target test parameters to the model boundary conditions of the initial digital twin model to obtain the target digital twin model; The virtual loading module is used to acquire the actual excitation signal of the physical test bench where the target physical test object is located, and drive the target digital twin model to perform synchronous virtual loading based on the actual excitation signal to obtain virtual response data; The test analysis module is used to modify the target digital twin model based on the real response data and the virtual response data of each physical sensor to obtain the modified twin model, and to obtain the test results corresponding to the target physical test object based on the modified twin model.

[0014] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when loaded and executed by a processor, performs the steps of the method described in any one of the first aspects.

[0015] A fourth aspect of this application provides an electronic device, specifically comprising: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, the processor being configured to load and execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of the first aspects.

[0016] In summary, this application includes at least one of the following beneficial technical effects: Based on the CAE model of the target physical test object, natural frequencies and mode shapes are extracted through modal analysis, thereby constructing an initial digital twin model to provide a benchmark for subsequent dynamic correction. On this basis, a mapping relationship is established between each physical sensor deployed on the physical test bench and the corresponding simulation nodes in the digital twin model. Simultaneously, the target boundary conditions of the virtual test are converted into model boundary conditions, resulting in the target digital twin model, ensuring the consistency between the target digital twin model and the physical entity in terms of geometry and mechanical topology. When an excitation signal is applied to the actual physical test bench, the system acquires the actual excitation signal in real time and synchronously drives the target digital twin model to perform virtual loading, ensuring that the simulation model and the physical test are under the same excitation input, achieving synchronous "physical-virtual" operation. Furthermore, by comparing the real response data of the physical sensors with the virtual response data of the digital twin model, the target digital twin model is dynamically corrected, enabling it to continuously absorb measured information during the physical test process, automatically calibrate model parameters and boundary conditions, and thus continuously approximate the real mechanical behavior of the physical entity. Ultimately, based on the corrected twin model, the experimental results were output. By constructing a closed-loop architecture of "modal analysis - frequency response function - digital twin modeling - physical sensor mapping - real-time excitation drive - dynamic correction", the deep coupling and synchronous evolution of the simulation model and the physical test object were realized, which significantly improved the accuracy of the simulation test results and enabled the digital twin model to accurately reflect the structural mechanical response of the target physical test object under real working conditions. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a virtual test analysis method for structural mechanical properties provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the relationship between the parts of interest and the usage scenarios of interest, provided in an embodiment of this application. Figure 3 This is a schematic diagram of a virtual test analysis system for structural mechanical properties provided in an embodiment of this application; Figure 4 This is a schematic diagram of another virtual test analysis system for structural mechanical properties provided in this application embodiment.

[0018] Figure labeling: 11. Data acquisition module; 12. Model building module; 13. Model optimization module; 14. Virtual loading module; 15. Experimental analysis module; 16. Mesh subdivision module. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0020] In the description of the embodiments of this application, words such as "exemplarily," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplarily," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of words such as "exemplarily," "for example," or "for instance" is intended to present the relevant concepts in a specific manner.

[0021] In the description of the embodiments of this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, or A and B existing simultaneously. Furthermore, unless otherwise stated, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0022] See Figure 1 This application discloses a flowchart of a virtual test analysis method for structural mechanical properties, which can also be implemented using a computer program or run on a virtual test analysis system for structural mechanical properties based on the von Neumann architecture. This computer program can be integrated into an application or run as a standalone tool application, specifically including: S101: Obtain the CAE model corresponding to the target physical test object, and obtain the target test parameters of the virtual test of the target physical test object.

[0023] Specifically, in this embodiment of the application, the execution subject of a virtual test analysis method for structural mechanical properties is a server. The server is wirelessly connected to the user's terminal, which can be a personal computer or a tablet computer. The terminal has a client related to virtual test analysis installed on it. The server is the backend server of the client, which can be an independent physical server or a cluster of multiple physical servers.

[0024] The target physical test object is the test object to be subjected to vibration testing, which can be a general structural component or a mechanical part. A Computer-Aided Engineering (CAE) model refers to a simulation analysis model formed by geometric modeling, mesh generation, and material property assignment of the target physical test object based on computer-aided engineering technology. It is used to simulate the physical characteristics and mechanical behavior of the target physical test object. This model can be built using mainstream CAE simulation software such as ANSYS, ABAQUS, and HyperWorks. Virtual testing refers to a testing method that simulates the actual test process of the target physical test object in a computer virtual environment based on a CAE model, obtaining relevant test data without conducting physical tests. Target test parameters refer to the key parameters that need to be preset during virtual testing, including but not limited to boundary conditions (such as constraint methods, support positions, and loading directions), test loading parameters or excitation-related parameters (such as excitation force, excitation frequency, and excitation location information), sampling settings, test accuracy requirements, and test termination conditions. These parameters are determined according to the test purpose of the target physical test object (such as strength testing, vibration testing, or fatigue testing) and relevant industry standards. In practice, a full-size scan of the target physical test object is performed using 3D scanning equipment (such as laser scanning equipment or CT scanning equipment) to acquire its 3D geometric data. This data is then imported into CAE simulation software, where redundant geometric features (such as irrelevant chamfers and small protrusions) are removed, simplifying and reconstructing the geometric model. The reconstructed geometric model is then meshed (the mesh type is selected as tetrahedral or hexahedral according to experimental requirements, and the mesh size is adjusted according to the experimental accuracy requirements; the higher the accuracy requirement, the smaller the mesh size). Based on the actual material properties of the target physical test object (such as elastic modulus, Poisson's ratio, density, yield strength, etc.), corresponding material parameters are assigned to the CAE model, completing the construction of the CAE model. This model is then stored in a database, generating a corresponding model call address. When a virtual test is needed on the target physical test object, the CAE model corresponding to the target physical test object is retrieved through the model call address, preparing for the subsequent virtual test. It should be noted that excitation-related parameters refer to various control parameters set in physical vibration tests or virtual simulations to characterize the excitation characteristics when applying external dynamic loads to the test object. Excitation intensity: also known as excitation amplitude, refers to the magnitude of the dynamic load applied to the excitation point of the test object; excitation frequency: refers to the vibration frequency of the excitation signal, used to characterize the speed of the excitation; excitation location information: also known as excitation point information, refers to the specific spatial location where the excitation is applied to the test object, used to define the area of ​​effect of the excitation. Additionally, sampling setting information refers to the various acquisition configuration parameters set in physical experiments for collecting the dynamic response signals (such as acceleration, strain, displacement, etc.) of the test object, including but not limited to sampling location and sampling frequency.

[0025] Simultaneously, it is necessary to obtain the target test parameters for the virtual test of the target physical test object. One feasible method for obtaining these parameters is as follows: With the target test parameters set as sampling information, historical usage records cached in the database are used to obtain the historical locations where damage has occurred on multiple historical objects. These historical locations may be identical. The historical usage records include, but are not limited to, the locations where damage occurred on historical objects within a preset historical timeframe, along with the usage scenarios in which the damage occurred. The preset historical timeframe can be 2 years or other reasonable periods. The historical objects are of the same type as the target physical test object. For example, the target physical test object is a general-purpose metal mounting bracket (used for equipment fixation, support, and vibration damping), and the historical objects are general-purpose metal mounting brackets of the same model, size, and material as the target physical test object. The damaged locations can be the edges of the bracket bolt holes, and the usage scenarios can be applications such as air conditioner outdoor units (outdoor, high-altitude, long-term vibration), car chassis (bumps, mud, impact), etc.

[0026] Furthermore, the frequency of a single historical part recurring within multiple historical parts is statistically analyzed. If the frequency exceeds a preset frequency threshold, it indicates that the historical part has been damaged more frequently. Therefore, this historical part is identified as a part requiring attention; that is, a part of the target physical test object that is prone to damage during actual use, and at least one part requiring attention exists. Then, based on the aforementioned historical usage records, multiple historical usage scenarios where damage has occurred to a single part requiring attention are obtained. The frequency of a single historical usage scenario recurring within these scenarios is statistically analyzed. If the frequency exceeds a preset frequency threshold, this historical usage scenario is identified as a usage scenario requiring attention for the single part requiring attention; that is, a usage scenario where damage to the part requiring attention is likely to occur during actual use. Each part requiring attention corresponds to at least one usage scenario requiring attention.

[0027] Furthermore, a first risk value for damage to a single area of ​​concern is assessed. This first risk value is the ratio of the first frequency of recurrence of the single area of ​​concern to the sum of the first frequencies of recurrence of all areas of concern, representing the likelihood of damage to the single area of ​​concern in the target physical test object during use. Next, a second risk value for damage to the single area of ​​concern is assessed in each corresponding usage scenario of concern. This second risk value is the ratio of the second frequency of recurrence of the single usage scenario of concern to the sum of the second frequencies of recurrence of all usage scenarios of concern, representing the likelihood of damage to the single area of ​​concern in the target physical test object within the usage scenario of concern. For example, there are two areas of concern, B1 and B2. The first frequency of occurrence of area B1 is 45 times, and the first frequency of occurrence of area B2 is 55 times. The first risk value of area B1 is: 45 times / (45 times + 55 times) = 0.45. Area B1 corresponds to usage scenarios B11 and B12, and area B2 corresponds to usage scenarios B21 and B22, etc. The second frequency of occurrence of usage scenario B11 is 50 times, and the second frequency of occurrence of usage scenario B12 is 50 times. Therefore, the second risk value of usage scenario B11 is: 50 times / (50 times + 50 times) = 0.5. See [link to relevant documentation] for details. Figure 2 .

[0028] Furthermore, the actual usage scenario of the target physical test object is obtained, specifically through user uploads via terminal. Based on the actual usage scenario, the first risk value of a single area of ​​concern, and the second risk values ​​of each usage scenario corresponding to a single area of ​​concern, the sampling settings information for the virtual test of the target physical test object is determined. One feasible implementation method is as follows: If there are actual usage scenarios among the various usage scenarios corresponding to a single area of ​​concern, then the area of ​​concern is determined as a reference area. That is, the area where the target physical test object is highly likely to be damaged, and there is at least one reference area. Then, the first risk value of the reference area is multiplied by the second risk value of the actual usage scenario corresponding to the reference area to obtain the risk index of damage to the reference area in actual use. The risk index characterizes the probability of damage to the reference area of ​​the target physical test object in actual use.

[0029] Each risk index is compared with a preset index threshold. If the risk index exceeds the threshold, it indicates a higher probability of damage to the target physical test object at the corresponding reference location during actual use. Therefore, the corresponding reference location is designated as the sampling location, i.e., the specific location on the target physical test object where physical sensors (such as accelerometers and strain gauges) are placed; at least one sampling location exists. Then, using a preset risk-frequency mapping table, the sampling frequency for each sampling location is matched to its risk index. The higher the risk index, the higher the corresponding sampling frequency. The risk-frequency mapping table includes different risk index ranges and their corresponding sampling frequencies: risk index range 0-0.2 corresponds to a sampling frequency of 0.5-1kHz; risk index range 0.2-0.4 corresponds to a sampling frequency of 1-2kHz; risk index range 0.4-0.6 corresponds to a sampling frequency of 2-5kHz, and so on. If the risk index is between 0-0.2, the corresponding sampling frequency range is 0.5-1kHz. Finally, each sampling location and its corresponding sampling frequency are determined as the sampling setting information.

[0030] In other embodiments, if the risk index does not exceed the index threshold, it indicates that the probability of damage to the target physical test object at the corresponding reference location is relatively low during actual use. Therefore, this reference location is determined as the location to be analyzed. A target location set is selected from multiple locations to be analyzed, where each location to be analyzed is located in the same continuous region, and the area of ​​the continuous region is less than a preset area threshold. Further, the risk indices corresponding to each location to be analyzed in the target location set are summed to obtain the overall risk index, which characterizes the overall probability of damage to the continuous region in the target physical test object. If the overall risk index exceeds the index threshold, the overall probability is relatively high. Therefore, the continuous region is determined as the sampling location, and the sampling frequency of the continuous region is determined based on the overall risk index and the aforementioned risk-frequency mapping table. In another embodiment, the target test parameters can also be obtained by the user through a terminal upload.

[0031] In one embodiment, when the target test parameter is excitation location information, multiple historical excitation points that can cover a single area of ​​interest in the historical object in historical virtual tests are obtained based on historical test records cached in the database. These historical test records include, but are not limited to, information such as multiple historical excitation points in the virtual tests of the historical object and the area covered by each historical excitation point. It should be noted that multiple historical excitation points that can cover a single area of ​​interest can be understood as applying excitation at multiple historical excitation points so that the single area of ​​interest can receive effective excitation, thereby accurately simulating its stress state for virtual testing.

[0032] Furthermore, the third frequency of a single historical stimulus point is counted among multiple historical stimulus points. If the third frequency exceeds a corresponding preset frequency threshold, then that historical stimulus point is identified as a stimulus point of interest corresponding to a single area of ​​interest, i.e., a stimulus point that is likely to cover a single area of ​​interest in the target physical test object. Each area of ​​interest corresponds to at least one stimulus point of interest. Next, the weights of each stimulus point of interest corresponding to a single area of ​​interest are evaluated. The weight is the ratio of the third frequency of a single stimulus point of interest to the sum of the third frequencies of all stimulus points of interest, representing the probability of covering the single area of ​​interest when stimulus is applied at that stimulus point.

[0033] The first risk value of a single sampling location is multiplied by the weights of its corresponding stimulus points, resulting in multiple multiplication results for that single sampling location. Each multiplication result represents the probability of coverage of the single sampling location when stimulus is applied at the corresponding stimulus point. Then, the multiplication results involving the same stimulus point are summed across all sampling locations to obtain the coverage index of that stimulus point. The coverage index represents the overall probability of coverage of the sampling location within the target physical test object when stimulus is applied at that stimulus point. Furthermore, if the coverage index exceeds a preset threshold, it indicates a high probability of coverage of the sampling location when stimulus is applied at the corresponding stimulus point during virtual testing of the target physical test object. This makes the virtual test more closely resemble actual service scenarios, more targeted, and improves both efficiency and accuracy. Therefore, the corresponding stimulus point is identified as the stimulus location information. Additionally, the stimulus mode for each stimulus point can be sinusoidal, impact, or pulse stimulus, etc.

[0034] S102: Based on the CAE model, determine the natural frequencies and mode shapes of the target physical test object through modal analysis. Based on the natural frequencies and mode shapes, determine the target frequency response function of the target physical test object. Based on the target frequency response function, determine the initial digital twin model of the target physical test object.

[0035] Specifically, the CAE model is imported into a professional CAE simulation tool. Using the software's built-in mesh generation function (manual / automatic mesh refinement, local mesh thinning), the CAE model is discretized using the finite element method (dividing the continuous structure into a finite number of elements). Then, based on structural mechanics principles, the corresponding structural dynamics equations are automatically derived and generated. This finite element discretization process is existing technology and will not be elaborated upon here. The professional CAE simulation tool used is either ANSYS or ABAQUS.

[0036] The structural dynamics equations are obtained as follows: In the formula, M represents the mass matrix, characterizing the mass distribution of nodes in the CAE model; C represents the damping matrix, and K represents the stiffness matrix, characterizing the structural resistance to deformation of the target physical test object. u(t) represents the node displacement vector, characterizing the displacement of all nodes in the CAE model at time t. Represents the nodal velocity vector. represents the nodal acceleration vector. f(t) represents the external load vector, characterizing the external forces, excitations, impacts, and vibration loads applied to the nodes. Based on the established structural dynamics equations, the mass matrix and stiffness matrix corresponding to the CAE model are further extracted and determined. The mass matrix is ​​obtained by assembling the element mass matrices, and the stiffness matrix is ​​obtained by assembling the element stiffness matrices. The assembly process strictly follows the finite element assembly rules to ensure that the matrices can accurately reflect the mass distribution and stiffness characteristics of the structure. Specifically, the stiffness matrix and mass matrix are determined using ANSYS or ABAQUS tools. Finally, the obtained mass matrix and stiffness matrix are substituted into the preset undamped free vibration characteristic equation, which is: In the formula, This represents the i-th natural frequency, i.e., the natural frequency. Let represent the vector of the i-th mode shape, i.e., the mode shape.

[0037] The characteristic equation is solved using the eigenvalue solver module of ANSYS or ABAQUS tools, yielding several eigenvalues ​​and corresponding eigenvectors. The square root of the eigenvalues ​​is then used to obtain the natural frequencies of the CAE model. The eigenvectors are then converted into the corresponding mode shapes to complete the modal analysis, providing a foundation for determining the target frequency response function.

[0038] It should be noted that "modal analysis" refers to the analytical method of obtaining the inherent vibration characteristics (natural frequencies, mode shapes) of a structure. It is the core step in structural dynamics analysis and is used to reveal the vibration laws of the target physical test object. "Finite element discretization" refers to dividing a continuous CAE model (corresponding to the continuous structure of the target physical test object) into several interconnected discrete elements (such as tetrahedral elements and hexahedral elements). By superimposing the mechanical properties of the discrete elements, the mechanical behavior of the continuous structure is approximately simulated. It is a prerequisite for solving the structural dynamics equations. "Structural dynamics equations" refer to the differential equations that describe the balance relationship between inertial forces, damping forces, elastic forces and external excitations during the vibration of the target physical test object. They are used to characterize the dynamic response characteristics of the structure.

[0039] Furthermore, based on the natural frequencies and the mode shapes, the target frequency response function corresponding to the target physical test object is determined. The target frequency response function is a frequency response function that reflects the relationship between the response and the excitation in the virtual test. One feasible way to determine it is to obtain the damping ratios (ζ1, ζ2, ..., ζ) of each mode of the target physical test object. m , where ζᵢ is the damping ratio of the i-th mode, typically ranging from 0.01 to 0.1); based on the natural frequency, mode shape vector, and modal damping ratio obtained above, the modal superposition method is used, combined with a preset frequency response function calculation formula, to construct the target frequency response function corresponding to the target physical test object. The target frequency response function is as follows:

[0040] In the formula, H(ω) is the target frequency response function matrix, ω is the excitation frequency (which can be obtained by uploading through the user's terminal), j is the imaginary unit, φᵢᵀ is the transpose of the i-th mode shape vector φᵢ, the denominator is the characteristic equation term of the i-th mode, which is used to characterize the response characteristics of the i-th mode at different excitation frequencies, and n represents the maximum order.

[0041] After determining the target frequency response function, key parameters characterizing the material and structure of the target physical test object are obtained. One method is to acquire these parameters through material testing (e.g., elastic modulus and Poisson's ratio obtained through tensile testing, and material density obtained through density testing), and structural parameters (e.g., geometric dimensions, wall thickness, connection positions, etc.) obtained through 3D scanning and dimensional measurement. All material and structural parameters are then integrated to form key parameters characterizing the material and structure of the target physical test object. These key parameters are further arranged in a preset order to construct a vector θ corresponding to each key parameter. Then, the target frequency response function is used as the mathematical expression of the surrogate model, and vector θ is used as the independent variable in the mathematical expression to obtain an initial digital twin model. This initial digital twin model reflects the dynamic characteristics, material and structural characteristics of the target physical test object. This model can simulate the response law of the target physical test object under different excitation frequencies, providing a core framework for the subsequent construction of the target digital twin model. The mathematical expression of the initial digital twin model is: ,in, Let m represent the frequency response function matrix, and m represent the maximum order.

[0042] S103: Construct the mapping relationship between each physical sensor set on the target physical test object and the corresponding simulation model node in the initial digital twin model, and map the boundary conditions in the target test parameters to the model boundary conditions of the initial digital twin model to obtain the target digital twin model.

[0043] Specifically, "physical sensors" refer to detection devices installed on or inside the target physical test object to collect real-time response data (such as displacement, velocity, acceleration, stress, strain, etc.) of the target physical test object during the test. These include, but are not limited to, accelerometers, strain gauges, and displacement sensors, and their installation locations are the sampling sites defined above. "Simulation model nodes" refer to the basic unit nodes formed after the initial digital twin model has undergone finite element discretization. Each node corresponds to a specific spatial location on the target physical test object, and virtual response data corresponding to that location can be output through simulation calculations. In the initial digital twin model, simulation model nodes corresponding to the three-dimensional coordinates of each physical sensor are located using ANSYS tools to ensure a precise correspondence between the node positions and the physical sensor installation positions. Subsequently, a one-to-one mapping relationship is established between the two, clarifying which simulation model node in the initial digital twin model corresponds to the virtual response data collected by each physical sensor. Subsequently, the boundary conditions in the target test parameters are retrieved, and the boundary conditions of the real test (such as the support method, fixed position, constraint force, etc. of the target physical test object) are parameterized and transformed. For example, fixed constraints are converted into fixed degree-of-freedom constraint parameters that can be recognized by simulation software, and hinged constraints are converted into degree-of-freedom restriction parameters in the corresponding direction. The transformed model boundary conditions are assigned to the corresponding positions of the initial digital twin model to complete the mapping of boundary conditions and obtain the target digital twin model.

[0044] S104: Obtain the actual excitation signal of the physical test bench where the target physical test object is located, and drive the target digital twin model to perform synchronous virtual loading based on the actual excitation signal to obtain virtual response data.

[0045] Specifically, a "physical test bench" refers to a specialized test device used to place the target physical test object, provide a stable test environment, and apply test excitation. It can output controllable external excitation according to the requirements of the target test parameters, while providing preset support and constraints for the target physical test object to ensure the smooth conduct of the actual test. The "actual excitation signal" refers to the signal corresponding to the external excitation (such as vibration excitation, impact excitation, load excitation, etc.) applied to the target physical test object by the physical test bench during the actual test. This signal is the core input that drives the target physical test object to generate a response. Its parameters (such as excitation frequency, excitation amplitude, excitation duration, and excitation waveform) are consistent with the loading parameters in the target test parameters and can be acquired by excitation sensors.

[0046] The physical test bench applies an actual excitation signal to the target physical test object according to the preset loading requirements in the target test parameters. The excitation sensor collects this actual excitation signal in real time. The collected signal is initially a raw signal containing external interference noise, which needs to be preprocessed. This preprocessing includes three steps: filtering, noise reduction, and amplitude calibration. The filtering process uses a low-pass filtering algorithm to remove high-frequency noise generated by external environmental vibrations and equipment operating noise. The noise reduction process uses a mean filtering algorithm to eliminate random interference in the signal. Amplitude calibration corrects the amplitude of the collected signal based on a standard excitation signal to ensure that the processed actual excitation signal can accurately reflect the excitation state applied by the physical test bench. After preprocessing, the actual excitation signal (time domain signal or frequency domain signal) is converted into a loading signal that can be recognized by the target digital twin model. During the conversion of commands and loading parameters, it is ensured that core parameters such as excitation frequency, excitation amplitude, excitation duration, and excitation waveform are completely consistent with the actual excitation signal to avoid parameter distortion. Subsequently, based on the converted loading commands and loading parameters, the target digital twin model is driven to perform synchronous virtual loading. The start time of virtual loading and the timing of excitation parameter changes are completely synchronized with the actual loading on the physical test bench to ensure the consistency between the virtual loading process and the real experimental loading process. During the loading process, the operating status of the target digital twin model is monitored in real time to avoid problems such as model errors and abnormal responses. During the synchronous virtual loading process, the response data of all simulation model nodes in the target digital twin model are collected in real time using ANSYS tools, with a focus on collecting the response data of simulation model nodes corresponding to physical sensors to obtain virtual response data.

[0047] In other embodiments, the stress-strain analysis module of ANSYS tools is used to calculate the stress-strain data of each region of the target digital twin model during synchronous virtual loading. Based on this stress-strain data, the stress-strain gradient of each region is calculated. Specifically, the calculation method is as follows: select any two adjacent discrete elements in the target digital twin model, calculate the stress-strain difference between the two elements, divide it by the center distance between the two elements to obtain the stress-strain gradient of the adjacent region, and calculate the stress-strain gradient of all adjacent regions in turn to complete the calculation of the stress-strain gradient of the entire model. Then, the calculated stress-strain gradient of each region is compared with a preset gradient threshold one by one. If the stress-strain gradient of a certain region exceeds the preset gradient threshold, the region is determined to be the target. The key areas of the digital twin model are recorded, including their location and extent. Next, a mesh refinement algorithm is activated to subdivide the identified key areas into smaller, more precise meshes (mesh size reduced to 1 / 2-1 / 4 of the original size, adjusted according to accuracy requirements). This ensures the subdivided mesh accurately captures stress and strain changes in the key areas while maintaining the mesh size in non-key areas to avoid excessive computational load and reduced efficiency due to overly dense overall meshes. After subdivision, the model undergoes mesh quality checks (e.g., mesh distortion and aspect ratio checks) to ensure the subdivided mesh quality meets simulation requirements, resulting in the subdivided twin model. It should be noted that "stress-strain gradient" refers to the ratio of the stress-strain difference between adjacent regions (or adjacent discrete elements) in the target digital twin model to the distance between regions. It is used to characterize the rate of change of stress-strain distribution in the target digital twin model. The larger the gradient, the more uneven the stress-strain distribution in that region, and the more prone it is to deformation or damage. "Preset gradient threshold" refers to a pre-set critical value used to determine whether the stress-strain distribution in the target digital twin model is uniform and whether mesh refinement is required. "Synchronous virtual loading" refers to converting the actual excitation signal of the physical test bench into a loading signal that the model can recognize in real time, driving the subdivided twin model to synchronously simulate the loading process of the real test, ensuring the consistency between virtual loading and real loading.

[0048] S105: Based on the real and virtual response data of each physical sensor, the target digital twin model is corrected to obtain the corrected twin model, and based on the corrected twin model, the test results corresponding to the target physical test object are obtained.

[0049] Specifically, the virtual response dataset stored during the virtual loading process and the real response dataset collected by physical sensors are retrieved. The two sets of data are matched one-to-one according to the acquisition time node and monitoring location. For each physical sensor and its corresponding simulation model node, the deviation value between the virtual response data and the real response data is calculated using methods such as absolute deviation, relative deviation, or root mean square deviation. All deviation values ​​are counted, and it is determined whether the deviation value exceeds the preset allowable range. If the deviation value exceeds the allowable range, the model correction process is initiated. Specifically, the correction is performed through parameter inversion algorithms (such as gradient descent optimization or Bayesian update), which is existing technology and will not be elaborated here. The correction priority is as follows: material property parameters, boundary condition parameters, and mesh size. Priority is given to fine-tuning the material property parameters of the target digital twin model (such as elastic modulus, Poisson's ratio, modal damping ratio, etc.). After each adjustment, the model is re-driven for virtual loading to obtain new virtual response data, which is then compared with the actual response data. If the deviation is still not within the acceptable range, the boundary condition parameters (such as constraint strength and support position) are fine-tuned. If the deviation still exceeds the range, the mesh in the area with the larger deviation is locally refined, and the above adjustment-loading-comparison process is repeated until all deviation values ​​are within the preset allowable range, completing the correction of the target digital twin model and obtaining the corrected twin model. If the deviation values ​​do not exceed the allowable range, no model correction is needed, and the target digital twin model is directly used as the corrected twin model. After correction, based on the corrected twin model, the complete test conditions of the target physical test object are simulated, and various response data output by the model are continuously collected. The data is analyzed, organized, and filtered to extract key data that can characterize the physical performance and test effect of the target physical test object, forming complete test results. In other embodiments, the subdivided twin model can be modified based on the real and virtual response data of each physical sensor to obtain a modified twin model.

[0050] The implementation principle of the virtual test analysis method for structural mechanical properties in this application is as follows: Based on the CAE model of the target physical test object, natural frequencies and mode shapes are extracted through modal analysis, and an initial digital twin model is constructed to provide a benchmark for subsequent dynamic correction. On this basis, a mapping relationship is established between each physical sensor deployed on the physical test bench and the corresponding simulation nodes in the digital twin model. Simultaneously, the target boundary conditions of the virtual test are converted into model boundary conditions to obtain the target digital twin model, ensuring the consistency between the target digital twin model and the physical entity in terms of geometry and mechanical topology. When an excitation signal is applied to the actual physical test bench, the system acquires the actual excitation signal in real time and synchronously drives the target digital twin model to perform virtual loading, so that the simulation model and the physical test are under the same excitation input, achieving synchronous "physical-virtual" operation. Furthermore, by comparing the real response data of the physical sensors with the virtual response data of the digital twin model, the target digital twin model is dynamically corrected, enabling the target digital twin model to continuously absorb measured information during the physical test process, automatically calibrate model parameters and boundary conditions, and thus continuously approximate the real mechanical behavior of the physical entity. Ultimately, based on the corrected twin model, the experimental results were output. By constructing a closed-loop architecture of "modal analysis - frequency response function - digital twin modeling - physical sensor mapping - real-time excitation drive - dynamic correction", the deep coupling and synchronous evolution of the simulation model and the physical test object were realized, which significantly improved the accuracy of the simulation test results and enabled the digital twin model to accurately reflect the structural mechanical response of the target physical test object under real working conditions.

[0051] The following are system embodiments of this application, which can be used to execute the method embodiments of this application. For details not disclosed in the system embodiments of this application, please refer to the method embodiments of this application.

[0052] Please see Figure 3 This is a schematic diagram of the structure of the virtual test analysis system for structural mechanical properties provided in this application embodiment. This virtual test analysis system for structural mechanical properties can be implemented as all or part of a system through software, hardware, or a combination of both. The system includes a data acquisition module 11, a model construction module 12, a model optimization module 13, a virtual loading module 14, and a test analysis module 15.

[0053] The data acquisition module 11 is used to acquire the CAE model corresponding to the target physical test object and to acquire the target test parameters of the virtual test of the target physical test object; The model building module 12 is used to determine the natural frequencies and mode shapes of the target physical test object based on the CAE model through modal analysis, determine the target frequency response function of the target physical test object based on the natural frequencies and mode shapes, and determine the initial digital twin model of the target physical test object based on the target frequency response function. The model optimization module 13 is used to construct the mapping relationship between each physical sensor set on the target physical test object and the corresponding simulation model node in the initial digital twin model, and to map the boundary conditions in the target test parameters to the model boundary conditions of the initial digital twin model to obtain the target digital twin model. Virtual loading module 14 is used to acquire the actual excitation signal of the physical test bench where the target physical test object is located, and drive the target digital twin model to perform synchronous virtual loading based on the actual excitation signal to obtain virtual response data; The test analysis module 15 is used to correct the target digital twin model based on the real response data and virtual response data of each physical sensor, to obtain the corrected twin model, and based on the corrected twin model, to obtain the test results corresponding to the target physical test object.

[0054] Optional, model building module 12, specifically used for: The CAE model is discretized using the finite element method to obtain the structural dynamics equations corresponding to the target physical test object. Based on the structural dynamics equations, determine the mass matrix and stiffness matrix corresponding to the CAE model; Substituting the mass matrix and stiffness matrix into the preset characteristic equation of undamped free vibration, the natural frequencies and mode shapes corresponding to the target physical test object are obtained.

[0055] Optional, model building module 12, specifically used for: To obtain key parameters characterizing the materials and structures in the target physical test object; Based on key parameters and the target frequency response function, the initial digital twin model corresponding to the target physical test object is determined. The mathematical expression of the initial digital twin model is as follows: ; In the formula, Represents the frequency response function matrix. Indicates the excitation frequency. Represents the i-th natural frequency. Denotes the vector of the i-th mode shape. Let represent the damping ratio of the i-th mode, θ represent the vector corresponding to the key parameters, and j represent the imaginary unit.

[0056] Optional, such as Figure 4 As shown, the system also includes a mesh subdivision module 16, specifically used for: Obtain the stress-strain gradient of each region in the target digital twin model; When the stress-strain gradient exceeds a preset gradient threshold, the corresponding region is identified as the key region of the target digital twin model. By using a grid densification algorithm, key areas are subdivided into local grids to obtain a subdivided twin model.

[0057] Optionally, the data acquisition module 11 is specifically used for: When the target test parameters are sampling settings, acquire the historical locations where multiple historical objects have been damaged, and determine at least one location of concern from each historical location. The historical objects are of the same type as the target physical test objects. Obtain the historical usage scenarios of the historical objects when the parts of interest have been damaged, and determine at least one usage scenario of interest from each historical usage scenario; Assess the first risk value for damage to the areas requiring attention, and assess the second risk value for damage to each usage scenario requiring attention; Obtain the actual usage scenario of the target physical test object, and based on the actual usage scenario, the first risk value, and each second risk value, determine the sampling setting information of the virtual test of the target physical test object.

[0058] Optionally, the data acquisition module 11 is also used for: If there are actual usage scenarios among the various usage scenarios that need attention, then the parts that need attention will be determined as reference parts; Based on the first risk value of the reference part and the second risk value of the actual use scenario, a risk index for damage to the reference part is obtained in actual application. The risk index of each reference part is compared with the preset index threshold. When the risk index exceeds the index threshold, the corresponding reference part is determined as the sampling part. Based on the risk index of the sampling part, the sampling frequency of the sampling part is determined, and the sampling part and sampling frequency are determined as the sampling setting information.

[0059] Optionally, the data acquisition module 11 is also used for: When the target test parameter is the excitation location information, multiple historical excitation points that cover the area of ​​interest in the historical virtual test are obtained, and at least one excitation point of interest is determined from the multiple historical excitation points. Evaluate the weight of each incentive point that needs attention. The weight represents the probability that the incentive point will cover the area that needs attention. Multiply the first risk value of a single sampling location by the weights of the corresponding incentive points that need attention to obtain multiple multiplication results for a single sampling location; Summing the multiplication results of all sampling locations for the same stimulus point, we obtain the coverage index of the corresponding stimulus point. The coverage index represents the overall probability of covering the sampling location. If the coverage index exceeds the preset threshold, the corresponding incentive points that need attention will be determined as incentive location information.

[0060] It should be noted that the above-described virtual test analysis system for structural mechanical properties, when executing the virtual test analysis method for structural mechanical properties, is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the equipment can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the virtual test analysis system for structural mechanical properties and the embodiment of the virtual test analysis method for structural mechanical properties provided above belong to the same concept, and their implementation process is detailed in the method embodiment, which will not be repeated here.

[0061] This application also discloses a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements a virtual test analysis method for structural mechanical properties as described in the above embodiments.

[0062] The computer program can be stored in a computer-readable medium. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or certain middleware. The computer-readable medium includes any entity or system capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the computer-readable medium includes, but is not limited to, the above-mentioned components.

[0063] The above-described method for virtual test analysis of structural mechanical properties is stored in the computer-readable storage medium and loaded and executed on the processor to facilitate the storage and application of the method.

[0064] This application also discloses an electronic device in which a computer program is stored in a computer-readable storage medium. When the computer program is loaded and executed by a processor, it implements the above-mentioned virtual test analysis method for structural mechanical properties.

[0065] The electronic device can be a desktop computer, a laptop computer, or a cloud server, and includes, but is not limited to, a processor and a memory. For example, the electronic device may also include input / output devices, network access devices, and buses.

[0066] The processor can be a central processing unit (CPU). Of course, depending on the actual use, it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it.

[0067] The memory can be an internal storage unit of an electronic device, such as a hard disk or RAM, or an external storage device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD), or flash memory card (FC) equipped on the electronic device. Furthermore, the memory can be a combination of an internal storage unit and an external storage device. The memory is used to store computer programs and other programs and data required by the electronic device. The memory can also be used to temporarily store data that has been output or will be output. This application does not limit this.

[0068] In this electronic device, a virtual test analysis method for structural mechanical properties according to the above embodiment is stored in the memory of the electronic device and loaded and executed on the processor of the electronic device for convenient use.

[0069] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for virtual test analysis of structural mechanical properties, characterized in that, The method includes: Obtain the CAE model corresponding to the target physical test object, and obtain the target test parameters of the virtual test of the target physical test object; Based on the CAE model, the natural frequencies and mode shapes of the target physical test object are determined through modal analysis. Based on the natural frequencies and mode shapes, the target frequency response function of the target physical test object is determined. Based on the target frequency response function, the initial digital twin model of the target physical test object is determined. A mapping relationship is constructed between each physical sensor set on the target physical test object and the corresponding simulation model node in the initial digital twin model, and the boundary conditions in the target test parameters are mapped to the model boundary conditions of the initial digital twin model to obtain the target digital twin model; The actual excitation signal of the physical test bench where the target physical test object is located is obtained, and the target digital twin model is driven to perform synchronous virtual loading based on the actual excitation signal to obtain virtual response data; Based on the real response data and virtual response data of each physical sensor, the target digital twin model is modified to obtain a modified twin model, and based on the modified twin model, the test results corresponding to the target physical test object are obtained.

2. The method of claim 1, wherein, The determination of the natural frequencies and mode shapes of the target physical test object based on the CAE model through modal analysis specifically includes: The CAE model is discretized using the finite element method to obtain the structural dynamics equations corresponding to the target physical test object; Based on the structural dynamics equations, determine the mass matrix and stiffness matrix corresponding to the CAE model; Substituting the mass matrix and the stiffness matrix into the preset characteristic equation of undamped free vibration, the natural frequencies and mode shapes corresponding to the target physical test object are obtained.

3. The method of claim 1, wherein, The step of determining the initial digital twin model corresponding to the target physical test object based on the target frequency response function specifically includes: Obtain key parameters characterizing the materials and structure of the target physical test object; Based on the key parameters and the target frequency response function, an initial digital twin model corresponding to the target physical test object is determined. The mathematical expression of the initial digital twin model is as follows: ； wherein represents a frequency response function matrix, represents an excitation frequency, represents the i-th order natural frequency, represents the i-th order modal shape vector, represents the i-th order modal damping ratio, θ represents a vector corresponding to the key parameter, and j represents an imaginary unit.

4. The virtual test analysis method for structural mechanical properties according to claim 1, characterized in that, After obtaining virtual response data by synchronously virtual loading of the target digital twin model based on actual excitation signals, the process further includes: Obtain the stress-strain gradient of each region in the target digital twin model; When the stress-strain gradient exceeds a preset gradient threshold, the corresponding region is identified as the key region of the target digital twin model. The key area is subdivided into local grids using a grid encryption algorithm to obtain a subdivided twin model. The process of correcting the target digital twin model based on the real and virtual response data from each physical sensor to obtain the corrected twin model specifically includes: Based on the real and virtual response data of each physical sensor, the subdivided twin model is corrected to obtain the corrected twin model.

5. The virtual test analysis method for structural mechanical properties according to claim 1, characterized in that, The acquisition of the target test parameters for the virtual test of the target physical test object specifically includes: When the target test parameters are sampling setting information, multiple historical locations where damage has occurred on historical objects are obtained, and at least one location requiring attention is determined from each of the historical locations. The historical objects are of the same type as the target physical test object. When the area of ​​concern has been damaged, the historical usage scenarios of the historical object are obtained, and at least one usage scenario of concern is determined from each of the historical usage scenarios; Assess a first risk value for damage to the areas of concern, and assess a second risk value for damage to each of the usage scenarios of concern; Obtain the actual usage scenario of the target physical test object, and based on the actual usage scenario, the first risk value, and each of the second risk values, determine the sampling setting information of the virtual test of the target physical test object.

6. The virtual test analysis method for structural mechanical properties according to claim 5, characterized in that, The step of determining the sampling settings information for the virtual test of the target physical test object based on the actual usage scenario, the first risk value, and each of the second risk values ​​specifically includes: If the actual usage scenario mentioned above exists in each of the aforementioned usage scenarios that require attention, then the part that requires attention will be determined as the reference part. Based on the first risk value of the reference part and the second risk value of the actual use scenario, a risk index for damage to the reference part during actual application is obtained. The risk index of each reference part is compared with a preset index threshold. When the risk index exceeds the index threshold, the corresponding reference part is determined as a sampling part. Based on the risk index of the sampling part, the sampling frequency of the sampling part is determined, and the sampling part and the sampling frequency are determined as sampling setting information.

7. The virtual test analysis method for structural mechanical properties according to claim 5, characterized in that, The acquisition of the target test parameters for the virtual test of the target physical test object specifically includes: When the target test parameter is the excitation location information, multiple historical excitation points that cover the part of interest in the historical virtual test are obtained, and at least one excitation point of interest is determined from the multiple historical excitation points. Evaluate the weight of each of the aforementioned stimulus points, whereby the weight represents the likelihood that the stimulus point will cover the area of ​​interest. Multiply the first risk value of a single sampling location by the weights of the corresponding incentive points that need attention to obtain multiple multiplication results corresponding to a single sampling location; Summing the multiplication results corresponding to all the sampling locations, the multiplication results involving the same stimulus point of interest are obtained to get the coverage index of the corresponding stimulus point of interest. The coverage index represents the overall probability of covering the sampling location. If the coverage index exceeds a preset threshold, the corresponding incentive point to be focused on will be determined as incentive location information.

8. A virtual test analysis system for structural mechanical properties, characterized in that, include: The data acquisition module (11) is used to acquire the CAE model corresponding to the target physical test object and to acquire the target test parameters of the virtual test of the target physical test object; The model building module (12) is used to determine the natural frequency and mode shape of the target physical test object through modal analysis based on the CAE model, determine the target frequency response function of the target physical test object based on the natural frequency and mode shape, and determine the initial digital twin model of the target physical test object based on the target frequency response function. The model optimization module (13) is used to construct the mapping relationship between each physical sensor set on the target physical test object and the corresponding simulation model node in the initial digital twin model, and to map the boundary conditions in the target test parameters to the model boundary conditions of the initial digital twin model to obtain the target digital twin model. The virtual loading module (14) is used to obtain the actual excitation signal of the physical test bench where the target physical test object is located, and drive the target digital twin model to perform synchronous virtual loading based on the actual excitation signal to obtain virtual response data; The test analysis module (15) is used to modify the target digital twin model based on the real response data and the virtual response data of each physical sensor to obtain the modified twin model, and to obtain the test results corresponding to the target physical test object based on the modified twin model.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it implements the method of any one of claims 1-7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor loads and executes the computer program, it implements the method of any one of claims 1-7.

Images