Multi-scene intelligent manufacturing teaching practical training method and system based on digital twinning technology, storage medium and electronic equipment

Abstract

The invention provides a multi-scene intelligent manufacturing teaching practical training method and system based on a digital twinning technology, a storage medium and electronic equipment. The method comprises the steps of data acquisition and processing, intelligent manufacturing digital twinning model construction, multi-scene integration, practical training process development, resource issuing and sharing, teaching management and the like. A digital twinborn model of equipment, process and environment is constructed and integrated into a three-dimensional virtual scene, and a graphical process design tool and a cloud platform are combined, so that standardized development, online deployment and intelligent evaluation of teaching resources are realized. The method improves the flexibility and efficiency of practical training, and is suitable for the construction and management of intelligent manufacturing multi-type teaching practical training scenes.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing education technology, and in particular to a multi-scenario intelligent manufacturing teaching and training method, system storage medium and electronic equipment based on digital twin technology. Background Art

[0002] Currently, intelligent manufacturing is the product of the deep integration of new-generation information technology and advanced manufacturing technology, and it runs through all aspects of manufacturing activities, including design, production, management, and service. Strengthening the training of intelligent manufacturing professionals and improving their capabilities are urgent tasks for the high-quality development of intelligent manufacturing.

[0003] Experimental teaching is a crucial component of cultivating talent for intelligent manufacturing. Establishing an intelligent manufacturing training base is the most common method of experimental teaching. Intelligent manufacturing training bases are purpose-built practical teaching venues designed to cultivate skilled personnel in intelligent manufacturing, providing students with a realistic production environment and hands-on opportunities. Through practical training at these training bases, students can learn about intelligent manufacturing process flows, technical principles, equipment operation, production management, and other aspects, improving their practical skills and employability. However, the smooth implementation of intelligent manufacturing experimental teaching is hindered by issues such as the large footprint, large number of units, high cost, and difficulty ensuring student safety.

[0004] Furthermore, training bases face issues such as outdated equipment, backward management concepts and systems, a relative lack of practical opportunities, a disconnect between talent development and actual industry needs, and insufficient integration of industry and education. The leading role of enterprises in developing manufacturing talent has yet to be fully realized, hindering the pace of development. To ensure that the effectiveness of intelligent manufacturing training matches the needs of enterprises, it is necessary to develop more comprehensive training resources, integrate them with modern educational concepts and enterprise production practices, and develop a more digitalized intelligent manufacturing training model.

[0005] With the development of new-generation information technologies such as virtual reality, the metaverse, and digital twins, digital modeling and simulation are enabling the rapid restoration of the production system state of intelligent manufacturing training bases on computers, as well as the simulation and visualization of the system's dynamic behavior. By combining virtual simulation teaching software with actual intelligent manufacturing production lines, an intelligent manufacturing virtual simulation teaching and training platform has been constructed. This platform provides richer teaching resources and more convenient training conditions. Through a "virtual-real" teaching model, it aims to cultivate interdisciplinary technical personnel with expertise in intelligent equipment design and operation and maintenance, intelligent factory management, and system integration and control. However, current intelligent manufacturing virtual teaching and training platforms suffer from a single production scenario, poor virtual-real interaction, and an inability to fully assess student capabilities. This results in students being unable to obtain effective feedback from the real world and having difficulty accurately understanding the theoretical basis behind each training operation. Summary of the Invention

[0006] The purpose of this invention is to provide a multi-scenario intelligent manufacturing teaching and training method, system storage medium and electronic equipment based on digital twin technology, which can realize the virtualization, scenario and interactivity of intelligent manufacturing teaching and training, provide learners with a rich and realistic learning experience, and at the same time reduce the cost of training and improve teaching efficiency and effectiveness.

[0007] The technical solution adopted in the present invention is: The multi-scenario intelligent manufacturing teaching and training method based on digital twin technology includes the following steps: Step a, Data Collection and Processing: Based on the training course requirements, determine the types of smart manufacturing equipment and physical quantities that need to be collected, including equipment appearance and structural parameters, operating status parameters, and environmental parameters; select and install temperature, pressure, vibration, and visual sensors to collect the above physical quantities, and pre-process the data using data cleaning, format conversion, time alignment, signal fusion, and normalization to construct a high-quality data set as the data foundation for the construction of the smart manufacturing digital twin model; Step b: Constructing a smart manufacturing digital twin model: First, based on the data set obtained in step a, combined with CAD two-dimensional drawings, PLC control logic diagrams, and process flow chart engineering data, clarify the structural information and parameter relationships of the equipment, process, and environment. Use 3D modeling software to build a geometric model, and add the model's structural attributes, motion constraints, communication interface information, and operating logic to realize the construction of the equipment model, process model, and environment model, thereby forming a smart manufacturing digital twin model that reflects the actual system behavior. Construct a smart manufacturing digital twin model to accurately reflect the structure and operating status of the physical system of the smart manufacturing equipment. Step c, intelligent manufacturing multi-scenario integration: Based on the process flow template and logic control rule library constructed by production process specifications, equipment control logic and historical operation data, the intelligent manufacturing digital twin model constructed in step b is combined to quickly and conveniently build multiple intelligent manufacturing production scenarios. Different scenarios can be freely switched and combined, providing a three-dimensional virtual scene for subsequent experimental training processes, allowing learners to experience the workflow and operation requirements in different intelligent manufacturing environments; Step d. Custom development of experimental training processes: Develop an experimental training process construction tool, import the 3D virtual scene built in step c, and customize the experimental training process based on different experimental training content. Select templates from the school-enterprise cooperation template library (which stores enterprise-certified process templates, such as standard automotive welding processes) or the AI-generated template pool (based on personalized processes recommended by the DQN algorithm). Configure interactive controls (such as the same HMI interface components as the enterprise production line) and scoring nodes (such as the capacity compliance rate indicator that connects to the enterprise KPI). The AI-generated template pool supports automatic process generation through natural language input (such as generating a battery assembly fault handling process), and verifies logical feasibility in real time through a digital twin model (verification takes less than 5 minutes). It also supports style adjustment and teaching element layout, allowing for rapid construction of teaching and training processes and the rapid development of various experimental training processes. Step e: Publishing and sharing experimental training resources: Integrate the various intelligent manufacturing 3D virtual scenes independently built in step c with the experimental training processes custom developed in step d to form virtual simulation resources containing teaching experimental training processes; publish the virtual simulation resources to the resource market of the cloud platform, and transform the developer's resource results in the resource market. For resources that do not involve commercial secrets or intellectual property disputes, implement an open access policy, allowing users to use and share them online for free or at low cost, breaking geographical restrictions, realizing "co-construction and sharing" of resources, and improving the utilization rate of virtual experimental training resources; Step f, experimental training teaching management: set up automatic assessment and learning tracking functions for each link of experimental training, use big data analysis, machine learning, knowledge graph construction, behavior trajectory mining and other technical means to conduct intelligent statistical analysis of learners' assessment data, support automatic scoring by step, automatic summary of experimental training, homework, and assessment results, generate analysis reports, and perform visual presentation; trace and track each training link of learners, and automatically form a technical skill portrait of students based on objective data such as learning time, training process and mastery status, and feed back the results of learners' training in each link to the teacher. Teachers can answer questions or consolidate explanations in a targeted manner in class based on the feedback results.

[0008] The step b specifically comprises the following steps: First, based on the BIM parametric template library, the 3D virtual scene model skeleton is generated by combining CAD 2D drawings and collected data analysis. Second, the Newton-Euler method and other kinematic inverse solution algorithms are used to automatically configure the degrees of freedom of the equipment joints, supporting a six-axis robotic arm trajectory simulation error of less than 0.1mm. The integrated ANSYS finite element analysis engine simulates the thermal deformation and stress distribution of the equipment in real time. Then, the parameters of the intelligent manufacturing digital twin model were dynamically corrected through the LSTM neural network, increasing the simulation accuracy to 97.3%, completing the three-dimensional geometric modeling of the intelligent manufacturing production equipment and assembly line; Subsequently, by docking with the data interface of the acquisition system in step a, the real-time physical data is connected and integrated into the corresponding digital twin model to realize data synchronization between the physical world and the virtual world; finally, the physical process of the simulation equipment is simulated using the platform's simulation software, and the intelligent manufacturing digital twin model is presented to the user in a graphical manner through the three-dimensional visualization platform. The user operates the construction of the intelligent manufacturing digital twin model through the human-computer interaction interface, and can use PLC visual programming to observe and predict the behavior status of the intelligent manufacturing equipment in a visual manner, and then optimize the intelligent manufacturing digital twin model to further improve its accuracy and practicality; repeating this step can obtain multiple intelligent manufacturing digital twin models to form an intelligent manufacturing digital twin model library.

[0009] The step c includes the following steps: based on the intelligent manufacturing digital twin model library constructed in step b, in accordance with the process flow template and logic control rule library constructed by production process specifications, equipment control logic and historical operation data, the configuration content defined is encapsulated by the process modeling language, and with the support of the digital twin cloud platform, the twin models in the intelligent manufacturing digital twin model library are combined and connected by dragging and dropping, and a semantic conversion tool is developed that supports 12 mainstream industrial communication protocols including Modbus-TCP, Profinet, EtherCAT, OPC UA, CANopen, EtherNet / IP, SERCOS, BACnet, HART, DeviceNet, IEC61850, and Profibus, and the data is encapsulated in JSON-LD format to convert data between different devices and different protocols into a unified format; Based on the genetic algorithm, the layout of production line equipment is optimized for multiple objectives, namely minimizing logistics costs and maximizing production capacity. It also supports drag-and-drop adjustments, and finally completes the layout of the site and equipment to build various intelligent manufacturing production scenarios.

[0010] The development experiment and training process construction tool in step d adopts a front-end and back-end separation architecture. The front-end is based on the Vue.js framework to design a graphical user interface, and the back-end is based on Node.js to implement process analysis and operation control, supporting visual configuration and logical arrangement of various elements in the experiment and training process; The development experiment and training process construction tool has a scene import function, which loads the three-dimensional virtual scene constructed in step c into the process editing interface as a teaching operation environment; it also has a page creation function, which quickly generates a training interaction page through preset teaching templates or custom components; supports adding various UI controls, and provides property setting and style adjustment functions; has a built-in flowchart-style DSL language and atomic instruction blocks of various standardized experiment and training processes; users build control processes by dragging and dropping combinations to realize data interaction and function calls between modules; select or define a suitable business process framework according to different experiment and training contents, integrate the scene and template, and adjust the style to realize customized development of the experiment and training process.

[0011] The multi-scenario intelligent manufacturing teaching and training system based on digital twin technology includes the following modules: Data acquisition and processing module, used to collect and process physical quantity data of intelligent manufacturing equipment, processes, and environment; The intelligent manufacturing digital twin model construction module is used to perform geometric modeling and rendering based on the collected and processed physical quantity data of intelligent manufacturing equipment, processes, and environments. It also adds structural information attributes, motion parameters, control logic, communication data, behavioral rules, and operating status tags to the twin model to build an intelligent manufacturing digital twin model with dynamic response capabilities. The intelligent manufacturing multi-scenario integration module is used to quickly and conveniently build a variety of intelligent manufacturing production scenarios based on process flow templates and logic control rule libraries constructed from production process specifications, equipment control logic, and historical operation data. It encapsulates the defined configuration content through a process modeling language and combines them on the basis of the constructed intelligent manufacturing digital twin model. This allows for free switching and combination between different scenarios. The experimental training process custom development module is used to import the constructed 3D virtual scene, customize the experimental training process according to different experimental training contents, select the appropriate teaching resource template, integrate the scene and template, and realize the rapid development of various experimental training processes; The experimental training resource publishing and sharing module is used to integrate various independently built intelligent manufacturing digital twin application scenarios with custom-developed experimental training processes, forming virtual simulation resources containing teaching experimental training processes and publishing them to the resource market of the cloud platform, realizing "co-construction and sharing" of resources and improving the utilization rate of virtual experimental training resources; The experimental training teaching management module is used to conduct intelligent statistical analysis and tracking of the assessment data of each link of the learners, automatically form a technical skill portrait of the students, and feedback the results of the learners' training in each link to the teacher.

[0012] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the device where the computer-readable storage medium is located executes the multi-scenario intelligent manufacturing teaching and training method based on digital twin technology.

[0013] An electronic device includes: a memory and a processor, wherein the memory stores a program that can be run on the processor, and when the processor executes the program, it implements the multi-scenario intelligent manufacturing teaching and training method based on digital twin technology.

[0014] This invention constructs digital twin models of multiple intelligent manufacturing scenarios to form an intelligent manufacturing digital twin model library. Through a training process construction tool, it provides a variety of standardized experimental training process modules, allowing teachers and students to quickly and conveniently build a variety of intelligent manufacturing three-dimensional virtual scenarios. By developing custom construction tools for experimental training processes, it provides a variety of standardized experimental training processes, and provides experimental training effect statistics, tracking, and feedback functions, providing learners with a comprehensive and immersive teaching and training experience. This invention not only reduces training costs and safety risks, but also improves teaching efficiency and effectiveness, providing strong support for the cultivation of intelligent manufacturing talents. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 is a flow chart of the present invention; Figure 2 1 is an electrical schematic diagram of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0018] like Figure 1 and 2As shown, the present invention includes digital twin models of multiple intelligent manufacturing scenarios, forming an intelligent manufacturing digital twin model library, supporting teachers and students to quickly and conveniently build a variety of intelligent manufacturing three-dimensional virtual scenes, and by developing custom construction tools for experimental (training) processes, providing a variety of standardized experimental (training) processes, thereby improving the teaching efficiency and effectiveness of experiments (training).

[0019] Through the above-mentioned method, the present invention can provide a variety of standardized experimental (training) processes, providing learners with a full-dimensional and immersive teaching and training experience, which not only reduces the training cost and safety risks, but also improves the teaching efficiency and effect, and provides strong support for the cultivation of intelligent manufacturing talents.

[0020] See also Figure 1 The present invention provides a multi-scenario intelligent manufacturing teaching and training method based on digital twin technology, comprising the following steps: a. Data collection and processing: Based on specific application requirements, determine the intelligent manufacturing equipment and its physical quantities that need to be collected, select the corresponding data collection equipment and install and debug it, collect the physical quantities of relevant equipment, and perform data preprocessing such as cleaning, conversion, fusion, and normalization to provide a data source for the subsequent construction of the intelligent manufacturing digital twin model.

[0021] During implementation, appropriate sensors (temperature, pressure, vibration, etc.) were selected based on the specific teaching and training course objectives and the specific production equipment and production line conditions to monitor the equipment's operating status and physical parameters in real time. Sensors were installed and debugged to ensure comprehensive and accurate data collection of equipment physical parameters. Redundant sensor nodes were deployed, using a star-mesh hybrid network that supports 5G / Wi-Fi 6 dual-mode communication. A single node had a coverage radius of 50 meters and a data packet loss rate of less than 0.5%. The OPC UA protocol was used to enable real-time device data collection. A wavelet-based noise filtering algorithm was designed to perform frequency domain denoising on vibration signals. Multi-sensor data was fused using a Kalman filter to ensure a sampling error of ≤1%, providing a data source for the subsequent construction of the intelligent manufacturing digital twin model. This step ensured that the collected data was consistent with the real physical world, achieving data synchronization and virtual-real linkage.

[0022] b. Construct a digital twin model for intelligent manufacturing: Based on the data collected and processed in step a, combine the two-dimensional drawings of the equipment, process, environment, etc. and other relevant information to determine the structure and specific information of the equipment, process, and environment. Use three-dimensional modeling technology or software tools (such as Blender, SolidWorks) to construct a three-dimensional geometric model of the equipment, process, and environment and render it. Add relevant information attributes, motion attributes, and communication data of the model to construct a digital twin model for intelligent manufacturing, specifically including equipment models, process models, and environment models, to accurately reflect the structure and operating status of the physical system of intelligent manufacturing equipment.

[0023] During the specific implementation, first, based on the BIM parametric template library, combined with CAD two-dimensional drawings and collected data analysis, a three-dimensional model skeleton is generated. The kinematic inverse solution algorithms such as the Newton-Euler method are used to automatically configure the device joint degrees of freedom, supporting the six-axis robotic arm trajectory simulation error of less than 0.1mm. The ANSYS finite element analysis engine is integrated to simulate the thermal deformation and stress distribution of the equipment in real time. The model parameters are dynamically corrected through the LSTM neural network, so that the simulation accuracy is improved to 97.3%, completing the three-dimensional geometric modeling of the intelligent manufacturing production equipment and assembly line; then, through the data interface, the real-time data collected in step a is integrated into the digital twin model to realize data synchronization between the physical world and the virtual world; finally, the platform's simulation software is used to simulate the physical process of the simulation equipment, and the digital twin model is presented to the user in a graphical manner through the three-dimensional visualization platform. The user operates the digital twin model through the human-computer interaction interface and can use PLC visual programming to visually observe and predict the behavior status of the intelligent manufacturing equipment, thereby optimizing the digital twin model to further improve its accuracy and practicality. By repeating this step, multiple intelligent manufacturing digital twin models can be obtained to form an intelligent manufacturing digital twin model library.

[0024] c. Intelligent manufacturing multi-scenario integration: Based on the process templates and rule base constructed by production process specifications and equipment control logic, they are combined on the basis of the intelligent manufacturing digital twin model constructed in step b to quickly and conveniently build a variety of intelligent manufacturing production scenarios. Different scenarios can be freely switched and combined, providing three-dimensional application scenarios for subsequent experimental (practical training) processes, allowing learners to experience the workflow and operational requirements in different intelligent manufacturing environments.

[0025] In specific implementation, based on the smart manufacturing digital twin model library constructed in step b, according to logical processing algorithms and process flow definitions, real-time data synchronization between the physical workshop and the virtual scene is achieved through the WebSocket protocol. Supported by the digital twin cloud platform, models in the smart manufacturing digital twin model library can be combined and connected through a drag-and-drop method. A semantic conversion tool has been developed that supports 12 mainstream industrial communication protocols, including Modbus-TCP, Profinet, EtherCAT, OPC UA, CANopen, EtherNet / IP, SERCOS, BACnet, HART, DeviceNet, IEC 61850, and Profibus. Data is encapsulated in the JSON-LD format, converting data between different devices and protocols into a unified format. A genetic algorithm is used to optimize the production line equipment layout for multiple objectives (minimizing logistics costs and maximizing production capacity), supporting drag-and-drop adjustments. Ultimately, the layout of the site and equipment is completed, creating various smart manufacturing production scenarios.

[0026] d. Customized Experimental (Practical Training) Process Development: Develop an experimental (practical training) process construction tool with drag-and-drop scene editing capabilities. Import the 3D virtual scene created in step c) and customize the experimental (practical training) process based on the specific experimental (practical training) content. Users can use the "School-Enterprise Cooperation Template Interface" to import real-world production line process files (such as a new energy battery assembly template for an automotive company) or utilize the "AI Template Generation Engine" to automatically generate personalized processes based on historical operational data (e.g., extracting high-frequency operation paths through a hidden Markov model). After importing the 3D virtual scene created in step c, integrate it with the template through parametric configuration (such as defining device model variables and mapping sensor data interfaces) to achieve rapid process development.

[0027] In specific implementation, this embodiment develops an experiment (training) process construction tool with a drag-and-drop scene editing function, which supports adding, deleting or modifying elements such as device models, sensors, control nodes in the scene by dragging and dropping. Customize elements such as scene import, page creation, button addition, experiment (training) step design, panel property setting, and text box content input. Design a flowchart DSL with built-in atomic instruction blocks of various standardized experiment (training) processes. Generate complex control logic by dragging and dropping combinations, increase code generation efficiency by 80%, realize data interaction and function call between modules, provide the function of custom development of experiment (training) processes, import the three-dimensional virtual scene built in step c, select or define the appropriate business process framework according to different experiment (training) contents, integrate the scene with the template, and adjust the style to realize custom development of experiment (training) processes.

[0028] e. Publishing and Sharing Experimental (Practical Training) Resources: Integrate the various intelligent manufacturing digital twin application scenarios independently constructed in step c with the custom experimental (practical training) processes developed in step d to create virtual simulation resources that encompass these teaching experimental (practical training) processes. These virtual simulation resources will be published to the cloud platform's resource market, where developers' resource achievements will be transformed. For resources not involving commercial secrets or intellectual property disputes, an open access policy will be implemented, allowing users to use and share them online for free or at low cost. This will break geographical restrictions, enable resource co-construction and sharing, and increase the utilization rate of virtual experimental (practical training) resources.

[0029] f. Experimental (Practical Training) Teaching Management: Automated assessment and learning tracking functions are implemented for each step of the experimental (practical training) process. Using technologies such as big data analysis, intelligent statistical analysis of learners' assessment data is performed. Automatic scoring by step is supported, and experimental (practical training) results, homework, and assessment results are automatically summarized, generating analytical reports and presenting them visually. A hidden Markov model is used to analyze student operation sequences, tracing and tracking each step of the learner's practical training. Based on objective data such as learning duration, training process, and mastery, a profile of the learner's technical skills is automatically generated. The results of each step of the learner's practical training are then fed back to the teacher. Based on this feedback, the teacher can provide targeted answers to questions or provide consolidation in the classroom. The DQN algorithm is used to recommend personalized learning paths, shortening the skill achievement cycle by 20%.

[0030] In specific implementation, the method proposed in the technical solution of the present invention can be automatically run by those skilled in the art using computer software technology. System devices that implement the method, such as computer-readable storage media that store the corresponding computer program of the technical solution of the present invention and computer equipment that runs the corresponding computer program, should also be within the scope of protection of the present invention.

[0031] This invention constructs digital twin models of multiple intelligent manufacturing scenarios to form an intelligent manufacturing digital twin model library. Through a training process construction tool, it provides a variety of standardized experimental training process modules, allowing teachers and students to quickly and conveniently build a variety of intelligent manufacturing three-dimensional virtual scenarios. By developing custom construction tools for experimental training processes, it provides a variety of standardized experimental training processes, and provides experimental training effect statistics, tracking, and feedback functions, providing learners with a comprehensive and immersive teaching and training experience. This invention not only reduces training costs and safety risks, but also improves teaching efficiency and effectiveness, providing strong support for the cultivation of intelligent manufacturing talents.

[0032] In some possible embodiments, a multi-scenario intelligent manufacturing teaching and training system based on digital twin technology is provided, including the following modules: The first module, the data acquisition and processing module, is used to collect and process physical quantity data of intelligent manufacturing equipment, processes, and environments; The second module, the intelligent manufacturing digital twin model construction module, is used to perform geometric modeling and rendering based on the collected and processed physical quantity data of intelligent manufacturing equipment, processes, and environments, and add relevant information attributes, motion attributes, and communication data of the model to build an intelligent manufacturing digital twin model; The third module, the multi-scenario integration module, is used to combine process templates and rule bases built according to production process specifications and equipment control logic on the basis of the intelligent manufacturing digital twin model built in step b, so as to quickly and conveniently build multiple intelligent manufacturing production scenarios and realize free switching and combination between different scenarios; The fourth module is the custom development module for experimental (training) processes. It is used to import the constructed 3D virtual scenes, customize the experimental (training) processes according to different experimental (training) contents, select appropriate teaching resource templates, integrate the scenes and templates, and realize the rapid development of various experimental (training) processes. The fifth module, the publishing and sharing module for experimental (practical training) resources, is used to integrate various independently built intelligent manufacturing digital twin application scenarios with custom-developed experimental (practical training) processes to form virtual simulation resources containing teaching experimental (practical training) processes and publish them to the resource market of the cloud platform, realizing the "co-construction and sharing" of resources and improving the utilization rate of virtual experimental (practical training) resources; The sixth module, the experimental (practical training) teaching management module, is used to conduct intelligent statistical analysis and tracking of the assessment data of each link of the learners, automatically form a technical skill portrait of the students, and feedback the results of the learners' practical training in each link to the teacher.

[0033] A computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the device containing the computer-readable storage medium executes the multi-scenario intelligent manufacturing teaching and training method based on digital twin technology as described above. The computer program includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory, and other memory devices.

[0034] An electronic device includes: a memory and a processor, wherein the memory stores a program that can be run on the processor, and when the processor executes the program, it implements the multi-scenario intelligent manufacturing teaching and training method based on digital twin technology as described above.

[0035] If the modules / units integrated in the electronic device described in this application are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the present application can implement all or part of the processes in the above-mentioned embodiment methods by instructing the relevant hardware devices to complete them through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above-mentioned various method embodiments.

[0036] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.

[0037] The computer-readable storage medium stores computer-readable instructions, which are processed by the electronic device. The following is a further explanation with specific examples; Example 1: Virtual Automobile Assembly Line Training Scenario description: This system simulates the entire intelligent assembly process of a complete vehicle, covering process links such as body welding, chassis assembly, interior installation, and vehicle inspection. It allows users to freely build equipment layouts by dragging and dropping, and compare production efficiency and troubleshooting processes under different parameter configurations.

[0038] Step a: Data Collection and Processing: Laser ranging sensors (accuracy ±2mm) were deployed on the conveyor belt (model: TGW-AGV-100) to collect real-time workpiece transmission speed and position data. A current sensor (accuracy ±0.5A) was installed on the welding robot (model: KUKA KR 1000) to monitor welding current and voltage parameters. Device data was collected using the OPC UA protocol (frequency: 500Hz). Vibration signal spectra were analyzed using Fourier transforms, and multi-source data was fused using a particle filter algorithm to ensure assembly position error ≤0.2mm and welding parameter error ≤1%.

[0039] Step b: Build a digital twin model for intelligent manufacturing: Based on the CATIA parametric template library and CAD drawings, a 3D model of the body welding fixture, chassis assembly platform, and other components was constructed. The Denavit-Hartenberg (DH) parameter method was used to configure the robot's joint degrees of freedom, achieving a trajectory simulation error of less than 0.15mm. The COMSOL multiphysics simulation engine was integrated to simulate welding thermal deformation and stress distribution. A BP neural network was used to dynamically correct weld strength parameters, improving simulation accuracy to 98.2%. Data synchronization between the virtual model and physical equipment was achieved via the MQTT protocol (latency less than 30ms). The Unity 3D platform visualized the assembly process, supporting real-time adjustment of conveyor speed (0.5-5m / s) and robot trajectory.

[0040] Step c: Smart Manufacturing Multi-Scenario Integration: Models of robots, conveyors, welding machines, and other equipment can be accessed from the "virtual equipment library" via drag-and-drop on the digital twin cloud platform. Magnetic interfaces automatically connect device logic (e.g., intelligently aligning the conveyor end with the robot's gripping station). A semantic conversion tool supporting 12 industrial communication protocols has been developed, with a unified data format of JSON-LD. A genetic algorithm is used to optimize equipment layout for multiple objectives (minimizing logistics costs and maximizing space utilization), automatically generating solutions such as "traditional layout" and "U-shaped layout." A comparison shows that the U-shaped layout shortens logistics paths by 25% and increases production capacity by 20%. Different configurations, such as "Scheme A" and "Scheme B," can be saved, allowing for one-click switching between scenarios and real-time monitoring of production data.

[0041] Step d, Custom Development of the Experimental (Practical Training) Process: Develop a visual editor based on Vue.js / Node.js. After importing a 3D scene, add "Manual Teach" and "Automatic Assembly" buttons. Set a scoring node for "Assembly Cycle Optimization" (weighted 40%). The system automatically calculates output per unit time and recommends parameter adjustments. Develop a "Fault Injection" control that can trigger preset fault scenarios such as "conveyor belt jam" and "welding parameter anomalies." Generate fault handling logic by dragging and dropping standardized instruction blocks (such as "robot speed adjustment" and "weld point parameter reset"), improving code generation efficiency by 85%.

[0042] Step e. Publishing and Sharing Experimental (Practical Training) Resources: Package these resources into a web application and publish them to a cloud platform. Accessible from PCs and VR devices (such as the HTC VIVE), they are differentiated into "teaching mode" (which locks basic configurations) and "research mode" (which opens up full parameter editing). Free trials are available for non-confidential scenarios, and users can download scenario configuration files (.dtconfig format) to enable cross-team reuse and version management.

[0043] Step f: Experimental (practical) teaching management: Using spatiotemporal trajectory visualization technology to replicate student operation sequences, the rationality of assembly logic was analyzed using a hidden Markov model. A "Robot Kinematics" learning package was implemented to address issues such as "path planning errors." The DQN algorithm was used to recommend personalized optimization paths, reducing the production capacity target cycle time by 22%. A multi-scheme comparison report (including production capacity, energy consumption, and failure rate) was generated, and equipment utilization was displayed as a heat map to assist teachers in explaining targeted layout optimization strategies.

[0044] Example 2: Industrial Robot Virtual Disassembly and Programming Training Scenario Description: Targeting the core course "Industrial Robot Technology" for university mechatronics majors, this course simulates the entire process of mechanical structure disassembly and assembly, servo system debugging, and PLC programming for a six-axis industrial robot (such as the ABB IRB 120). This allows users to complete robot disassembly, joint axis calibration, and trajectory programming experiments in a virtual environment, avoiding the risks and damage of real equipment disassembly and assembly.

[0045] Step a, Data Collection and Processing: Virtual torque sensors (accuracy of ±0.1 N·m) were deployed at the virtual robot joints to collect real-time force data on each axis during assembly and disassembly. Current / voltage sensors were embedded in the servo motor model to monitor motor operating parameters. Data was collected using the OPC UA protocol (200 Hz frequency). Multi-joint force data was fused using a Kalman filter to ensure torque simulation error of ≤1%. Wavelet transforms were used to analyze the motor current spectrum and identify abnormal load conditions.

[0046] Step b: Build a digital twin model for intelligent manufacturing: A 1:1 parametric model of the robot, control cabinet, and teach pendant was constructed using SolidWorks. Joint degrees of freedom were configured using the DH parameter method, achieving a kinematic simulation error of less than 0.5mm. The ANSYS finite element engine was integrated to simulate the stress distribution in the gearbox. "Magnetic" assembly and disassembly interaction logic was developed using Unity 3D. When a virtual tool approaches a bolt, the attachment effect is triggered. A "force threshold" interactive feedback mechanism (e.g., excessive torque triggers a red warning) was set to enhance operational realism.

[0047] Step c: Integrate multiple scenarios for intelligent manufacturing: Combine modules such as the "disassembly and assembly workbench," "parts storage area," and "PLC control cabinet" within the digital twin platform. Develop a virtual I / O interface supporting the Modbus-RTU protocol to enable simulation linking robot I / O signals with PLC programs. Three scenarios are divided: "structural recognition," "disassembly and assembly training," and "programming practice." The structural recognition scenario highlights key components (such as the harmonic reducer), while the disassembly and assembly training scenario sets standard disassembly and assembly sequence constraints (e.g., the wrist cover must be removed before the motor).

[0048] Step d: Customize the experimental training process: Develop a visual editor based on Vue.js. After importing a 3D scene, add a "disassembly and assembly step guide" control. The system generates a standardized disassembly and assembly process in the order of "base → upper arm → lower arm → wrist." Teachers can also add a custom scoring node for "safety operation check" (weighted 30%). An integrated PLC visual programming module (similar to the S7-1200 programming interface) allows users to start and stop programs on the robot and view I / O signal status in real time. The system automatically detects program syntax errors and logic vulnerabilities.

[0049] Step e. Publishing and Sharing Experimental Training Resources: These resources are published as PC executables, supporting offline use (ideal for computer lab teaching). They offer both "teacher-side" and "student-side" modes, with the teacher side unlocking the "fault settings" feature (e.g., virtual gearbox stalling). These resources are released alongside the course materials, with each disassembly and assembly step corresponding to a textbook chapter's QR code. Scanning the code allows users to view principle animations (e.g., harmonic reducer transmission principle).

[0050] Step f: Experimental Training Management: A hidden Markov model analyzes user disassembly and assembly sequences, automatically identifying "illegal operations" (such as disassembly without powering off), deducting points, and generating a "Disassembly and Assembly Process Compliance Report." To address issues such as "large joint calibration errors," instructional videos are provided, and the DQN algorithm is used to recommend optimal calibration parameter combinations, reducing calibration time by 40%.

[0051] Example 3: Modular cognitive training of intelligent production lines Scenario Description: Targeting the "Intelligent Manufacturing System Integration" course design, this project builds a modular production line consisting of virtual processing cells, logistics conveyor lines, and visual inspection stations. This allows users to reorganize the production line layout by dragging and dropping, validating concepts such as "flexible manufacturing" and "agile switching." This system is suitable for team collaboration to complete system integration projects.

[0052] Step a: Data Collection and Processing: Virtual vibration sensors (accuracy of ±0.5μm) were deployed in the virtual machining center to collect tool wear data. Photoelectric sensors were installed on the conveyor line to monitor workpiece position. Data was collected using the OPC UA protocol. The vibration signal spectrum was analyzed using Fourier transform to identify abnormal tool wear characteristics. Kalman filtering was then used to fuse multi-sensor position data to ensure workpiece positioning error of ≤1mm.

[0053] Step b: Build a digital twin model for intelligent manufacturing: Based on the BIM parametric template library, standard units such as the "processing module," "inspection module," and "logistics module" are constructed. Each module encapsulates independent I / O interfaces and control logic (for example, the processing module supports the replacement of different tool models). A "status bar" is designed for each module in Unity 3D to display real-time data such as processing progress and equipment energy consumption. The module's internal structure can be viewed by clicking on it (for example, a diagram of the spindle components of a machining center).

[0054] Step c: Smart Manufacturing Multi-Scenario Integration: Modules are combined by dragging and dropping on the digital twin cloud platform. A module interface supporting the Profinet protocol is developed to enable communication between equipment models from different manufacturers (e.g., data exchange between a Siemens PLC and a Fanuc robot). Typical layouts, such as "single-row," "U-shaped," and "island" layouts, are built in. Using a genetic algorithm, the system automatically calculates the logistics efficiency of each layout and generates comparative reports (e.g., a U-shaped layout reduces handling distance by 30% compared to a single-row layout).

[0055] Step d: Customize the development of the experimental training process: Users drag components such as "processing module" and "conveyor line" onto the canvas, and the system automatically connects the logic, completing the production line layout without coding. In the scheduling algorithm panel, select a preset strategy such as "Shortest Path" from the drop-down menu. Adjust the parameters and the system will automatically run. Click to check faults such as "Tool Wear" and drag modules such as "Tool Replacement" to handle them. The system will automatically score them. Mark relationships such as "Inspection Completed → Material Transfer" and the system will automatically generate code and verify the validity of the logic.

[0056] Step e. Publishing and Sharing Experimental Training Resources: These resources are published as WebGL applications, supporting multi-person online collaborative design (similar to Minecraft's building mode). Teachers can view each group's production line design progress in real time and provide comments and suggestions. A "real enterprise production line import" interface is provided, allowing CAD drawings of an automotive welding line to be converted into virtual modules, bringing enterprise projects to campus.

[0057] Step f: Experimental Training Management: Using spatiotemporal trajectory visualization technology, the debugging process of each production line group is recorded. Based on the knowledge graph, the user's "module combination strategies" and "troubleshooting methods" are analyzed to generate a radar chart of team collaboration capabilities. The system automatically identifies "atypical layout solutions" (such as a mixed ring and branch line layout). If the efficiency is better than the preset solution, it is marked as an "innovative solution" and promoted to the course forum for presentation.

[0058] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the module division is merely a logical function division, and other division methods may be used in actual implementation.

[0059] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0060] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A multi-scenario intelligent manufacturing teaching and training method based on digital twin technology, characterized by: The following steps are included: Step a, Data Collection and Processing: Based on the training course requirements, determine the types of smart manufacturing equipment and physical quantities that need to be collected, including equipment appearance and structural parameters, operating status parameters, and environmental parameters; select and install temperature, pressure, vibration, and visual sensors to collect the above physical quantities, and pre-process the data using data cleaning, format conversion, time alignment, signal fusion, and normalization to construct a high-quality data set as the data foundation for the construction of the smart manufacturing digital twin model; Step b: Constructing a smart manufacturing digital twin model: First, based on the data set obtained in step a, combined with CAD two-dimensional drawings, PLC control logic diagrams, and process flow chart engineering data, clarify the structural information and parameter relationships of the equipment, process, and environment. Use 3D modeling software to build a geometric model, and add the model's structural attributes, motion constraints, communication interface information, and operating logic to realize the construction of the equipment model, process model, and environment model, thereby forming a smart manufacturing digital twin model that reflects the actual system behavior. Construct a smart manufacturing digital twin model to accurately reflect the structure and operating status of the physical system of the smart manufacturing equipment. Step c, intelligent manufacturing multi-scenario integration: Based on the process flow template and logic control rule library constructed by production process specifications, equipment control logic and historical operation data, the intelligent manufacturing digital twin model constructed in step b is combined to quickly and conveniently build multiple intelligent manufacturing production scenarios. Different scenarios can be freely switched and combined, providing a three-dimensional virtual scene for subsequent experimental training processes, allowing learners to experience the workflow and operation requirements in different intelligent manufacturing environments; Step d. Custom development of experimental training processes: Develop an experimental training process construction tool, import the 3D virtual scene built in step c, and customize the experimental training process based on different experimental training content. Select templates from the school-enterprise cooperation template library (which stores enterprise-certified process templates, such as standard automotive welding processes) or the AI-generated template pool (based on personalized processes recommended by the DQN algorithm). Configure interactive controls (such as the same HMI interface components as the enterprise production line) and scoring nodes (such as the capacity compliance rate indicator that connects to the enterprise KPI). The AI-generated template pool supports automatic process generation through natural language input (such as generating a battery assembly fault handling process), and verifies logical feasibility in real time through a digital twin model (verification takes less than 5 minutes). It also supports style adjustment and teaching element layout, allowing for rapid construction of teaching and training processes and the rapid development of various experimental training processes. Step e: Publishing and sharing experimental training resources: Integrate the various intelligent manufacturing 3D virtual scenes independently built in step c with the experimental training processes custom developed in step d to form a virtual simulation resource containing the teaching experimental training process; Publish virtual simulation resources to the cloud platform's resource market, where developers' resource achievements can be transformed. For resources that do not involve commercial secrets or intellectual property disputes, an open access policy is implemented, allowing users to use and share them online for free or at low cost, breaking geographical restrictions, achieving "co-construction and sharing" of resources, and improving the utilization rate of virtual experimental training resources. Step f, experimental training teaching management: set up automatic assessment and learning tracking functions for each link of experimental training, use big data analysis, machine learning, knowledge graph construction, behavior trajectory mining and other technical means to conduct intelligent statistical analysis of learners' assessment data, support automatic scoring by step, automatic summary of experimental training, homework, and assessment results, generate analysis reports, and perform visual presentation; trace and track each training link of learners, and automatically form a technical skill portrait of students based on objective data such as learning time, training process and mastery status, and feed back the results of learners' training in each link to the teacher. Teachers can answer questions or consolidate explanations in a targeted manner in class based on the feedback results.

2. The multi-scenario intelligent manufacturing teaching and training method based on digital twin technology according to claim 1 is characterized by: The step b specifically comprises the following steps: First, based on the BIM parametric template library, the 3D virtual scene model skeleton is generated by combining CAD 2D drawings and collected data analysis. Second, the Newton-Euler method and other kinematic inverse solution algorithms are used to automatically configure the degrees of freedom of the equipment joints, supporting a six-axis robotic arm trajectory simulation error of less than 0.1mm. The integrated ANSYS finite element analysis engine simulates the thermal deformation and stress distribution of the equipment in real time. Then, the parameters of the intelligent manufacturing digital twin model were dynamically corrected through the LSTM neural network, increasing the simulation accuracy to 97.3%, completing the three-dimensional geometric modeling of the intelligent manufacturing production equipment and assembly line; Subsequently, by docking with the data interface of the acquisition system in step a, the real-time physical data is connected and integrated into the corresponding digital twin model to realize data synchronization between the physical world and the virtual world; finally, the physical process of the simulation equipment is simulated using the platform's simulation software, and the intelligent manufacturing digital twin model is presented to the user in a graphical manner through the three-dimensional visualization platform. The user operates the construction of the intelligent manufacturing digital twin model through the human-computer interaction interface, and can use PLC visual programming to observe and predict the behavior status of the intelligent manufacturing equipment in a visual manner, and then optimize the intelligent manufacturing digital twin model to further improve its accuracy and practicality; repeating this step can obtain multiple intelligent manufacturing digital twin models to form an intelligent manufacturing digital twin model library.

3. The multi-scenario intelligent manufacturing teaching and training method based on digital twin technology according to claim 2 is characterized by: The step c includes the following steps: based on the intelligent manufacturing digital twin model library constructed in step b, in accordance with the process flow template and logic control rule library constructed by production process specifications, equipment control logic and historical operation data, the configuration content defined is encapsulated by the process modeling language, and with the support of the digital twin cloud platform, the twin models in the intelligent manufacturing digital twin model library are combined and connected by dragging and dropping, and a semantic conversion tool is developed that supports 12 mainstream industrial communication protocols including Modbus-TCP, Profinet, EtherCAT, OPC UA, CANopen, EtherNet / IP, SERCOS, BACnet, HART, DeviceNet, IEC 61850, and Profibus, and the data is encapsulated in JSON-LD format to convert data between different devices and different protocols into a unified format; Based on the genetic algorithm, the layout of production line equipment is optimized for multiple objectives, namely minimizing logistics costs and maximizing production capacity. It also supports drag-and-drop adjustments, and finally completes the layout of the site and equipment to build various intelligent manufacturing production scenarios.

4. The multi-scenario intelligent manufacturing teaching and training method based on digital twin technology according to claim 3 is characterized by: The development experiment and training process construction tool in step d adopts a front-end and back-end separation architecture. The front-end is based on the Vue.js framework to design a graphical user interface, and the back-end is based on Node.js to implement process analysis and operation control, supporting visual configuration and logical arrangement of various elements in the experiment and training process; The development experiment and training process construction tool has a scene import function, which loads the three-dimensional virtual scene constructed in step c into the process editing interface as a teaching operation environment; it also has a page creation function, which quickly generates a training interaction page through preset teaching templates or custom components; supports adding various UI controls, and provides property setting and style adjustment functions; has a built-in flowchart-style DSL language and atomic instruction blocks of various standardized experiment and training processes; users build control processes by dragging and dropping combinations to realize data interaction and function calls between modules; select or define a suitable business process framework according to different experiment and training contents, integrate the scene and template, and adjust the style to realize customized development of the experiment and training process.

5. A multi-scenario intelligent manufacturing teaching and training system based on digital twin technology, characterized by: Includes the following modules, Data acquisition and processing module, used to collect and process physical quantity data of intelligent manufacturing equipment, processes, and environment; The intelligent manufacturing digital twin model construction module is used to perform geometric modeling and rendering based on the collected and processed physical quantity data of intelligent manufacturing equipment, processes, and environments. It also adds structural information attributes, motion parameters, control logic, communication data, behavioral rules, and operating status tags to the twin model to build an intelligent manufacturing digital twin model with dynamic response capabilities. The intelligent manufacturing multi-scenario integration module is used to quickly and conveniently build a variety of intelligent manufacturing production scenarios based on process flow templates and logic control rule libraries constructed from production process specifications, equipment control logic, and historical operation data. It encapsulates the defined configuration content through a process modeling language and combines them on the basis of the constructed intelligent manufacturing digital twin model. This allows for free switching and combination between different scenarios. The experimental training process custom development module is used to import the constructed 3D virtual scene, customize the experimental training process according to different experimental training contents, select the appropriate teaching resource template, integrate the scene and template, and realize the rapid development of various experimental training processes; The experimental training resource publishing and sharing module is used to integrate various independently built intelligent manufacturing digital twin application scenarios with custom-developed experimental training processes, forming virtual simulation resources containing teaching experimental training processes and publishing them to the resource market of the cloud platform, realizing "co-construction and sharing" of resources and improving the utilization rate of virtual experimental training resources; The experimental training teaching management module is used to conduct intelligent statistical analysis and tracking of the assessment data of each link of the learners, automatically form a technical skill portrait of the students, and feedback the results of the learners' training in each link to the teacher.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the device where the computer-readable storage medium is located executes the multi-scenario intelligent manufacturing teaching and training method based on digital twin technology as described in any one of claims 1 to 4.

7. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a program that can be run on the processor, and when the processor executes the program, it implements the multi-scenario intelligent manufacturing teaching and training method based on digital twin technology as described in any one of claims 1 to 4.

Images