A digital twin system for industrial CT system and its construction method

Abstract

The present invention provides a digital twin system for industrial CT system and a construction method thereof, belonging to the field of digital twin. The framework includes four parts: a physical space module, a virtual space module, a function and application layer, and a connection module; the physical space module is a three-dimensional field, including physical entities, physical behaviors and real data; the virtual space module is a mirror image of the physical space module, which is the same as the physical space module, and is also a three-dimensional field, corresponding to the physical entities, physical behaviors and real data in the physical space module, and the virtual space module includes digital models, simulation behaviors and virtual data. The present invention also proposes a construction method, including: step 1, defining the physical space module; step 2, constructing the virtual space module through the static connection module; step 3, realizing the function and application layer through the dynamic connection module. The present invention can obtain the best scanning mode and parameters, reduce the dependence on professional knowledge, and optimize the parameter setting to ensure scanning consistency, improve imaging efficiency and repeatability of results.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital twins, and in particular relates to a digital twin system for an industrial CT system and a construction method thereof. Background Art

[0002] Industrial computed tomography (CT) is an advanced imaging technology for industrial applications that uses the penetrating power of X-rays to capture data from different angles and reconstruct a three-dimensional image of an object. Industrial CT is a typical non-destructive testing technology that can reveal the internal structure of an object without physically dissecting or disassembling it.

[0003] However, although traditional industrial CT technology is quite mature and has been widely used in many fields such as aerospace, automobile manufacturing and electronics, there are still some problems:

[0004] The operation process and parameter settings are complex: This complexity requires the operator to have professional training and expertise. For different materials or workpieces, the scanning methods and scanning parameters required to achieve the best imaging effect are different, which will lead to low efficiency and inconsistent imaging results.

[0005] High cost: Industrial CT equipment is expensive to purchase, maintain, and operate. In addition, radiation safety factors must be considered, which further increases the cost.

[0006] Presence of artifacts and measurement errors: Artifacts due to factors such as beam hardening or motion can distort images and complicate image interpretation. Measurement uncertainties caused by the system, the scanned object or the environment can affect the reliability of the results. These issues affect the quality of industrial CT images as well as the accuracy and traceability of the assessment. Summary of the invention

[0007] In order to overcome the shortcomings of the prior art, the present invention proposes a digital twin system for an industrial CT system and a construction method thereof.

[0008] The technical solution is as follows: A digital twin system for industrial CT system, the digital twin framework includes four parts: physical space module, virtual space module, function and application layer, connection module;

[0009] The physical space module is a three-dimensional domain, including physical entities, physical behaviors, and real data. Physical entities are the basis of the physical space module. Each entity has its own inherent properties, which determine the performance and function of the physical entity in the physical world.

[0010] The virtual space module is a mirror image of the physical space module. Like the physical space module, it is also a three-dimensional domain. It corresponds to the physical entities, physical behaviors and real data in the physical space module. The virtual space module includes digital models, simulated behaviors and virtual data.

[0011] In the function and application layer, function refers to its effect on other parts within the framework, that is, to achieve collaboration and iterative optimization between physical space modules and virtual space modules through connecting modules; application refers to its effect outside the framework, that is, to transform the operational results of the digital twin framework into practical solutions for users;

[0012] The connection module is the link for realizing information exchange between the physical space module, virtual space module and the function and application layer in the digital twin framework; the connection module reflects the interactive relationship between the physical space module, virtual space module and the function and application layer, and is also the channel for data flow, ensuring that data can communicate and synchronize seamlessly between different spaces and levels.

[0013] The present invention also proposes a method for constructing a digital twin system for an industrial CT system, the method comprising the following steps:

[0014] Step 1: Define the physical space module;

[0015] The definition of the physical space module includes all the key components and processes of the industrial CT system, while removing or simplifying non-essential elements to avoid over-complication;

[0016] For industrial CT systems, the physical entities include the testing equipment, testing environment, testing personnel and the object to be tested; the key components include the testing equipment and the object to be tested, and the testing equipment includes the X-ray source, detector and mechanical system;

[0017] For industrial CT systems, physical behaviors include system operation procedures and CT physical imaging processes; system operation procedures include setting up and calibrating scanning equipment, configuring scanning parameters, starting and real-time monitoring of scanning execution, completing scanning and viewing reconstruction results; CT physical imaging processes include the generation of X-rays by the ray source, the attenuation of X-rays through objects to varying degrees, the detector receiving the attenuated X-rays and converting them into digital images, and reconstructing CT images from digital images scanned at multiple angles;

[0018] For industrial CT systems, real data include operating data of system equipment, projection data obtained from scanning the inspected object, and CT data reconstructed from these scans;

[0019] Step 2: Build a virtual space module through static connection;

[0020] The virtual space module is constructed based on the previous definition of the physical space module, and the mapping relationship between the virtual space module and the physical space module is established through static connections. These static connections maintain the fidelity of the virtual representation with its physical counterpart;

[0021] For industrial CT systems, digital models include virtual inspection equipment, virtual environment, virtual operators, and virtual inspected objects. The construction of these digital models includes: geometric shapes and their unique attributes. These attributes are organized into attribute dictionaries, represented in the form of "key-value" pairs, for subsequent simulation and analysis in the virtual space module.

[0022] For industrial CT systems, the simulation behavior includes simulated operation of the industrial CT system, where a virtual device performs a scan on a digital object in a controlled environment; this includes simulated X-ray generation and detection, where virtual X-rays pass through the digital object, mimicking the absorption and interaction patterns observed in the physical space module; the results of these simulations are virtual projection data, which are then processed to reconstruct a virtual CT image;

[0023] For industrial CT systems, virtual data is data generated by the virtual space module, including simulated CT scans and reconstructions, as well as virtual system operation data;

[0024] Step 3: Implement the function and application layer through dynamic connection;

[0025] The function and application layer receives information from the physical and virtual space modules through dynamic connections; the function and application layer comprehensively analyzes, processes and enhances the received information; the results of analysis, processing and enhancement are then fed back to the respective spaces through dynamic connections, which helps to continuously update the physical and virtual space modules; in addition, this layer delivers information to users and connects the digital twin framework with real applications and facilitates user interaction;

[0026] In the process of industrial CT inspection, the function and application layers are sequentially divided into three stages: before inspection, during inspection, and after inspection;

[0027] The pre-detection functional and application layers are used for: simulation analysis, decision making, and data generation;

[0028] In the pre-inspection stage, the digital twin framework is performing simulations to analyze different scanning methods, scanning parameters, reconstruction algorithms, and reconstruction parameters;

[0029] The functions and application layers in the detection are used for: real-time monitoring and fault prediction;

[0030] Throughout the inspection process, the digital twin framework provides real-time monitoring, and the data collected through this process can not only be delivered to the user, but also used by the digital twin framework to perform predictive analysis of potential failures;

[0031] The post-detection functions and application layers are used for: data analysis and equipment maintenance;

[0032] In the post-detection phase, a comprehensive data analysis is performed to combine the empirical data collected from the physical space module with the simulation data generated by the virtual space module.

[0033] The beneficial effects of the present invention are:

[0034] 1. Simplify the operation process and parameter settings of industrial CT. Through simulation analysis, the framework can obtain the best scanning mode and parameters, reduce the dependence on the operator's professional knowledge, and optimize the parameter settings to ensure that the scanning process of different materials or workpieces is highly consistent, thereby improving the imaging efficiency and the repeatability of the results.

[0035] 2. Through simulation and testing of virtual space modules, the need for physical equipment is reduced, thereby reducing purchase and maintenance costs. Predictive maintenance using collected data avoids unexpected downtime and reduces maintenance costs. Through simulation in a virtual environment, operational training and process optimization can be carried out without compromising radiation safety.

[0036] 3. Through the optimization of simulation behavior, the framework can predict and correct factors that may cause artifacts, thereby improving image quality. The data analysis function of the framework can identify and correct measurement uncertainties caused by the system, workpiece or environment, improving the reliability of measurement results. By recording and analyzing all data during operation, the framework enhances the transparency and traceability of the evaluation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Schematic diagram of the digital twin system of the present invention;

[0038] Figure 2 Schematic diagram of the digital twin system for industrial CT system of the present invention;

[0039] Figure 3 This is a flow chart of a method for constructing a digital twin system for an industrial CT system according to the present invention;

[0040] Figure 4 Schematic diagram of the software platform of the digital twin system for industrial CT system of the present invention. DETAILED DESCRIPTION

[0041] The present invention will be further described below with reference to the accompanying drawings and examples.

[0042] like Figure 1 As shown in FIG. , the present invention proposes a digital twin system for an industrial CT system, such as Figure 1As shown, it includes four parts: physical space module, virtual space module, function and application layer, and connection module.

[0043] The physical space module is a three-dimensional domain that includes physical entities, physical behaviors, and real data.

[0044] Physical entity: Physical entity is the basis of physical space module, which can be components, facilities, etc. Each entity has its inherent properties, such as size, shape, material composition, etc. These properties determine the performance and function of the entity in the physical world.

[0045] Physical behavior: Physical entities are not static, they exhibit various behaviors according to the internal physical laws and external conditions. These behaviors include not only the individual actions of the entities, but also the interactions between the entities.

[0046] Real data: Activities and interactions in the physical space module generate a large amount of data. This data is a direct record of the entity's behavior and interaction. Real data provides a basis for understanding and analyzing the behavior of physical entities.

[0047] These three components interact with each other to form an organic set. In this set, physical entities change or interact with other entities according to their inherent properties and behavioral rules. These changes and interactions are not isolated. They affect each other and act together on the physical space module, ultimately producing various phenomena or outputs.

[0048] The virtual space module is a mirror image of the physical space module. Like the physical space module, it is also a three-dimensional field. Corresponding to the physical entities, physical behaviors and real data in the physical space module, the virtual space module includes digital models, simulated behaviors and virtual data.

[0049] Digital Model: Digital models are exact replicas of physical entities in virtual space modules. They include not only the geometry and dimensions of the entity, which can be created and represented by computer-aided design (CAD) software, but also the material properties and other key features of the entity, which can be described in detail by property dictionaries.

[0050] Simulation behavior: Simulation behavior is the behavior of digital models in a virtual environment. They are achieved through mathematical models and simulation algorithms that apply physical laws. Simulation behavior transforms the static characteristics of digital models into dynamic performance, providing power for analysis and optimization.

[0051] Virtual data: Corresponding to the real data in the physical space module, the data generated by the virtual space module is called virtual data. These data are the result of simulating and analyzing the simulation behavior of the digital model. Virtual data is not only used to evaluate the effectiveness of the simulation behavior, but can also be fed back to the digital model for further optimization and adjustment.

[0052] The digital model is the basis of all activities in the virtual space module, and provides necessary information for the generation of simulation behavior and virtual data; the simulation behavior shows the behavior and response of the digital model under different conditions; the virtual data is the output of the digital model through simulation behavior, and also provides feedback and guidance for the evaluation and optimization of the digital model and simulation behavior.

[0053] The function and application layer is the core part of the entire digital twin framework. "Function" refers to its role in other parts of the framework, that is, to achieve efficient collaboration and iterative optimization between physical space modules and virtual space modules through dynamic connections. "Application" refers to its role outside the framework, that is, to transform the operational results of the digital twin framework into practical solutions for users. These two aspects overlap to a large extent, so they are combined into the function and application layer.

[0054] The function and application layer implements the following functions:

[0055] Simulation analysis: Use digital models to simulate under various conditions, predict the behavior and response of physical entities, and provide a scientific basis for decision-making.

[0056] Decision making: Based on the results of simulation analysis, assist or automate the decision-making process to improve the efficiency and accuracy of decision-making.

[0057] Data generation: Through simulation and monitoring activities, data is continuously collected and generated, providing a rich source of information for the digital twin model.

[0058] Real-time monitoring: Track the status of physical entities in real time and respond quickly to any abnormalities or deviations from expected behavior.

[0059] Error prediction: Predict potential failures or problems by analyzing trends and patterns, enabling preventive maintenance.

[0060] Data analysis: Deeply explore and analyze collected data to extract valuable insights and optimize models and processes.

[0061] Equipment maintenance: Plan and execute maintenance activities based on data analysis and prediction results to reduce downtime and extend equipment life.

[0062] The connection modules are the link for information exchange between the physical space module, virtual space module, and function and application layers in the digital twin framework. They not only reflect the interactive relationship between the various parts, but also serve as channels for data flow, ensuring that information can be seamlessly communicated and synchronized between different spaces and levels.

[0063] The connection module includes static connection and dynamic connection.

[0064] Static connections are defined at the beginning of the framework design and remain stable during system operation. The main function of static connections is to establish a one-to-one mapping relationship between physical space modules and virtual space modules, ensuring that digital models, simulated behaviors, and virtual data can accurately reflect their corresponding physical entities, physical behaviors, and real data. Static connections provide the necessary structural foundation and consistency for the digital twin framework.

[0065] Dynamic connections enable interaction between the physical and virtual space modules through the function and application layer, and change as the framework evolves. During operation, the information generated by the physical and virtual space modules is transmitted to the function and application layer through dynamic connections. This layer integrates, processes and analyzes information, provides user-oriented solutions to the outside world, and also provides feedback to the physical and virtual space modules to achieve continuous self-learning and adjustment.

[0066] Static connections ensure the stability and consistency of the framework foundation, while dynamic connections give the framework adaptability and flexibility. This combination enables the digital twin framework to not only provide support during the design and deployment phases, but also continuously optimize operations, improve efficiency and responsiveness throughout the life cycle.

[0067] like Figure 2 As shown in the figure, the present invention proposes a digital twin system for industrial CT system. Figure 1 Based on the digital twin framework of the industry, the various links involved in industrial CT are added. Detailed information will be combined with Figure 3 Give a description.

[0068] like Figure 3 As shown, the construction of the digital twin system for the industrial CT system of the present invention is realized through the following process:

[0069] Step 1: Define the physical space module.

[0070] The definition of the physical space module must comprehensively include all key components and processes of the industrial CT system, while removing or simplifying non-essential elements to avoid over-complication.

[0071] For industrial CT systems, the physical entities include detection equipment, detection environment, detection personnel and objects to be detected. The key components include detection equipment (X-ray source, detector and mechanical system) and objects to be detected.

[0072] For industrial CT systems, physical behaviors include system operation procedures and CT physical imaging processes. The operation procedures include setting up and calibrating scanning equipment, configuring scanning parameters, starting and real-time monitoring of scanning execution, completing scanning and viewing reconstruction results. The CT physical imaging process includes the generation of X-rays by the ray source, the attenuation of X-rays through objects to varying degrees, the detector receiving the attenuated X-rays and converting them into digital images, and reconstructing CT images from digital images scanned at multiple angles.

[0073] For industrial CT systems, real data includes the operating data of system equipment, the projection data obtained from scanning the inspected object, and the CT data reconstructed from these scans. These data not only reflect the current operating status of the system, but also provide a basis for subsequent data analysis and image data reconstruction.

[0074] Step 2: Build the virtual space module through static connection.

[0075] The virtual space modules are built upon the prior definition of the physical space modules, ensuring an accurate and consistent representation. The mapping between these two domains is established through static connections that serve as a robust link to maintain the fidelity of the virtual representation with its physical counterpart.

[0076] For industrial CT systems, digital models include virtual inspection equipment, virtual environments, virtual operators, and virtual inspected objects. The construction of these digital models includes two main aspects: geometry (usually represented by CAD models of physical entities) and their unique properties (directly corresponding to the properties found in physical entities). These properties are organized into attribute dictionaries and represented in the form of "key-value" pairs to facilitate subsequent simulation and analysis in virtual space modules.

[0077] Two approaches were adopted to construct the geometry of the digital model: a drawing-based approach and a measurement-based approach.

[0078] The drawing-based method refers to the use of existing design drawings and technical documents to construct the geometric information of the digital model. Through 3D modeling software, the information in the design drawings and technical documents can be converted into the geometric shape of the digital model.

[0079] The measurement-based method directly obtains geometric data from physical objects, and then uses this data to create geometric information of digital models; for objects that only need external information, a depth camera is used to generate point cloud data, and for objects that require external and internal information, CT scanning can be used to obtain point cloud data. The point cloud data is denoised, registered, and then converted into a triangular mesh model to obtain the geometric shape of the digital model.

[0080] For industrial CT systems, simulation activities include simulated operation of industrial CT systems, where virtual equipment performs scans on digital objects in a controlled environment. This includes simulated X-ray generation and detection, where virtual X-rays pass through digital objects, mimicking the absorption and interaction patterns observed in physical space modules. The result of these simulations is virtual projection data, which is then processed to reconstruct virtual CT images. These images are used for a variety of purposes, including operator training, system testing, and algorithm development, without the cost or risk of physical experiments.

[0081] For industrial CT systems, virtual data refers to data generated by the virtual space module, including simulated CT scans and reconstructions, as well as virtual system operation data. These data are essential for verifying the accuracy of the virtual model and help to refine the digital twin to better simulate its physical counterpart. In addition, these data can provide improved insights into the design and operation of the physical system, facilitating a continuous feedback loop between the virtual and physical space modules.

[0082] Step 3: Implement the function and application layer through dynamic connection.

[0083] The Function and Application layer defines the main uses and expected impacts of the digital twin framework. This layer receives information from the physical and virtual space modules through dynamic connections. It performs comprehensive analysis, processing, and enhancement of the received information. The results of these operations are then fed back to the respective spaces through dynamic connections, which helps to continuously update the physical and virtual space modules. In addition, this layer delivers information to users, effectively connecting the digital twin framework to real-world applications and facilitating user interaction.

[0084] In the industrial CT detection process, the function and application layers can be divided into three stages in sequence: before detection, during detection and after detection.

[0085] Pre-detection functions and applications: simulation analysis, decision making, and data generation.

[0086] In the pre-inspection stage, the digital twin framework plays a key role in conducting simulations to analyze how different scanning methods, scanning parameters, reconstruction algorithms, and reconstruction parameters affect the final results of CT imaging. This comprehensive simulation enables users to select the most effective factors and then develop a perfect inspection plan.

[0087] Functions and applications in detection: real-time monitoring and fault prediction.

[0088] Throughout the inspection process, the digital twin framework provides real-time monitoring, and the data collected through this process can not only be delivered to the user, but also used by the digital twin framework to perform predictive analysis on potential failures.

[0089] Functions and applications after detection: data analysis and equipment maintenance.

[0090] In the post-test phase, a comprehensive data analysis was performed to combine the empirical data collected from the physical space module with the simulated data generated by the virtual space module. This comprehensive analysis facilitated careful comparison and analysis of any discrepancies in the data sets, and equipment maintenance and calibration based on the test results and equipment operation data.

[0091] like Figure 4 The figure shows a schematic diagram of the software platform of the digital twin framework for the industrial CT system of the present invention. The software platform is the carrier of the digital twin framework of the present invention and the bridge for interaction with users. The software platform consists of four modules: algorithm module, visualization module, communication module and data module.

[0092] The algorithm module is responsible for simulation, data processing and data analysis.

[0093] The visualization module uses virtual reality technology to provide real-time 3D display of digital models. In addition to traditional interaction methods such as mouse, keyboard and display, it also supports virtual reality devices, including head-mounted displays, controllers and sensors. Users can rotate, move, scale, click and drag objects in the virtual environment and receive highly realistic visual, auditory and tactile feedback.

[0094] The communication module ensures data transmission and information exchange between various components. The communication within the virtual space module is handled by inter-process communication within the computer. In the physical space module, the computer and the detector are connected through an image acquisition card, which converts the analog signal received by the detector into a digital image and then outputs the digital image to the computer. The communication between the computer X-ray source and the mechanical system is achieved through a network switch.

[0095] The data module is responsible for the storage, management and access of data. The data includes not only data from the physical space module, but also data generated by the virtual space module and historical data. The data module uses efficient data structures and database management systems to support fast data retrieval and processing while ensuring data integrity and security.

[0096] In summary, the digital twin framework for industrial CT systems proposed in the present invention and its construction method, by integrating simulation analysis and parameter optimization technology, effectively reduce the dependence on professional training of operators, while improving the consistency and efficiency of the imaging process. The simulation and predictive maintenance of virtual space modules within the framework reduce the dependence on physical equipment, thereby reducing the overall operating cost. In addition, by comprehensively analyzing virtual and real data, the present invention significantly improves the image quality, and ensures the stable operation of the system and enhances reliability with the help of real-time monitoring and continuous performance evaluation. The application of digital twin technology not only promotes the innovation of industrial CT detection technology, but also provides a new perspective and tools for technological progress and application practice in related fields, and contributes important forces to the sustainable development of the industrial detection field and the realization of intelligent manufacturing.

Claims

1. A digital twin system for industrial CT system, characterized in that: The digital twin system consists of four parts: physical space module, virtual space module, function and application layer, and connection module. The physical space module is a three-dimensional domain, including physical entities, physical behaviors, and real data. Physical entities are the basis of the physical space module. Each entity has its own inherent properties, which determine the performance and function of the physical entity in the physical world. The virtual space module is a mirror image of the physical space module. Like the physical space module, it is also a three-dimensional domain. It corresponds to the physical entities, physical behaviors and real data in the physical space module. The virtual space module includes digital models, simulated behaviors and virtual data. In the function and application layer, function refers to its effect on other parts within the framework, that is, to achieve collaboration and iterative optimization between physical space modules and virtual space modules through connecting modules; application refers to its effect outside the framework, that is, to transform the operational results of the digital twin framework into practical solutions for users; The connection module is the link for information exchange between the physical space module, virtual space module, and function and application layer in the digital twin framework; the connection module reflects the interactive relationship between the physical space module, virtual space module, and function and application layer, and is also the channel for data flow, ensuring that data can communicate and synchronize seamlessly between different spaces and levels; Physical behavior refers to the various behaviors that physical entities exhibit according to internal physical laws and external conditions; these behaviors include not only the individual actions of entities, but also the interactions between entities; Real data refers to data generated by activities and interactions in the physical space module. These data are direct records of entity behaviors and interactions. Real data provides a basis for understanding and analyzing the behavior of physical entities. Physical entities, physical behaviors, and real data form a set in which physical entities change or interact with other physical entities according to their inherent properties and behavioral rules. These changes and interactions affect each other and act on the physical space module together, ultimately producing various phenomena or outputs. Digital models are replicas of physical entities in virtual space modules; they include not only the geometry and dimensions of the physical entity, which are created and represented by computer-aided design software, but also the material properties and other key characteristics of the physical entity, which are described in detail by a property dictionary; Simulation behavior is the behavior of digital models in a virtual environment. They are achieved through mathematical models and simulation algorithms that apply physical laws. Simulation behavior transforms the static characteristics of digital models into dynamic performance, providing power for analysis and optimization. Virtual data corresponds to real data in the physical space module. The data generated by the virtual space module is called virtual data. These data are the results of simulating and analyzing the simulation behavior of the digital model. Virtual data is not only used to evaluate the effectiveness of the simulation behavior, but also fed back to the digital model for further optimization and adjustment. The digital model is the basis of activities in the virtual space module, providing necessary information for the generation of simulation behavior and virtual data; simulation behavior shows the behavior and response of the digital model under different conditions; virtual data is the output of the digital model through simulation behavior, and also provides feedback and guidance for the evaluation and optimization of the digital model and simulation behavior; The function and application layer implements the following functions: Simulation analysis: Use digital models to simulate under various conditions, predict the behavior and response of physical entities, and provide a scientific basis for decision-making; Decision making: Based on the results of simulation analysis, assist or automate the decision-making process to improve the efficiency and accuracy of decision-making; Data generation: Through simulation and monitoring activities, data is continuously collected and generated, providing a rich source of information for the digital twin framework; Real-time monitoring: Track the status of physical entities in real time and respond quickly to any abnormalities or deviations from expected behavior; Error prediction: predict potential failures or problems by analyzing trends and patterns, enabling preventive maintenance; Data analysis: Deeply explore and analyze the collected data, extract valuable insights, and optimize models and processes; Equipment maintenance: Plan and execute maintenance activities based on data analysis and prediction results to reduce downtime and extend equipment life; The connection module includes static connection and dynamic connection; Static connections are defined at the beginning of framework design and remain stable during the operation of the CT system. Static connections are used to establish a one-to-one mapping relationship between physical space modules and virtual space modules, ensuring that digital models, simulation behaviors, and virtual data can accurately reflect their corresponding physical entities, physical behaviors, and real data. Static connections provide a structural foundation and consistency for the digital twin framework. Dynamic connections enable interaction between the physical space module and the virtual space module through the function and application layer, and change as the framework evolves; during operation, the information generated by the physical space module and the virtual space module is transmitted to the function and application layer through dynamic connections; the function and application layer integrates, processes and analyzes information, provides user-oriented solutions to the outside world, and also provides feedback to the physical space module and the virtual space module to achieve continuous self-learning and adjustment.

2. A method for constructing a digital twin system for an industrial CT system, characterized in that: The method comprises the following steps: Step 1: Define the physical space module; The definition of the physical space module includes all the key components and processes of the industrial CT system, while removing or simplifying non-essential elements to avoid over-complication; For industrial CT systems, the physical entities include the testing equipment, testing environment, testing personnel and the object to be tested; the key components include the testing equipment and the object to be tested, and the testing equipment includes the X-ray source, detector and mechanical system; For industrial CT systems, physical behaviors include system operation procedures and CT physical imaging processes; system operation procedures include setting up and calibrating scanning equipment, configuring scanning parameters, starting and real-time monitoring of scanning execution, completing scanning and viewing reconstruction results; CT physical imaging processes include the generation of X-rays by the ray source, the attenuation of X-rays through objects to varying degrees, the detector receiving the attenuated X-rays and converting them into digital images, and reconstructing CT images from digital images scanned at multiple angles; For industrial CT systems, real data include operating data of system equipment, projection data obtained from scanning the inspected object, and CT data reconstructed from these scans; Step 2: Build a virtual space module through static connection; The virtual space module is constructed based on the previous definition of the physical space module, and the mapping relationship between the virtual space module and the physical space module is established through static connections. These static connections maintain the fidelity of the virtual representation with its physical counterpart; For industrial CT systems, digital models include virtual inspection equipment, virtual environment, virtual operators and virtual inspected objects. The construction of these digital models includes: geometric shapes and their unique attributes. These attributes are organized into attribute dictionaries, represented in the form of "key-value" pairs, for subsequent simulation and analysis in the virtual space module. For industrial CT systems, the simulation behavior includes simulated operation of the industrial CT system, where a virtual device performs a scan on a digital object in a controlled environment; this includes simulated X-ray generation and detection, where virtual X-rays pass through the digital object, mimicking the absorption and interaction patterns observed in the physical space module; the results of these simulations are virtual projection data, which are then processed to reconstruct a virtual CT image; For industrial CT systems, virtual data is data generated by the virtual space module, including simulated CT scans and reconstructions, as well as virtual system operation data; Step 3: Implement the function and application layer through dynamic connection; The function and application layer receives information from the physical and virtual space modules through dynamic connections; the function and application layer comprehensively analyzes, processes and enhances the received information; the results of analysis, processing and enhancement are then fed back to the respective spaces through dynamic connections, which helps to continuously update the physical and virtual space modules; in addition, this layer delivers information to users and connects the digital twin framework with real applications and facilitates user interaction; In the process of industrial CT inspection, the function and application layers are sequentially divided into three stages: before inspection, during inspection, and after inspection; The pre-detection functional and application layers are used for: simulation analysis, decision making, and data generation; In the pre-inspection stage, the digital twin framework is performing simulations to analyze different scanning methods, scanning parameters, reconstruction algorithms, and reconstruction parameters; The functions and application layers in the detection are used for: real-time monitoring and fault prediction; Throughout the inspection process, the digital twin framework provides real-time monitoring, and the data collected through this process can not only be delivered to the user, but also used by the digital twin framework to perform predictive analysis of potential failures; The post-detection functions and application layers are used for: data analysis and equipment maintenance; In the post-detection phase, a comprehensive data analysis is performed to combine the empirical data collected from the physical space module with the simulation data generated by the virtual space module.

Images