Anti-collision wall stress test digital twin system and method

Abstract

The invention relates to the technical field of computer simulation, in particular to an anti-collision wall stress test digital twin system and method. According to the system, a physical anti-collision wall module is provided with a plurality of sensors to collect actual data; the data acquisition module is responsible for preprocessing and transmitting sensor data; the data processing module simulates, analyzes and optimizes the stress condition of the anti-collision wall by constructing and calibrating a parameterized finite element model and combining an artificial intelligence algorithm; the user interaction module realizes result display and parameter input interaction; the method is based on the system and covers a complete process from model construction to result output and feedback. According to the invention, high-precision and high-real-time mechanical property test and evaluation of the anti-collision wall can be realized, the test cost can be effectively reduced, the efficiency can be improved, powerful support is provided for structural safety guarantee and optimization design of the anti-collision wall, and the method is suitable for structural safety detection and performance prediction of the anti-collision wall in scenes such as highways, bridges, tunnels and the like.

Description

Technical Field

[0001] The present invention relates to the field of computer simulation technology, and in particular to a digital twin system and method for force testing of a crash barrier. Background Art

[0002] As a key protective component for road traffic, industrial plants, and critical infrastructure, crash barriers' mechanical properties during accidental impacts are directly related to personnel safety and structural integrity. Traditional crash barrier stress testing methods rely primarily on physical vehicle crash tests or bench tests. These methods are not only costly, time-consuming, and labor-intensive, requiring demanding site conditions, but also inherently dangerous, making it difficult to conduct a comprehensive and systematic assessment of multiple operating conditions. Furthermore, the limited number of sensors deployed during physical testing makes it difficult to obtain a complete picture of the stress and strain distribution and damage evolution within the structure.

[0003] In recent years, numerical simulation techniques such as finite element analysis have been widely used in structural crash simulation. However, the accuracy of pure numerical simulations is highly dependent on the precision of model parameters, which often contain uncertainties, resulting in significant deviations between simulation results and actual conditions. To improve the predictive accuracy of simulation models, model calibration is often required, but this presents challenges in the absence of sufficient experimental data.

[0004] Digital twin technology bridges the physical and virtual worlds, integrating real-time data from physical entities, simulation models, and intelligent algorithms to provide a new solution for monitoring, analyzing, predicting, and optimizing complex systems. Digital twin technology leverages real-time sensor data from physical objects to simulate behavior and monitor operations throughout their lifecycle.

[0005] Applying digital twin technology to stress testing of crash barriers has the potential to overcome the limitations of traditional physical testing and simple numerical simulation. By constructing a high-fidelity digital twin model of the crash barrier and calibrating it with a small amount of targeted physical test data, it is possible to accurately predict the mechanical response of the crash barrier under various virtual impact conditions. This not only significantly reduces testing costs and cycle time, improving testing efficiency, but also provides insights into the damage mechanisms and weaknesses of the crash barrier, providing strong support for its optimized design and safety assessment. Summary of the Invention

[0006] The present invention provides a digital twin system and method for crash barrier force testing. Through the deep integration of physical entities and virtual models, combined with artificial intelligence algorithms, high-precision simulation, real-time monitoring and intelligent optimization of the mechanical properties of crash barriers can be achieved, solving the problems of high cost, low precision and poor real-time performance in the existing technology.

[0007] A digital twin system for crash wall stress testing, comprising: Physical crash barrier module: A physical crash barrier installed in a pre-set test site or actual application scenario, wherein a sensor network is deployed on the physical crash barrier, and the sensor network includes at least one selected from a strain sensor, an acceleration sensor, a displacement sensor, and a temperature sensor; Data acquisition module: connected to the sensor network, used to collect sensor data from the sensor network and pre-process the data; Data processing module: connected to the data acquisition module, used to receive, process and analyze sensor data and drive the operation of the digital twin model; User interaction module: provides a graphical user interface for displaying the simulation result data and the optimization solution generated by the artificial intelligence, and receives virtual impact parameters input by the user.

[0008] Preferably, the data processing module includes: Model construction and maintenance unit: constructs and maintains a parameterized finite element model corresponding to the physical crash barrier as the digital twin core model of the physical crash barrier. The parameterized finite element model includes a macrostructure model and a micromechanical model. Data receiving and preprocessing unit: receives the sensor data transmitted by the data acquisition module and performs preprocessing operations; Model calibration unit: performing preliminary physical testing to obtain baseline sensor data, the preliminary physical testing comprising at least one of nondestructive testing, static loading testing, modal testing, or low-energy impact testing; simulating the working conditions of the preliminary physical test on the digital twin core model to obtain an initial simulation response; comparing the initial simulated response to the baseline sensor data; Iteratively adjusting one or more preset parameters of the digital twin core model using an optimization algorithm until the difference between a subsequent simulation response of the digital twin core model and the baseline sensor data is less than a preset threshold, thereby obtaining a calibrated digital twin model; An impact simulation unit is configured to simulate, using the calibrated digital twin model, a mechanical response of the physical crash barrier under an impact event defined by user-defined virtual impact parameters, wherein the virtual impact parameters include at least one of the impactor velocity, mass, angle, and surface material properties; Result generation and analysis unit: generates and analyzes the simulation result data of the mechanical response and extracts key performance indicators; Artificial intelligence-assisted decision-making unit: Based on the deep learning convolutional neural network (CNN), feature extraction and pattern recognition are performed on the sensor data and simulation result data to automatically identify abnormal stress areas and potential structural weaknesses; Predicting the performance of the physical crash barrier under future working conditions using a long short-term memory network (LSTM); Genetic algorithms are used to generate optimization plans including material replacement and structural reinforcement locations; Multi-scale modeling and updating unit: A micromechanical model is established for the local structural weaknesses of the crash barrier, and updated in coordination with the macrostructural model; The physical sensor data and virtual sensor node data are fused based on the Kalman filter algorithm, and the update time interval and weight distribution are automatically adjusted according to the data change frequency and fluctuation characteristics. The calculation formula is: ; Where, for The updated weight at time, is the current data standard deviation, The largest standard deviation in history, is the frequency of data change, To preset the maximum frequency, and is an adjustable coefficient, and .

[0009] Preferably, the artificial intelligence-assisted decision-making unit includes: Data preprocessing layer: normalizes and reduces noise on raw sensor data; Feature extraction layer: automatically extract the spatiotemporal features of force data through the CNN network; Classification and recognition layer: Identify different types of abnormal stress patterns based on the Softmax classifier; Explanation generation layer: SHAP method is used to generate interpretable risk assessment reports.

[0010] A digital twin method for force testing of a crash wall, comprising: Construct and maintain a parametric finite element model: Construct and maintain a parametric finite element model corresponding to a physical crash barrier as the core digital twin model of the physical crash barrier, wherein the physical crash barrier is provided with a sensor network, wherein the sensor network includes at least one selected from a strain sensor, an acceleration sensor, a displacement sensor, and a temperature sensor; Receiving and preprocessing sensor data: receiving sensor data from a data acquisition module connected to the sensor network and preprocessing the sensor data; Model calibration: performing, by the model calibration unit, a preliminary physical test on the physical crash barrier to obtain baseline sensor data, the preliminary physical test comprising at least one of a nondestructive test, a static load test, a modal test, or a low-energy impact test; simulating the working conditions of the preliminary physical test on the digital twin core model to obtain an initial simulation response; comparing the initial simulated response to the baseline sensor data; Iteratively adjusting one or more preset parameters of the digital twin core model using an optimization algorithm until the difference between a subsequent simulation response of the digital twin core model and the baseline sensor data is less than a preset threshold, thereby obtaining a calibrated digital twin model; Receiving user-defined virtual impact parameters: receiving virtual impact parameters input by the user, wherein the virtual impact parameters include at least one of the impactor speed, mass, angle, and surface material characteristics; Performing an impact simulation process: simulating the mechanical response of the physical crash barrier under the impact event defined by the virtual impact parameters using the calibrated digital twin model; Artificial intelligence-assisted analysis and prediction: By means of the artificial intelligence-assisted decision-making unit, feature extraction and anomaly identification are performed on the sensor data and simulation result data based on a deep learning CNN network to generate a heat map containing risk levels and locations; The force trend under future working conditions is predicted by using the long short-term memory network (LSTM), and a genetic algorithm is used to generate an optimization plan for the crash barrier structure. Analyzing and outputting simulation result data: analyzing and outputting the simulation result data of the mechanical response to evaluate the stress performance of the physical crash barrier; Result visualization and feedback: The simulation result data and the optimization plan generated by the artificial intelligence are visualized through a graphical user interface, and user feedback is received.

[0011] Preferably, the model calibration further comprises a microstructure response updating sub-step: Establish a micromechanical model for the critical connection parts and material defect areas of the crash barrier; According to the macroscopic stress field distribution, the microscopic model is updated through multi-scale calculation methods; The microstructural evolution results are fed back into the macroscopic model to correct the overall mechanical property predictions.

[0012] Preferably, the artificial intelligence-assisted analysis and prediction further includes: Construct a generative adversarial network (GAN) to generate virtual test data with different damage levels and expand training samples; A transfer learning method is used to apply the trained model to the performance evaluation of different types of crash barriers, reducing training time and data requirements.

[0013] Preferably, the digital twin method for crash barrier force testing further comprises the following steps: synchronously collecting real-time sensor data when performing a physical impact test on the physical crash barrier; and using the real-time sensor data to further calibrate the digital twin model; The simulation results of the calibrated digital twin model under the same working conditions of the physical impact test are compared and verified with the real-time sensor data, and the simulation results are used to provide internal state information of the physical crash barrier in the area not covered by the sensor.

[0014] Compared with the prior art, the advantages of the present invention are: Improve test efficiency and accuracy: Through virtual model simulation and analysis, the scale of physical experiments can be reduced, effectively lowering testing costs and shortening testing cycles. At the same time, with the help of precise model construction and calibration mechanisms, high-precision simulation and prediction of the mechanical properties of the crash barrier can be achieved, providing reliable support for its structural safety assessment.

[0015] Enhanced intelligent decision-making capabilities: Leveraging AI technologies such as deep learning, the system monitors and quickly identifies abnormal stress conditions in crash barriers in real time. Based on the analysis results, it generates specific, actionable, and quantitative optimization solutions, effectively assisting in the iterative design of crash barrier structures and improving their overall performance and safety.

[0016] Comprehensive evaluation and deep insights: It can simulate areas where sensors are difficult to deploy or physical quantities that are difficult to measure directly in physical experiments, providing a more comprehensive understanding of the failure mechanism of the crash barrier. DETAILED DESCRIPTION

[0017] To make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In one embodiment, a digital twin system for crash barrier force testing is provided, including a physical crash barrier module, a data acquisition module, a data processing module, and a user interaction module. The specific implementation details of each module are described in detail below: Physical crash barrier module: In actual application scenarios, such as crash barriers on highways, bridges, or tunnels, physical crash barriers are installed. A sensor network is carefully laid out on this physical crash barrier, and sensor types include strain sensors, accelerometers, displacement sensors, and temperature sensors. Strain sensors can be attached to key stress-bearing parts of the crash barrier, such as corners and connection points, to measure the strain of the wall during stress. Accelerometers are distributed in different locations to capture changes in acceleration at the moment of impact or under other dynamic conditions. Displacement sensors are responsible for monitoring possible displacements of the wall. Temperature sensors are used to obtain changes in ambient temperature and the wall's own temperature, as temperature has a certain impact on material properties and structural mechanical characteristics. These sensors work together to comprehensively collect various raw data reflecting the physical state of the crash barrier.

[0019] Data Acquisition Module: This module connects to the sensor network on the physical crash barrier module through wired or wireless connections. This module receives raw data from each sensor in real time and immediately performs preprocessing on this data. For example, analog sensor data is converted to digital signals using an analog-to-digital converter to facilitate subsequent transmission and processing. Furthermore, filtering algorithms (such as mean filtering and Kalman filtering) are used to remove noise and improve data quality, ensuring accurate and reliable data transmitted to the data processing module.

[0020] Data processing module: As the core hub of the entire system, the data processing module has multiple functional units that work together to realize the complex functions of the system. The details are as follows: Model building and maintenance unit: First, a parametric finite element model corresponding to the physical crash barrier was constructed using specialized finite element modeling software, based on the actual geometric dimensions of the physical crash barrier, material properties (such as the strength grade of the concrete and the elastic modulus of the steel), and detailed parameters such as the connection method. This model includes a macrostructural model to simulate the mechanical behavior of the entire crash barrier at a macro level, such as overall deformation and stress distribution; it also includes a micromechanical model that focuses on the mechanical response of the crash barrier's internal materials at the microstructural level, such as the propagation of microcracks in the concrete and the evolution of microscopic damage in the steel.

[0021] During system operation, the parametric finite element model is continuously maintained based on subsequent test data and model update requirements, such as timely updating of material parameters, geometric dimension changes and other information to ensure that the model can always accurately reflect the actual status of the physical crash barrier.

[0022] The data reception and preprocessing unit receives pre-processed sensor data from the data acquisition module via a pre-set data interface. Based on this, it further performs unified format conversion to ensure that all data conforms to the format requirements for subsequent processing and analysis. For example, data from different sensors is standardized according to timestamps and data types. Furthermore, data integrity checks are performed again, and missing or abnormal data points are interpolated or repaired to ensure data continuity and availability.

[0023] Model calibration unit:

[0024] Preliminary physical testing: A series of preliminary physical tests are conducted on the physical crash barrier, including non-destructive testing, static loading testing, modal testing or low-energy impact testing, to obtain baseline sensor data.

[0025] Simulation and comparative analysis: The operating conditions of the initial physical test are accurately replicated on the constructed digital twin core model (i.e., the parameterized finite element model). Finite element simulation is used to obtain the corresponding initial simulated responses. These initial simulated responses are then compared with the actual baseline sensor data collected to analyze the differences between the two.

[0026] Iterative parameter adjustment: Optimization algorithms (such as genetic algorithms and particle swarm optimization) are used to iteratively adjust one or more preset parameters of the core digital twin model based on these discrepancies. This process continues until the difference between the subsequent simulation response of the core digital twin model and the baseline sensor data is less than a preset threshold (e.g., a root mean square error of less than 5%). This results in a calibrated digital twin model that accurately simulates the actual mechanical properties of the physical crash barrier.

[0027] Impact simulation unit: Based on the virtual impact parameters input by the user through the user interaction module, these parameters specifically cover key factors such as the impact object's speed, mass, angle, and surface material properties. Utilizing the calibrated digital twin model and the dynamic analysis function of the finite element simulation software, the mechanical response of the physical crash barrier under the impact event defined by these virtual impact parameters is simulated, including but not limited to changes in stress distribution, the location of the strain peak, the overall deformation shape, and energy absorption.

[0028] The Result Generation and Analysis Unit conducts in-depth analysis of the mechanical response simulation data obtained by the Impact Simulation Unit, extracting a series of key performance indicators. For example, a post-processing algorithm calculates a stress distribution cloud map, visually displaying the stress magnitude and distribution of various parts of the crash barrier. The strain distribution is analyzed to identify strain concentration areas and determine potential damage locations. Deformation patterns are analyzed to observe the overall and local deformation characteristics of the crash barrier under impact. The damage level is assessed, and the presence and severity of damage are determined based on specific damage determination criteria (such as strain energy density and stress intensity factor). Energy absorption characteristics are analyzed to understand the crash barrier's ability to absorb and dissipate energy during an impact. Furthermore, indicators such as the penetration depth of the impacting object are also considered to comprehensively determine the crash barrier's protective performance.

[0029] Artificial intelligence-assisted decision-making unit: Data preprocessing layer: First, the original sensor data and simulation result data are normalized, and the data is mapped to a specific interval to eliminate the impact of differences between different data magnitudes; at the same time, the noise reduction algorithm is used to further remove residual noise components in the data, improve data quality, and lay a good foundation for subsequent deep learning analysis.

[0030] Feature extraction layer: The preprocessed data is input into the CNN network, and the network automatically learns and extracts the spatiotemporal characteristics of the force data, such as the stress change characteristics and strain change laws at different times and positions, to form a representative feature vector.

[0031] Classification and recognition layer: Based on the extracted feature vectors, the Softmax classifier is used to identify and classify different types of stress anomaly patterns. For example, it can distinguish whether the anomaly is caused by local stress concentration or due to internal defects in the material. At the same time, the corresponding probability value is given to facilitate accurate judgment of the category and probability of the anomaly.

[0032] Explanation Generation Layer: Using the SHAP method, the decisions made by the AI ​​model are explained, generating an explainable risk assessment report that clearly demonstrates the contribution of each input factor to the final risk assessment result, allowing users to intuitively understand the basis and process of the model's decision-making.

[0033] Multi-scale modeling and updating unit: Establishment and collaborative updating of micromechanical models: Based on the local structural weaknesses discovered through analysis of the crash barrier, a micromechanical model is established using the theory and methods of micromechanics to describe in detail the microstructural changes and damage evolution mechanisms of the local area during the stress process. Then, the micromechanical model is collaboratively updated with the macrostructural model to achieve the mutual correlation and information transmission between the overall mechanical response at the macro level and the local damage evolution at the micro level. For example, when the macroscopic stress field distribution changes, the micromodel is driven to update accordingly based on the multi-scale calculation method. At the same time, the microstructural evolution results are fed back to the macromodel to correct the overall mechanical properties predicted by the macromodel, ensuring that the entire model system can accurately reflect the changes in the real mechanical state of the crash barrier from macro to micro.

[0034] Data fusion and weight adjustment: The physical sensor data and virtual sensor node data are fused based on the Kalman filter algorithm. In the actual operation process, the frequency of data changes (for example, by counting the number of data fluctuations per unit time) and the fluctuation characteristics (such as the standard deviation of the data reflects the fluctuation amplitude) are monitored in real time. According to the formula , automatically adjust the update time interval and weight distribution. Among them, and is an adjustable coefficient (which can be set based on actual experience or preliminary test results, for example , ), is the current data standard deviation, The largest standard deviation in history, is the frequency of data change, This dynamic weight adjustment mechanism allows the model to rationally integrate multi-source data under different working conditions, achieving more accurate updates and simulations.

[0035] User Interaction Module: This module enables user-friendly interaction through a graphical user interface. It clearly displays simulation results (such as visualizations and animations of key performance indicators) generated by the result generation and analysis unit in the data processing module, along with optimization solutions generated by the AI-assisted decision-making unit, to the user, facilitating a comprehensive understanding of the crash barrier's mechanical performance and potential areas for improvement. Furthermore, it provides an input interface for user-entered virtual impact parameters, allowing users to flexibly set different impact conditions based on actual needs, thereby driving the impact simulation unit in the data processing module to carry out the corresponding simulation analysis.

[0036] In another embodiment, the following steps are provided for implementing the method flow corresponding to the system: Build and maintain a parametric finite element model: Using professional finite element modeling software, we combined the physical crash barrier's actual engineering drawings, material parameters, and detailed information about the installation environment to construct a corresponding parametric finite element model. This model encompasses both a macrostructural model and a micromechanical model. Throughout the system's operation, we continuously monitor potential structural changes and material degradation in the physical crash barrier, promptly updating and maintaining relevant model parameters to ensure the model accurately reflects the physical entity.

[0037] Receiving and preprocessing sensor data: The data acquisition module establishes a stable communication connection with the sensor network on the physical crash barrier module, receiving all types of raw sensor data in real time. Following the established data preprocessing process, the sensor data is preprocessed using analog-to-digital conversion, filtering algorithms, data format unification, integrity checks, and repairs. This ensures data accuracy and usability, providing a high-quality data foundation for subsequent model calibration and other analytical tasks.

[0038] Model calibration: Conduct preliminary physical testing: Perform various preliminary physical testing methods, including non-destructive testing, static loading testing, modal testing, and low-energy impact testing, on the physical crash barrier in accordance with regulatory requirements. This will obtain rich baseline sensor data and record the various parameter settings and corresponding sensor response data during each test in detail, providing a practical reference for subsequent model calibration.

[0039] Simulate working conditions and obtain initial simulation responses: The actual working conditions of the above preliminary physical tests are fully input into the constructed digital twin core model. The mechanical response under the corresponding working conditions is accurately simulated through finite element simulation software to obtain the initial simulation response data.

[0040] Comparative Analysis and Parameter Adjustment: Initial simulation response data is compared with actual baseline sensor data, and the model's accuracy is evaluated by calculating differential metrics (such as root mean square error and relative error). Subsequently, an optimization algorithm is used to iteratively adjust key parameters in the core digital twin model until the difference between the model's subsequent simulation response and the baseline sensor data falls below a pre-set threshold. This results in a calibrated digital twin model that accurately simulates the actual mechanical behavior of the physical crash barrier.

[0041] Receive user-defined virtual impact parameters: Through the graphical user interface provided by the user interaction module, the virtual impact parameters input by the user are received. These parameters cover multiple aspects such as the impactor's speed, mass, angle, and surface material properties. Users can flexibly set different virtual impact parameter combinations based on actual research needs, simulation scenarios, or safety assessment goals, so as to comprehensively examine the mechanical performance of the crash barrier under various possible impact conditions.

[0042] Execute the impact simulation process: The virtual impact parameters entered by the user are passed to the calibrated digital twin model. Using the dynamic analysis function of the finite element simulation software, the impact simulation calculation is started to simulate the complete mechanical response process of the physical crash barrier under the impact event defined by the corresponding virtual impact parameters. This includes the detailed mechanical behavior of each stage, from the propagation of stress waves at the moment of impact to the deformation development of the entire crash barrier structure and the final energy absorption and dissipation, to obtain comprehensive and detailed mechanical response simulation result data.

[0043] Artificial intelligence-assisted analysis and prediction: Feature extraction and anomaly identification: The preprocessed sensor data and the result data obtained from the impact simulation are input into the deep learning convolutional neural network in the artificial intelligence-assisted decision-making unit. With the powerful feature extraction capability of CNN, the spatiotemporal features in these data are automatically learned and extracted. Based on the extracted features, the classification and recognition algorithm is used to identify and judge abnormal stress areas and potential structural weaknesses, and generate a heat map containing risk levels and locations to intuitively display the distribution and severity of abnormal situations.

[0044] Performance Prediction and Optimization Solution Generation: Utilizing long-short-term memory networks (LSTMs) to analyze and learn from historical sensor data and current simulation results, the system predicts the stress trends of the physical crash barrier over a specific timeframe, providing early insight into potential safety hazards. Furthermore, incorporating genetic algorithms, the system generates optimization solutions with specific content, including recommendations for material replacement and structural reinforcement locations, with the goal of improving the crash barrier's mechanical properties and reducing damage risk. This provides strong decision-making support for the subsequent maintenance, renovation, and optimized design of the crash barrier.

[0045] Analyze and output simulation result data: Analyze the mechanical response simulation result data obtained during the impact simulation process, extract key performance indicators such as stress distribution, strain distribution, deformation mode, damage level, energy absorption characteristics, and impactor penetration depth, and present these indicators in an intuitive and easy-to-understand form through data visualization technology, so that users can fully understand the mechanical performance of the crash barrier under specific impact conditions.

[0046] Results visualization and feedback: The user interaction module's graphical user interface visualizes the simulation results obtained from the above analysis and the optimization solutions generated by the AI-assisted decision-making unit. These data are presented to users in the form of charts, graphs, and text descriptions, allowing them to clearly understand the stress conditions, potential risks, and improvement directions of the crash barrier. At the same time, a corresponding feedback channel is set up on the interface to receive user feedback such as questions, opinions, or further analysis requirements regarding the displayed content. Based on user feedback, the system's parameter settings, simulation conditions, or analysis methods are adjusted in a timely manner, achieving good interaction with users and continuously optimizing the application effects of the entire crash barrier stress test digital twin system and method.

[0047] Through the specific implementation of the above-mentioned system architecture and method flow, the digital twin system and method for force testing of crash barriers of the present invention can efficiently and accurately realize comprehensive testing, analysis and optimization decision-making of the mechanical properties of crash barriers, providing strong technical support for the safety assurance and performance improvement of crash barriers.

[0048] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0049] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A digital twin system for crash wall stress testing, characterized in that: include: Physical crash barrier module: A physical crash barrier installed in a pre-set test site or actual application scenario, wherein a sensor network is deployed on the physical crash barrier, and the sensor network includes at least one selected from a strain sensor, an acceleration sensor, a displacement sensor, and a temperature sensor; Data acquisition module: connected to the sensor network, used to collect sensor data from the sensor network and pre-process the data; Data processing module: connected to the data acquisition module, used to receive, process and analyze sensor data and drive the operation of the digital twin model; User interaction module: provides a graphical user interface for displaying the simulation result data and the optimization solution generated by the artificial intelligence, and receives virtual impact parameters input by the user.

2. A digital twin system for crash wall stress testing according to claim 1, characterized in that: The data processing module includes a model building and maintenance unit, a data receiving and preprocessing unit, a model calibration unit, an impact simulation unit, and a result generation and analysis unit; The model construction and maintenance unit is used to construct and maintain a parameterized finite element model corresponding to the physical crash barrier as the digital twin core model of the physical crash barrier, wherein the parameterized finite element model includes a macrostructure model and a micromechanical model; The data receiving and preprocessing unit is used to receive the sensor data transmitted by the data acquisition module and perform preprocessing operations; The model calibration unit is used to adjust one or more preset parameters of the digital twin core model using an optimization algorithm to calibrate the digital twin model; The impact simulation unit is configured to simulate, using the calibrated digital twin model, a mechanical response of the physical crash barrier under an impact event defined by user-defined virtual impact parameters, wherein the virtual impact parameters include at least one of impactor velocity, mass, angle, and surface material properties; The result generation and analysis unit is used to generate and analyze the simulation result data of the mechanical response and extract key performance indicators, wherein the key performance indicators include at least one of stress distribution, strain distribution, deformation mode, damage degree, energy absorption characteristics or impactor penetration depth.

3. A digital twin system for crash wall stress testing according to claim 1, characterized in that: The step of adjusting one or more preset parameters of the digital twin core model using an optimization algorithm to calibrate the digital twin model includes: performing preliminary physical testing to obtain baseline sensor data, the preliminary physical testing comprising at least one of nondestructive testing, static loading testing, modal testing, or low-energy impact testing; simulating the working conditions of the preliminary physical test on the digital twin core model to obtain an initial simulation response; comparing the initial simulated response to the baseline sensor data; An optimization algorithm is used to iteratively adjust one or more preset parameters of the digital twin core model until the difference between the subsequent simulation response of the digital twin core model and the baseline sensor data is less than a preset threshold, thereby obtaining a calibrated digital twin model.

4. A digital twin system for crash wall stress testing according to claim 1, characterized in that: The data processing module also includes an artificial intelligence-assisted decision-making unit: Based on the deep learning convolutional neural network (CNN), feature extraction and pattern recognition are performed on the sensor data and simulation result data to automatically identify abnormal stress areas and potential structural weaknesses; Predicting the performance of the physical crash barrier under future working conditions using a long short-term memory network (LSTM); Genetic algorithms are used to generate optimization solutions including material replacement and structural reinforcement locations.

5. A digital twin system for crash wall stress testing according to claim 4, characterized in that: The artificial intelligence-assisted decision-making unit includes: Data preprocessing layer: normalizes and reduces noise on raw sensor data; Feature extraction layer: automatically extract the spatiotemporal features of force data through the CNN network; Classification and recognition layer: Identify different types of abnormal stress patterns based on the Softmax classifier; Explanation generation layer: SHAP method is used to generate interpretable risk assessment reports.

6. A digital twin system for crash wall stress testing according to claim 1, characterized in that: The data processing module further includes a multi-scale modeling and updating unit: A micromechanical model is established for the local structural weaknesses of the crash barrier, and updated in coordination with the macrostructural model; The physical sensor data and virtual sensor node data are fused based on the Kalman filter algorithm, and the update time interval and weight distribution are automatically adjusted according to the data change frequency and fluctuation characteristics. The calculation formula is: ; Where, for The updated weight at time, is the current data standard deviation, The largest standard deviation in history, is the frequency of data change, To preset the maximum frequency, and is an adjustable coefficient, and .

7. A digital twin method for force testing of a crash wall, characterized in that: include: Construct and maintain a parametric finite element model: Construct and maintain a parametric finite element model corresponding to a physical crash barrier as the core digital twin model of the physical crash barrier, wherein the physical crash barrier is provided with a sensor network, wherein the sensor network includes at least one selected from a strain sensor, an acceleration sensor, a displacement sensor, and a temperature sensor; Receiving and preprocessing sensor data: receiving sensor data from a data acquisition module connected to the sensor network and preprocessing the sensor data; Model calibration: performing, by the model calibration unit, a preliminary physical test on the physical crash barrier to obtain baseline sensor data, the preliminary physical test comprising at least one of a nondestructive test, a static load test, a modal test, or a low-energy impact test; simulating the working conditions of the preliminary physical test on the digital twin core model to obtain an initial simulation response; comparing the initial simulated response to the baseline sensor data; Iteratively adjusting one or more preset parameters of the digital twin core model using an optimization algorithm until the difference between a subsequent simulation response of the digital twin core model and the baseline sensor data is less than a preset threshold, thereby obtaining a calibrated digital twin model; Receiving user-defined virtual impact parameters: receiving virtual impact parameters input by the user, wherein the virtual impact parameters include at least one of the impactor speed, mass, angle, and surface material characteristics; Performing an impact simulation process: simulating the mechanical response of the physical crash barrier under the impact event defined by the virtual impact parameters using the calibrated digital twin model; Artificial intelligence-assisted analysis and prediction: By means of the artificial intelligence-assisted decision-making unit, feature extraction and anomaly identification are performed on the sensor data and simulation result data based on a deep learning CNN network to generate a heat map containing risk levels and locations; The force trend under future working conditions is predicted by using the long short-term memory network (LSTM), and a genetic algorithm is used to generate an optimization plan for the crash barrier structure. Analyzing and outputting simulation result data: analyzing and outputting the simulation result data of the mechanical response to evaluate the stress performance of the physical crash barrier; Result visualization and feedback: The simulation result data and the optimization plan generated by the artificial intelligence are visualized through a graphical user interface, and user feedback is received.

8. A digital twin method for crash barrier stress testing according to claim 7, characterized in that: The model calibration further includes a microstructure response updating step: Establish a micromechanical model for the critical connection parts and material defect areas of the crash barrier; According to the macroscopic stress field distribution, the microscopic model is updated through multi-scale calculation methods; The microstructural evolution results are fed back into the macroscopic model to correct the overall mechanical property predictions.

9. A digital twin method for crash barrier stress testing according to claim 7, characterized in that: The artificial intelligence-assisted analysis and prediction further includes: Construct a generative adversarial network (GAN) to generate virtual test data with different damage levels and expand training samples; A transfer learning method is used to apply the trained model to the performance evaluation of different types of crash barriers, reducing training time and data requirements.

10. A digital twin method for crash barrier stress testing according to claim 7, characterized in that: Further including: When conducting a physical impact test of the physical crash barrier, synchronously collecting real-time sensor data; Using the real-time sensor data to further calibrate the digital twin model; The simulation results of the calibrated digital twin model under the same working conditions of the physical impact test are compared and verified with the real-time sensor data, and the simulation results are used to provide internal state information of the physical crash barrier in the area not covered by the sensor.

Images