A data processing method and system based on digital twin intelligent production line practical training

Abstract

The application discloses a kind of data processing method and system based on digital twin intelligent production line practical training, the method includes: obtaining the multi-source data set of intelligent production line practical training box in current teaching practical training process, and the student operation data set in historical practical training process, to adjust the first digital twin model of intelligent production line practical training box, obtain second digital twin model;Obtain the reference behavior data between each target component in second digital twin model under the preset standard control program, and simulate fault to second digital twin model, obtain third digital twin model;According to each student operation data set, obtain the target troubleshooting procedure corresponding to each target student, modification times and troubleshooting time consumption;Based on the actual running data of each target component under each target troubleshooting procedure, obtain a plurality of update state data corresponding to third digital twin model, compare with reference behavior data, generate the practical troubleshooting data of each target student, and send each practical troubleshooting data to teacher terminal.

Description

Technical Field

[0001] This application relates to the field of teaching and training technology, and more specifically, to a data processing method and system based on digital twin intelligent production line training. Background Technology

[0002] In the field of intelligent manufacturing, the teaching and training of intelligent production lines faces a prominent contradiction: high equipment costs, limited teaching scenarios, and a high degree of knowledge integration. Various educational institutions generally require students to complete a closed-loop training process, from production line planning and equipment debugging to troubleshooting, in a near-real industrial environment. However, due to the large footprint of real production lines, the inability to disassemble them at will, and the high cost of equipment that cannot withstand repeated trial and error by students, teaching scenarios have long been limited to cognitive visits or single-module demonstrations.

[0003] A typical solution to the aforementioned technical problems is to use a virtual simulation training platform based on digital twin technology. This solution constructs a virtual mirror of the production line, simulating the operation of the real production line in a virtual environment. Teachers can preset faults, and students can modify programs and troubleshoot through virtual interaction. However, this existing solution has significant drawbacks. First, there is a lack of real-time linkage between the virtual simulation and the real physical controller, meaning that the programs written by students cannot be verified on physical equipment, resulting in a disconnect between training results and actual working conditions. Second, the system uses a uniform fault scenario and interaction method for all students, failing to dynamically adapt to differences in student abilities, thus hindering the improvement of training efficiency and teaching effectiveness. Summary of the Invention

[0004] The main purpose of this application is to provide a data processing method and system based on digital twin intelligent production line training, so as to solve the technical problems in the existing technology where the lack of virtual-real linkage leads to the disconnect between training results and reality, and the single interaction method cannot adapt to the differences in students' abilities, resulting in poor teaching effect.

[0005] To achieve the above objectives, the first aspect of this application proposes a data processing method based on digital twin intelligent production line training, comprising:

[0006] Obtain multi-source datasets of the intelligent production line training box during the current teaching and training process, as well as student operation datasets of multiple historical target students during the historical training process; Based on the physical location of each target component in the multi-source dataset, and combined with the corresponding virtual expected location in the first digital twin model of the intelligent production line training box, a location compensation coefficient is generated to adjust the first digital twin model and obtain the second digital twin model. Acquire baseline behavior data between target components in the second digital twin model under a preset standard control program, perform fault simulation processing on the second digital twin model, and obtain a third digital twin model; Based on the student operation dataset, determine the target operation mode of the student terminal for each target student, so as to obtain the target troubleshooting program, number of modifications, and troubleshooting time for each target student. Obtain the actual operating data of each target component under each target troubleshooting procedure, and update the third digital twin model according to each actual operating data to obtain multiple corresponding update status data; Each updated status data is compared with the baseline behavior data to obtain the troubleshooting operation results for each target student. Combined with the corresponding number of modifications and troubleshooting time, the training troubleshooting data for each target student is generated and sent to the teacher terminal.

[0007] Secondly, this application provides a data processing system based on digital twin intelligent production line training, including: The acquisition module is used to acquire multi-source datasets of the intelligent production line training box during the current teaching and training process, as well as student operation datasets of multiple historical target students during the historical training process. The adjustment module is used to generate a position compensation coefficient based on the physical position of each target component in the multi-source dataset and the corresponding virtual expected position in the first digital twin model of the intelligent production line training box, so as to adjust the first digital twin model and obtain the second digital twin model. The simulation module is used to acquire the baseline behavior data between each target component in the second digital twin model under a preset standard control program, perform fault simulation processing on the second digital twin model, and obtain a third digital twin model. The determination module is used to determine the target operation mode of each target student's student terminal based on the student operation dataset, so as to obtain the target troubleshooting program, number of modifications and troubleshooting time corresponding to each target student. The update module is used to obtain the actual operating data of each target component under each target troubleshooting procedure, and update the third digital twin model according to each actual operating data to obtain multiple corresponding update status data. The generation module is used to compare each updated status data with the baseline behavior data to obtain the troubleshooting operation results corresponding to each target student. Combining the corresponding number of modifications and troubleshooting time, the module generates the training troubleshooting data for each target student and sends the training troubleshooting data to the teacher terminal.

[0008] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the steps of a data processing method based on digital twin intelligent production line training.

[0009] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a computer, implements the steps of a data processing method based on digital twin intelligent production line training.

[0010] The technical solutions provided by the embodiments of this application may include the following beneficial effects: In this application, by acquiring multi-source datasets of intelligent production line training boxes during current teaching and training processes and student operation datasets of multiple historical target students during historical training processes, a data foundation is provided for subsequent virtual-physical linkage and personalized teaching. Position compensation coefficients are generated based on the physical positions of each target component in the multi-source datasets and the corresponding virtual expected positions in the first digital twin model. The first digital twin model is then adjusted to obtain a second digital twin model, enabling the virtual model to follow the movement state of the physical training box in real time, establishing a closed-loop verification environment for virtual-physical linkage. Benchmark behavior data between each target component in the second digital twin model is extracted under a preset standard control program, and fault simulation processing is performed to obtain a third digital twin model. Fault simulation is then performed in the synchronized digital twin model, providing a quantifiable normal operating condition reference standard for troubleshooting assessment, transforming fault diagnosis training from abstract cognition to precise comparison based on evidence. Based on the student operation dataset, the target operation mode of each student's terminal is determined. This allows for the acquisition of the corresponding target troubleshooting program, number of modifications, and troubleshooting time for each student. This enables students of different ability levels to receive appropriate operational permissions and guidance on the same training platform, achieving dynamic adaptation of practical training. Actual operational data of each target component under each target troubleshooting program is collected. Based on this data, the third digital twin model is updated to obtain multiple updated state data, ensuring that the model state accurately reflects the actual impact of student troubleshooting operations on physical equipment. The updated state data is compared with baseline behavior data to obtain the troubleshooting operation results for each target student. Combined with the corresponding number of modifications and troubleshooting time, practical training troubleshooting data for each target student is generated and sent to the teacher's terminal. This allows troubleshooting scores to dynamically adjust evaluation weights based on fault type and parameters, providing objective and traceable practical training records to support teaching management. Attached Figure Description

[0011] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings: Figure 1 A flowchart of a data processing method based on digital twin intelligent production line training provided in this application; Figure 2 A schematic diagram of a data processing system based on digital twin intelligent production line training provided in this application. Detailed Implementation

[0012] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0013] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0014] In this application, the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing this application and its embodiments, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation.

[0015] Furthermore, in addition to indicating location or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0016] Furthermore, the terms "installation," "setup," "equipped with," "connection," "linked," and "socketing" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0017] The core of this application is to provide a data processing method for intelligent production line training based on digital twins. This method uses a computer system as the central hub to connect and coordinate data interaction and task collaboration among the intelligent production line training kit, student terminals, and teacher terminals. In the teaching scenario of intelligent production line debugging and troubleshooting in vocational schools or corporate training institutions, it constructs an adaptive training closed loop of "physical training, virtual simulation, and teaching management." On the one hand, the computer system collects multi-source data such as the physical position and operating speed of each moving part from the intelligent production line training kit in real time, driving the digital twin model to maintain synchronous linkage with the physical equipment. On the other hand, it retrieves operation datasets of each target student from a historical database to assess their proficiency level, automatically matching guided or autonomous differentiated operation modes for the student terminals. Students receive standard troubleshooting procedures and fault information through student terminals. They then write and submit their troubleshooting programs in the corresponding operating modes. The computer system sends these programs to the physical controller to drive the actual components of the training box. Simultaneously, it collects actual operating data to update the digital twin model. By comparing the updated model status with the baseline behavior data component by component and stage by stage, it generates the training troubleshooting data for each student and sends it to the teacher's terminal. This allows teachers to view the troubleshooting results and operation process data of all students on a unified interface, achieving a complete closed-loop training management system from production line planning and debugging to personalized troubleshooting training and teaching evaluation feedback.

[0018] A flowchart of one specific implementation method is shown below. Figure 1 As shown, the method includes: Step 101: Obtain the multi-source dataset of the intelligent production line training box in the current teaching and training process, and the student operation dataset of multiple historical target students in the historical training process.

[0019] Step 102: Based on the physical location of each target component in the multi-source dataset, and combined with the corresponding virtual expected location in the first digital twin model of the intelligent production line training box, generate a location compensation coefficient to adjust the first digital twin model and obtain the second digital twin model.

[0020] Step 103: Obtain the baseline behavior data between each target component in the second digital twin model under the preset standard control program, perform fault simulation processing on the second digital twin model, and obtain the third digital twin model.

[0021] Step 104: Based on the student operation dataset, determine the target operation mode of the student terminal for each target student, so as to obtain the target troubleshooting program, number of modifications, and troubleshooting time for each target student.

[0022] Step 105: Obtain the actual operating data of each target component under each target troubleshooting procedure, and update the third digital twin model according to each actual operating data to obtain the corresponding multiple updated status data.

[0023] Step 106: Compare each updated status data with the baseline behavior data to obtain the troubleshooting operation results for each target student. Combine the corresponding number of modifications and troubleshooting time to generate the practical troubleshooting data for each target student, and send each practical troubleshooting data to the teacher's terminal.

[0024] In step 101 above, the current teaching and training process refers to the ongoing training course period, which is the time interval from when the teacher starts the training task to when the training task ends. The historical training process refers to all training course periods that have been completed and have generated operation records before the start of this training course.

[0025] The intelligent production line training box refers to an integrated desktop training device that combines multiple moving parts such as conveyor belts, gripping devices, and sorting mechanisms, as well as physical sensors and physical controllers. It is used to simulate the material conveying, workpiece gripping and sorting processes of a real production line in a limited space, enabling students to complete production line debugging and troubleshooting training in a desktop environment.

[0026] Multi-source datasets refer to the collection of physical positions and running speeds of various moving parts at multiple sampling times, collected in real time by physical sensors arranged on each moving part of the intelligent production line training box at a preset sampling frequency.

[0027] Historical target students refer to those who have participated in at least one training task and generated operational records during the history training process. Target students refer to the students participating in this training task during the current teaching and training process.

[0028] The student operation dataset refers to the records of the operation accuracy, number of operation deviations, and task completion time of multiple historical target students when completing troubleshooting tasks in the historical training process. Each record corresponds to the operation performance of a historical target student in a training task.

[0029] One implementation method for acquiring data in this application embodiment is as follows: After the teacher starts the intelligent production line troubleshooting training task, the physical sensors such as encoders and grating rulers arranged on the conveyor belt, gripping device and sorting mechanism in the intelligent production line training box are used to collect the physical position and running speed of each moving part at each collection time in real time at a preset sampling frequency. The physical position is the displacement reading of each moving part along the two orthogonal directions of the working plane of the training box, and the running speed is the instantaneous rate reading of each moving part at the corresponding collection time. The physical position and running speed of each moving part at each collection time are summarized to obtain a multi-source dataset.

[0030] The system retrieves the operation accuracy rate, number of operation deviations, and task completion time of each target student participating in the current training from the historical database. The three types of recorded data are then organized according to student identity and timestamp to obtain the student operation dataset.

[0031] In step 102 above, firstly, based on the physical positions of each target component in the multi-source dataset and combined with the corresponding virtual expected positions in the first digital twin model of the intelligent production line training box, position compensation coefficients are generated. Then, the generated position compensation coefficients are used to correct the virtual positions of each target component in the first digital twin model, making the corrected virtual positions of each target component more consistent with the physical positions of the corresponding components in the physical training box. The corrected first digital twin model is then used as the second digital twin model. The first digital twin model refers to the initial virtual model of the intelligent production line training box pre-constructed before the start of the current teaching and training process. This model includes the three-dimensional geometry of each target component, assembly relationships, and the virtual expected positions calculated according to the ideal kinematic model. The second digital twin model refers to the digital twin model in which the virtual positions of each target component are synchronized in real time with the physical positions of the corresponding components in the physical training box after adjustment by the position compensation coefficients.

[0032] The step "Generating position compensation coefficients based on the physical locations of each target component in the multi-source dataset and the corresponding virtual expected locations in the first digital twin model of the intelligent production line training box" specifically includes the following steps: Step 201: Based on the physical location of each target component in the multi-source dataset and the corresponding virtual expected location in the first digital twin model of the intelligent production line training box, calculate the difference between the physical location of each target component and the corresponding virtual expected location at the same acquisition time to obtain the position deviation sequence of each target component at different acquisition times.

[0033] Step 202: Based on the distribution trend of the position deviation values ​​in each position deviation sequence with the acquisition time, determine the corresponding fitting curve, and extract the slope and intercept of each fitting curve as the position compensation coefficient of the corresponding target component.

[0034] In step 201 above, the target component refers to the various functional components in the intelligent production line training box that participate in the production line operation and troubleshooting, including the conveyor belt, gripping device and sorting mechanism.

[0035] Physical position refers to the displacement reading obtained by physical sensors arranged on each target component of the intelligent production line training box at a certain acquisition time. The displacement reading represents the coordinate value of the target component in the working plane of the training box along two orthogonal directions.

[0036] The virtual expected position refers to the theoretical coordinate value of the corresponding target component in the first digital twin model calculated according to the ideal kinematic equation at the same acquisition time.

[0037] The position deviation sequence refers to a set of deviation values ​​arranged in chronological order of acquisition time. Each deviation value is the difference between the physical position and the virtual expected position at the same acquisition time. The positive or negative sign of the deviation value indicates the direction of the offset of the physical position relative to the virtual expected position.

[0038] In this embodiment, the physical positions of the conveyor belt, gripping device, and sorting mechanism at the first, second, third, and up to the Nth acquisition time are extracted from a multi-source dataset. These physical positions are represented by coordinate values. Virtual expected positions of the conveyor belt, gripping device, and sorting mechanism at the same acquisition time are extracted from a first digital twin model. The difference between the physical position coordinates and the virtual expected position coordinates of the conveyor belt at each acquisition time is calculated to obtain the position deviation value of the conveyor belt at each acquisition time. All the position deviation values ​​of the conveyor belt are arranged chronologically according to the acquisition time to form a position deviation sequence. Similarly, the above position deviation value calculation is performed on the gripping device and the sorting mechanism respectively to obtain the position deviation sequence of the gripping device and the position deviation sequence of the sorting mechanism.

[0039] In step 202 above, the distribution trend refers to the overall trend of change when the position deviation value is plotted on a preset coordinate system with the acquisition time as the independent variable and the position deviation value as the dependent variable. The distribution trend reflects the change law of the position deviation value increasing, decreasing or remaining stable as the acquisition time progresses. The preset coordinate system refers to a plane rectangular coordinate system established with the acquisition time as the horizontal axis and the position deviation value as the vertical axis. The horizontal axis is used to mark the time sequence of each acquisition time, and the vertical axis is used to mark the numerical value of each position deviation value. The intersection of the horizontal axis and the vertical axis is the origin, which corresponds to the state where the acquisition time is zero and the position deviation value is zero.

[0040] A fitted curve is a straight line defined in a preset coordinate system. The sum of the squares of the vertical distances between each point on the line and the actual position deviation value corresponding to the acquisition time is minimized. This line can describe the main trend of the position deviation value changing with the acquisition time in a simplified form.

[0041] The position compensation coefficient is composed of the slope and the intercept, and is used to describe the linear change of the position deviation of the corresponding target component with the acquisition time. The slope refers to the average change of the position deviation value of the dependent variable when the acquisition time of the independent variable changes by one unit time in the fitted curve, reflecting the drift rate of the position deviation with the acquisition time. The intercept refers to the ordinate value of the intersection of the fitted curve and the vertical axis, reflecting the reference offset of the position deviation at the initial acquisition time.

[0042] In this embodiment, taking the position deviation sequence of a conveyor belt as an example, each acquisition time is used as the independent variable, and the corresponding position deviation value of the conveyor belt at each acquisition time is used as the dependent variable. Multiple data points are marked in a preset coordinate system, where the horizontal axis represents the acquisition time and the vertical axis represents the position deviation value. The overall distribution trend of these data points is analyzed. If the data points are roughly distributed along a straight line, a straight line is determined in the preset coordinate system such that the sum of the squares of the vertical distances from each data point to this line is minimized. This straight line is the fitting curve. The vertical coordinate value of any point on the fitting curve represents the position deviation value estimated by the fitting curve at the acquisition time corresponding to that point. The estimated position deviation value at any acquisition time is equal to the slope multiplied by the acquisition time corresponding to that point plus the intercept. The slope and intercept of the fitting curve are extracted as the position compensation coefficient of the conveyor belt. Similarly, the above fitting process is repeated for the position deviation sequences of the gripping device and the sorting mechanism, respectively, and their respective slopes and intercepts are extracted to obtain the position compensation coefficients of the gripping device and the sorting mechanism. The position compensation coefficients of each target component are applied to the virtual expected position of the corresponding component in the first digital twin model. That is, for each acquisition time, the correction value of the virtual expected position is equal to the original virtual expected position plus the estimated position deviation value corresponding to that acquisition time, so that the corrected virtual position is consistent with the physical position, thus obtaining the second digital twin model.

[0043] The embodiments of this application enable the adjusted second digital twin model to follow the movement state of the physical training box in real time. When students perform troubleshooting operations, the position of the component observed in the virtual model is consistent with the actual operating position of the physical training box, thus realizing synchronous linkage between virtual simulation and physical control.

[0044] In step 103 above, the baseline behavioral data between the target components in the second digital twin model under the preset standard control program is first obtained. Specifically, the preset standard control program is loaded into the virtual controller of the second digital twin model. The virtual controller drives the operation of each target component in the second digital twin model according to the instruction sequence of the preset standard control program. During the operation, the operating parameter sequence of each target component at different operating stages and the cooperative timing relationship between each target component are collected from the second digital twin model at a preset sampling frequency. The collected operating parameter sequence and cooperative timing relationship are used as the baseline behavioral data. The preset standard control program refers to the pre-written and verified normal operation control program of the production line. This program drives each target component in the second digital twin model to operate collaboratively according to the correct process flow, and the action timing and cooperative relationship between each target component conform to the standard specifications of the production line design. The baseline behavior data refers to the set of operating parameter sequences and inter-component collaboration timing relationships of each target component in the second digital twin model under the drive of the preset standard control program at different operating stages. This data reflects the standard operating state of the production line under fault-free conditions. The operating parameter sequence refers to the set of physical quantities of each target component arranged in the order of acquisition time in each operating stage. The collaboration timing relationship refers to the time sequence of action connection and signal transmission relationship between different target components during operation.

[0045] Then, fault simulation processing is performed on the second digital twin model to obtain the third digital twin model. This step specifically includes the following steps: Step 301: Extract the fault types and fault parameters of each target component corresponding to the current teaching and training process task from the pre-built fault simulation database.

[0046] Step 302: Based on the fault type and fault parameters of each target component, adjust the operating parameters and / or connection relationships of the corresponding target components in the second digital twin model to obtain the third digital twin model.

[0047] In step 301 above, the fault simulation database refers to a pre-built set of fault parameters. This database pre-stores the correspondence between various fault types and fault parameters. Each fault type corresponds to a different parameter adjustment method. When the training task starts, the corresponding fault parameters are extracted from this database according to the method set by the teacher's terminal or randomly selected.

[0048] The current teaching and training task refers to the troubleshooting training task set by the teacher's terminal. This task specifies the types and severity of faults that students need to troubleshoot in this training. The fault type of the target component refers to the specific way in which the target component deviates from its normal operating state, including at least one of the following: the target component's operating speed limit is reduced, there is a delay in signal transmission between the target component and other components, or the data link between the target component and other components is broken. The fault parameters of the target component refer to the specific values ​​of the abnormality corresponding to the fault type, used to quantitatively describe the severity of the fault.

[0049] In this embodiment of the application, based on the current teaching and training process task set by the teacher terminal, the fault type and fault parameters corresponding to the current teaching and training process task are extracted from the fault simulation database. For example, if the upper limit of the conveyor belt's running speed is reduced, the fault parameter is that the upper limit of the speed is reduced to 60% of the normal upper limit of the speed. At the same time, if the signal transmission delay between the gripping device and other components is increased, the fault parameter is that the delay time is increased by 300ms.

[0050] In step 302 above, the third digital twin model refers to the digital twin model that includes the abnormal state set in the current teaching and training process task after the fault simulation is completed, and is used to provide students with troubleshooting training scenarios. The operating parameters or connection relationships of each target component in the model have been modified. When the model is running, it will present abnormal phenomena corresponding to the fault type. The abnormal state can be understood as the abnormal production line state presented by the fault type and fault parameters of each target component applied in the second digital twin model through fault simulation, such as the slowdown of the conveyor belt, the delay of the gripping device action, or the sorting mechanism not responding to the trigger signal.

[0051] In this embodiment, the second digital twin model is adjusted according to the fault type and fault parameters, and the modified second digital twin model is used as the third digital twin model.

[0052] The third digital twin model in this application embodiment enables the troubleshooting training scenario to not only present the real fault performance linked with the physical training box, but also provide a referenceable normal operation standard. Students can conduct troubleshooting training in fault scenarios with clear objective references, and the judgment of troubleshooting results can be based on the comparison of benchmark data rather than subjective experience.

[0053] In step 104 above, the step "determine the target operation mode of each target student's terminal based on each student's operation dataset" specifically includes the following steps: Step 401: Extract the operation accuracy rate, number of operation deviations and task completion time of the corresponding historical target students in the history teaching and training process from each student operation dataset, and calculate the proficiency score of each historical target student in combination with the corresponding preset influence weight.

[0054] Step 402: Match each target student with each historical target student. Use the proficiency score of the historical target student corresponding to the target student whose identity is successfully matched as the corresponding proficiency score. Set the proficiency score of the target student whose identity is not matched as the initial proficiency score.

[0055] Step 403: Based on the proficiency scores of each target student, set the target operation mode for each target student's student terminal.

[0056] In step 401 above, the preset influence weight refers to the influence coefficients that are pre-set for operation accuracy, number of operation deviations, and task completion time, respectively, reflecting the importance of the above three indicators in the proficiency assessment. The sum of the preset influence weights is 1. The preset influence weights can be set according to the type of training task. For training tasks that emphasize operation standardization, the weights corresponding to operation accuracy and number of operation deviations can be set higher; for training tasks that emphasize efficiency, the weights corresponding to task completion time can be set higher. In this embodiment of the application, the numerical value of the preset influence weights corresponding to the above three indicators is not limited.

[0057] The operation accuracy rate is the average ratio of the number of correct operation steps to the total number of operation steps in the history teaching and training process for the target students. The number of operation deviations is the average cumulative number of deviations between the actual operation steps and the standard operation steps in the history teaching and training process. The task completion time is the average time from the start of the troubleshooting task to the submission of the troubleshooting program in the history teaching and training process.

[0058] The proficiency score is a weighted sum of the historical target student's operational accuracy, number of operational deviations, and task completion time. This score is used to quantitatively measure the overall operational level of the historical target student in troubleshooting training. The higher the proficiency score, the higher the level of proficiency.

[0059] In this embodiment of the application, the operation accuracy rate, number of operation deviations and task completion time of each historical target student in the historical teaching and training process are extracted from the student operation dataset. The operation accuracy rate, number of operation deviations and task completion time are weighted and summed according to the corresponding preset influence weights to obtain the proficiency score of each historical target student.

[0060] In step 402 above, the initial proficiency score refers to the preset starting score for new students who are participating in practical training for the first time among the target students and who do not have a matching record among the historical target students. This starting score is a fixed value that is preset to provide a uniform evaluation starting point for new students.

[0061] In this embodiment, the student ID or identity identifier of each target student in the current teaching and training process is compared one by one with the student ID or identity identifier of the historical target students. When the identity identifier of a target student has the same record in the historical target students, the identity matching is determined to be successful, and the proficiency score of the historical target student is directly used as the proficiency score of the target student. When the identity identifier of a target student has no record in the historical target students, the identity matching is determined to be unsuccessful, and a preset initial proficiency score is assigned to the target student.

[0062] In step 403 above, the target operation mode refers to the setting of the operation permission level and interface prompt level configured on the student terminal based on the proficiency score of the target student. Different proficiency scores correspond to different operation modes so that students with different ability levels can obtain an operation environment that is suitable for their abilities.

[0063] In this embodiment, the proficiency score of each target student is compared with a preset grading threshold. When a target student's proficiency score is lower than the preset grading threshold, the student terminal is set to guided operation mode. In this mode, the troubleshooting interface displays step-by-step operation prompts and restricts modification permissions for non-faulty sections of the standard troubleshooting basic program. When a target student's proficiency score is higher than or equal to the preset grading threshold, the student terminal is set to autonomous troubleshooting mode. In this mode, operation prompts are disabled and full modification permissions for the standard troubleshooting basic program are granted. The preset grading threshold refers to the proficiency score boundary used to divide the guided operation mode and the autonomous troubleshooting mode. This boundary is a pre-set fixed value that can be set according to the difficulty level of the training task. For more difficult training tasks, the preset grading threshold can be set higher to allow more students to enter the guided operation mode; for less difficult training tasks, the preset grading threshold can be set lower to allow more students to enter the autonomous troubleshooting mode. This embodiment does not limit the value of the preset grading threshold.

[0064] The step "Obtain the target troubleshooting program, number of modifications, and troubleshooting time for each target student" specifically includes the following steps: Step 411: Extract the standard troubleshooting basic program corresponding to the abnormal state of the third digital twin model from the preset fault program library, as well as the pre-stored fault point information and fault description associated with the standard troubleshooting basic program.

[0065] Step 412: Load the standard troubleshooting basic program and the corresponding pre-stored fault point information and fault description into each student terminal. Combined with the target operation mode of the student terminal, obtain the target troubleshooting program, number of modifications, and troubleshooting time corresponding to the modification operation of the standard troubleshooting basic program by each target student on the corresponding student terminal.

[0066] In step 411 above, the pre-built fault program library refers to a pre-constructed set of troubleshooting programs. This library pre-stores standard troubleshooting basic programs corresponding to various fault types. Each standard troubleshooting basic program is associated with corresponding pre-stored fault point information and fault description. The pre-stored fault point information refers to the location identifier of each fault code in the program segment of the standard troubleshooting basic program, such as the line number range or program block number of the fault code, used to indicate the specific program segment that students should pay attention to when troubleshooting. The fault description refers to the textual description of the abnormal phenomenon corresponding to each fault point in natural language, such as the conveyor belt running speed being lower than normal or the gripping device response signal having a delay, used to help students understand the correspondence between fault codes and abnormal phenomena on the production line.

[0067] The standard troubleshooting basic procedure refers to the production line control program that includes preset fault points. This program has embedded fault codes corresponding to the current training task on the basis of the normal control program. The target students need to locate and modify these fault codes in the current teaching and training to restore the normal operation of the production line.

[0068] In this embodiment, based on the fault type in the third digital twin model, a standard troubleshooting basic program matching the fault type is searched from a pre-set fault program library. For example, if the third digital twin model sets faults such as the upper limit of the conveyor belt running speed being reduced to 60% of the normal value and the signal transmission delay of the gripping device increasing by 300ms, then a standard troubleshooting basic program containing the above two fault codes is extracted from the fault program library. At the same time, the pre-stored fault point information and fault description associated with this program are extracted. The first fault point information points to the program line number where the upper limit of the conveyor belt speed assignment statement is located, and the corresponding fault description is that the upper limit of the conveyor belt running speed is limited; the second fault point information points to the program line number where the gripping device signal delay parameter assignment statement is located, and the corresponding fault description is that there is a delay in the gripping device response signal.

[0069] In step 412 above, the target troubleshooting program refers to the version of the troubleshooting program submitted by each target student after modifying the fault codes in the standard troubleshooting basic program in the target operation mode of their respective student terminal. Each target student corresponds to one target troubleshooting program, and different target troubleshooting programs are different due to the different modification methods and modification results of the students.

[0070] In this embodiment, the standard troubleshooting basic program and associated pre-stored fault point information and fault descriptions are distributed to the student terminals of each target student. In guided operation mode, the student terminal's troubleshooting interface displays step-by-step operation prompts, guiding the target student to locate the corresponding program lines for each fault point. A corresponding fault description is displayed at each fault point. Students can only modify the program code in the section marked as a fault point, and their modification permissions are limited to the line number range corresponding to the fault point. In self-directed troubleshooting mode, the student terminal's troubleshooting interface does not display operation prompts, only fault descriptions as reference information. Students can freely modify all sections of the standard troubleshooting basic program. After each target student completes the modification of the standard troubleshooting basic program in their respective target operation mode, they submit the modified troubleshooting program through their student terminal. The troubleshooting programs submitted by each target student are obtained as the corresponding target troubleshooting programs. Simultaneously, the number of times each target student modifies the program code during the modification process, and the time from the completion of loading the standard troubleshooting basic program to the submission of the target troubleshooting program are counted as the troubleshooting time.

[0071] This application embodiment quantifies and evaluates students' historical operation data and matches differentiated terminal operation modes, enabling students with existing training experience and high proficiency to explore troubleshooting with greater freedom in autonomous mode, while students who are participating for the first time or have lower proficiency can receive sufficient operation prompts in guided mode. This simultaneously meets the training needs of students with different ability levels, breaks the single teaching mode where all students face the same interaction method, and provides objective data for evaluating students' troubleshooting process by recording the number of modifications and troubleshooting time.

[0072] In step 105 above, the actual operating data of each target component under each target troubleshooting program is first obtained. Specifically, the target troubleshooting programs submitted by each target student are sent to the physical controller, so that the physical controller executes the target troubleshooting program to generate corresponding control commands, driving each target component in the intelligent production line training box to run. During the operation, physical sensors arranged on each target component collect the actual physical position and running speed of each target component at different sampling times according to a preset sampling frequency. The collected actual physical position and running speed are used as the actual operating data corresponding to the target troubleshooting program. Since the target troubleshooting programs submitted by each target student are different, each target troubleshooting program generates a set of actual operating data after running. Therefore, multiple sets of actual operating data corresponding one-to-one with each target troubleshooting program are finally obtained. Among them, the actual operating data refers to the set of actual physical position and running status of each target component collected in real time by physical sensors at different sampling times after each target troubleshooting program is sent to the physical controller and the physical controller executes the troubleshooting program to drive each target component to run.

[0073] Then, based on the actual operational data, the third digital twin model is updated to obtain multiple corresponding updated status data. This step specifically includes the following steps: Step 501: Extract the actual physical position of each target component from the actual operating data at different acquisition times, and calculate the deviation between the actual physical position of each target component at each acquisition time and the corresponding virtual position in the third digital twin model.

[0074] Step 502: Take the sum of the squares of the deviation values ​​corresponding to each target component at each acquisition time as the optimization target, and solve for the position compensation amount of each target component.

[0075] Step 503: Based on the position compensation amount of each target component, update the virtual position of each target component in the third digital twin model to obtain the fourth digital twin model, and obtain multiple update status data corresponding to the fourth digital twin model when running the troubleshooting program for each target.

[0076] In step 501 above, the actual physical position refers to the displacement readings obtained by physical sensors at each acquisition moment when the target troubleshooting program submitted by the student is sent to the physical controller and the physical controller drives the corresponding target component in the intelligent production line training box to actually run.

[0077] Virtual position refers to the spatial coordinates of the corresponding target component in the third digital twin model at the same acquisition time. The deviation value of the virtual position refers to the difference between the actual physical position and the virtual position at the same acquisition time. The sign of the deviation value indicates the direction of the offset of the actual physical position relative to the virtual position.

[0078] In this embodiment, the actual physical positions of the conveyor belt, gripping device, and sorting mechanism at each acquisition time are extracted from the actual operating data corresponding to each target troubleshooting procedure; the virtual positions of the conveyor belt, gripping device, and sorting mechanism at each acquisition time are extracted from the third digital twin model. Specifically, the difference between the actual physical position and the virtual position of the conveyor belt at each acquisition time is calculated to obtain the deviation value of the conveyor belt at each acquisition time. Similarly, the above deviation value calculation is also performed on the gripping device and the sorting mechanism to obtain the deviation values ​​corresponding to the gripping device and the sorting mechanism at each acquisition time.

[0079] In step 502 above, the position compensation amount refers to a set of correction values ​​that need to be solved to make the virtual position of each target component in the third digital twin model consistent with the actual physical position.

[0080] In this embodiment, taking a conveyor belt as an example, the squares of multiple deviation values ​​corresponding to the conveyor belt at each collection time are taken and then summed to obtain the sum of squares of the deviation values. Based on the least squares method, the position compensation amount of the conveyor belt is solved with the goal of minimizing the sum of squares of the deviation values. Similarly, the above solution process is repeated for the gripping device and the sorting mechanism to obtain the position compensation amount of the gripping device and the position compensation amount of the sorting mechanism.

[0081] In step 503 above, the fourth digital twin model refers to the digital twin model obtained after correcting the virtual positions of each target component in the third digital twin model with position compensation. The virtual positions of each target component in this model are consistent with the actual physical positions corresponding to the target troubleshooting program.

[0082] Updated status data refers to the set of virtual positions and operating states of each target component presented by the fourth digital twin model when running each target troubleshooting program. Each target troubleshooting program corresponds to a set of updated status data.

[0083] In this embodiment, the position compensation amount of the conveyor belt is applied to the virtual position of the conveyor belt in the third digital twin model, the position compensation amount of the gripping device is applied to the virtual position of the gripping device, and the position compensation amount of the sorting mechanism is applied to the virtual position of the sorting mechanism. This ensures that the virtual positions of each target component are consistent with their corresponding actual physical positions, and the corrected model is used as the fourth digital twin model. The target troubleshooting program is run again in the fourth digital twin model, and the virtual positions and operating states of each target component during the process are collected as update status data corresponding to the target troubleshooting program. The above update process is repeated for each target troubleshooting program. Since the target troubleshooting programs submitted by each target student are different, each troubleshooting program generates a set of update status data after running, ultimately obtaining multiple update status data corresponding one-to-one with each target troubleshooting program.

[0084] This application embodiment sends the student troubleshooting program to the physical controller to drive the real components to run, collects actual running data and updates the digital twin model with the goal of minimizing the sum of squared deviations, so that the state of the virtual model can accurately reflect the actual control effect of the student troubleshooting program on the physical equipment. The component running state observed by the student in the virtual model is consistent with the actual running results of the physical training box, thereby verifying the execution effect of the troubleshooting program written by the student on the real equipment.

[0085] In step 106 above, each updated status data is compared with the baseline behavior data to obtain the troubleshooting operation results for each target student. Then, combined with the corresponding number of modifications and troubleshooting time, the training troubleshooting data for each target student is generated and sent to the teacher terminal. The training troubleshooting data refers to the complete set of data related to the student's troubleshooting operation generated in a complete training task, including the troubleshooting operation results, number of modifications, and troubleshooting time, which is used to provide feedback to the teacher terminal on the completion status of each target student's troubleshooting training.

[0086] The step "comparing each updated status data with the baseline behavior data to obtain the troubleshooting operation results for each target student" specifically includes the following steps: Step 601: Extract the simulated operating parameter sequence of each target component under different operating stages in each target troubleshooting program from each updated status data, and extract the standard operating parameter sequence of each target component under different operating stages in the preset standard control program from the baseline behavior data.

[0087] Step 602: Calculate the difference between the simulated operating parameter sequence and the standard operating parameter sequence of the same target component in the same operating stage to obtain the parameter deviation sequence of each target component in different operating stages.

[0088] Step 603: Based on the multiple parameter deviation sequences and corresponding preset weights of each target component under each target troubleshooting procedure, calculate the troubleshooting score corresponding to each target student. Combined with the preset score threshold, generate the troubleshooting operation result corresponding to each target student. Each preset weight is set based on the fault type and fault parameter of each target component corresponding to the current teaching and training process task.

[0089] In step 601 above, the simulated operating parameter sequence refers to the set of operating parameter values ​​of each target component in each operating stage, arranged in chronological order of acquisition time, when the fourth digital twin model is running a certain target troubleshooting program. This sequence reflects the actual operating status of the production line driven by the target troubleshooting program.

[0090] The standard operating parameter sequence refers to the set of operating parameter values ​​of each target component in the baseline behavior data, which are arranged in chronological order of acquisition time during the corresponding operating phase under the drive of the preset standard control program. This sequence reflects the standard operating state of the production line under fault-free conditions.

[0091] The operation phase refers to the different process flow segments divided in the preset standard control program. Each operation phase corresponds to the time interval in which the production line completes a specific operation.

[0092] In this embodiment, for the updated status data corresponding to each target troubleshooting procedure, simulated operating parameter sequences for the conveyor belt during the transport phase, the gripping device during the gripping phase, and the sorting mechanism during the sorting phase are extracted according to the operating stage. Simultaneously, standard operating parameter sequences for the conveyor belt during the transport phase, the gripping device during the gripping phase, and the sorting mechanism during the sorting phase are extracted from the baseline behavior data. The extracted data are aligned according to the same acquisition time to ensure that the number of data points in the simulated operating parameter sequence and the standard operating parameter sequence are the same within the same operating stage, and that the acquisition times correspond one-to-one.

[0093] In step 602 above, the parameter deviation sequence refers to the numerical sequence formed by arranging the difference between the simulated operating parameter value and the standard operating parameter value at each acquisition time for the same target component in the same operating phase, according to the order of acquisition time. The magnitude of the parameter deviation value reflects the degree of deviation between the operating state of the target component in the operating phase and the standard operating state.

[0094] In this embodiment, for the conveyor belt, the difference between the simulated operating speed value in the simulated operating parameter sequence of the transport stage at each acquisition moment and the standard operating speed value in the standard operating parameter sequence is calculated to obtain the parameter deviation sequence of the conveyor belt during the transport stage. For the gripping device, the difference between the simulated displacement value in the simulated operating parameter sequence of the gripping stage at each acquisition moment and the standard displacement value in the standard operating parameter sequence is calculated to obtain the parameter deviation sequence of the gripping device during the gripping stage. For the sorting mechanism, the difference between the simulated action trigger time in the simulated operating parameter sequence of the sorting stage at each acquisition moment and the standard action trigger time in the standard operating parameter sequence is calculated to obtain the parameter deviation sequence of the sorting mechanism during the sorting stage. The above difference calculations are performed on the conveyor belt, gripping device, and sorting mechanism at all operating stages to obtain multiple parameter deviation sequences for each target component at different operating stages.

[0095] In step 603 above, the troubleshooting score refers to the value obtained by weighting and summing the parameter deviation sequence of each target component in each operating stage and the preset weight. It is used to quantify the degree of recovery of the production line operating status by the target troubleshooting program. The lower the troubleshooting score, the closer the production line operating status is to the standard operating status after troubleshooting.

[0096] The preset scoring threshold refers to a pre-set score limit used to distinguish between successful and unsuccessful troubleshooting. A troubleshooting score below or equal to the preset scoring threshold is considered successful, while a troubleshooting score above the preset scoring threshold is considered unsuccessful. The preset scoring threshold can be set according to the fault type and fault parameters of each target component. For example, for training tasks with many fault types or large deviations in fault parameters, the preset scoring threshold can be set higher to allow for a certain degree of parameter deviation; for training tasks with a single fault type or small deviations in fault parameters, the preset scoring threshold can be set lower to require higher troubleshooting accuracy. This application embodiment does not limit the numerical value of the preset scoring threshold.

[0097] The troubleshooting result refers to the conclusion drawn from the comparison and evaluation of whether the troubleshooting of the target student was successful, and the corresponding troubleshooting score.

[0098] In this embodiment of the application, for each target troubleshooting procedure, the average parameter deviation value of the conveyor belt is obtained by taking the absolute value of all parameter deviation values ​​in the parameter deviation sequence of the conveyor belt during the transportation stage; the average parameter deviation value of the gripping device is obtained by taking the absolute value of all parameter deviation values ​​in the parameter deviation sequence of the gripping stage; and the average parameter deviation value of the sorting mechanism is obtained by taking the absolute value of all parameter deviation values ​​in the parameter deviation sequence of the sorting stage.

[0099] Based on the fault types and fault parameters of each target component corresponding to the current teaching and training task, corresponding preset weights are set for the conveyor belt, gripping device, and sorting mechanism. These preset weights can be set according to whether each target component has been set with a fault and the degree of deviation of the fault parameters. The preset weight of the target component with a fault is higher than that of the target component without a fault. The target component with a greater degree of deviation of the fault parameters has a higher preset weight. For example, in the current training task, the conveyor belt is set with a fault that reduces the upper limit of the running speed, and the gripping device is set with a fault that delays the signal transmission. In this case, the preset weights of the conveyor belt and the gripping device are higher than the preset weights of the sorting mechanism.

[0100] The average parameter deviation values ​​of the conveyor belt, the gripping device, and the sorting mechanism corresponding to the current target troubleshooting procedure are weighted and summed according to their respective preset weights to obtain the troubleshooting score of the target troubleshooting procedure. Each troubleshooting score is compared with a preset scoring threshold to obtain the troubleshooting operation result of the target student. For example, if the troubleshooting score corresponding to a target student's generated target troubleshooting procedure is lower than or equal to the preset scoring threshold, the target student is deemed to have successfully troubleshooted the problem; if the troubleshooting score is higher than the preset scoring threshold, the target student is deemed to have failed to troubleshoot the problem.

[0101] The troubleshooting results of each target student are combined with the corresponding number of modifications and troubleshooting time to generate training troubleshooting data for each target student. This data is then sent to the teacher's terminal, which displays the data in a list format, including each student's troubleshooting success or failure status, troubleshooting score, number of modifications, and troubleshooting time. This allows the teacher to view and evaluate the overall performance of the training.

[0102] The troubleshooting assessment in this application embodiment can focus on examining the target student's repair effect on the core fault. At the same time, it integrates the troubleshooting operation results and operation process data to generate training troubleshooting data and automatically sends it to the teacher's terminal, providing the teacher with an objective and traceable teaching assessment basis, avoiding the traditional method in which the teacher needs to check the operation of each target student one by one.

[0103] Figure 2 This application provides a schematic diagram of the structure of a data processing system based on digital twin intelligent production line training, representing a specific implementation of the system. (Refer to...) Figure 2 The system may include: The acquisition module 21 is used to acquire the multi-source dataset of the intelligent production line training box in the current teaching and training process, and the student operation dataset of multiple historical target students in the historical training process. The adjustment module 22 is used to generate a position compensation coefficient based on the physical position of each target component in the multi-source dataset and the corresponding virtual expected position in the first digital twin model of the intelligent production line training box, so as to adjust the first digital twin model and obtain the second digital twin model. The simulation module 23 is used to acquire the baseline behavior data between each target component in the second digital twin model under the preset standard control program, perform fault simulation processing on the second digital twin model, and obtain the third digital twin model. The determination module 24 is used to determine the target operation mode of the student terminal of each target student based on the student operation dataset, so as to obtain the target troubleshooting program, number of modifications and troubleshooting time corresponding to each target student. The update module 25 is used to obtain the actual operating data of each target component under each target troubleshooting procedure, and update the third digital twin model according to each actual operating data to obtain multiple corresponding update status data. The generation module 26 is used to compare each updated status data with the baseline behavior data to obtain the troubleshooting operation results corresponding to each target student. Combined with the corresponding number of modifications and troubleshooting time, the module generates the training troubleshooting data for each target student and sends the training troubleshooting data to the teacher terminal.

[0104] This application provides a data processing system based on digital twin intelligent production line training to implement the aforementioned data processing method based on digital twin intelligent production line training. Therefore, the specific implementation of the data processing system based on digital twin intelligent production line training can be found in the embodiment section of the data processing method based on digital twin intelligent production line training described above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0105] This application also provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement the data processing method based on digital twin intelligent production line training as described above.

[0106] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a computer, implements any of the above-described data processing methods for intelligent production line training based on digital twins.

[0107] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0108] The embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described embodiments of a data processing method for training a digital twin intelligent production line.

[0109] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0110] Obviously, those skilled in the art should understand that the various units or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps into a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0111] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A data processing method based on digital twin intelligent production line training, characterized in that, include: Obtain multi-source datasets of the intelligent production line training box during the current teaching and training process, as well as student operation datasets of multiple historical target students during the historical training process; Based on the physical location of each target component in the multi-source dataset, and combined with the corresponding virtual expected location in the first digital twin model of the intelligent production line training box, a location compensation coefficient is generated to adjust the first digital twin model and obtain the second digital twin model. Acquire baseline behavior data between target components in the second digital twin model under a preset standard control program, perform fault simulation processing on the second digital twin model, and obtain a third digital twin model; Based on the student operation dataset, determine the target operation mode of the student terminal for each target student, so as to obtain the target troubleshooting program, number of modifications, and troubleshooting time for each target student. Obtain the actual operating data of each target component under each target troubleshooting procedure, and update the third digital twin model according to each actual operating data to obtain multiple corresponding update status data; Each updated status data is compared with the baseline behavior data to obtain the troubleshooting operation results for each target student. Combined with the corresponding number of modifications and troubleshooting time, the training troubleshooting data for each target student is generated and sent to the teacher terminal.

2. The method according to claim 1, characterized in that, Based on the physical location of each target component in the multi-source dataset, and combined with the corresponding virtual expected location in the first digital twin model of the intelligent production line training box, a location compensation coefficient is generated, including: Based on the physical location of each target component in the multi-source dataset and the corresponding virtual expected location in the first digital twin model of the intelligent production line training box, the difference between the physical location of each target component and the corresponding virtual expected location at the same acquisition time is calculated to obtain the position deviation sequence of each target component at different acquisition times. Based on the distribution trend of each position deviation value in each position deviation sequence with the acquisition time, the corresponding fitting curve is determined, and the slope and intercept of each fitting curve are extracted as the position compensation coefficient of the corresponding target component.

3. The method according to claim 1, characterized in that... The second digital twin model is subjected to fault simulation processing to obtain a third digital twin model, including: Extract the fault types and fault parameters of each target component corresponding to the current teaching and training process task from the pre-built fault simulation database; Based on the fault type and fault parameters of each target component, the operating parameters and / or connection relationships of the corresponding target components in the second digital twin model are adjusted to obtain the third digital twin model.

4. The method according to claim 1, characterized in that, Based on the student operation datasets, determine the target operation mode of each target student's terminal, including: Extract the operation accuracy rate, number of operation deviations and task completion time of the corresponding historical target students in the history teaching and training process from each student operation dataset, and calculate the proficiency score of each historical target student in combination with the corresponding preset influence weights. Each target student is matched with each historical target student. The proficiency score of the historical target student corresponding to the target student whose identity is successfully matched is used as the corresponding proficiency score. The proficiency score of the target student whose identity is not matched is set as the initial proficiency score. Based on the proficiency scores of each target student, the target operation mode of the student terminal for each target student is set.

5. The method according to claim 1, characterized in that, Obtain the target troubleshooting program, number of modifications, and troubleshooting time for each target student, including: Extract the standard troubleshooting basic program corresponding to the abnormal state of the third digital twin model from the pre-set fault program library, as well as the pre-stored fault point information and fault description associated with the standard troubleshooting basic program. The standard troubleshooting basic program and the corresponding pre-stored fault point information and fault description are loaded into each student terminal. Based on the target operation mode of the student terminal, the target troubleshooting program, the number of modifications, and the troubleshooting time corresponding to the modification operation of the standard troubleshooting basic program by each target student on the corresponding student terminal are obtained.

6. The method according to claim 1, characterized in that, Based on the actual operational data, the third digital twin model is updated to obtain multiple corresponding update status data, including: The actual physical positions of each target component at different acquisition times are extracted from the actual operation data, and the deviation values ​​between the actual physical positions of each target component at each acquisition time and the corresponding virtual positions in the third digital twin model are calculated. The sum of the squares of the deviation values ​​of each target component at each acquisition time is used as the optimization objective to solve for the position compensation of each target component. Based on the position compensation amount of each target component, the virtual position of each target component in the third digital twin model is updated to obtain the fourth digital twin model, and multiple update status data corresponding to the running of each target troubleshooting program in the fourth digital twin model are obtained.

7. The method according to claim 1, characterized in that, Each updated status data is compared with the baseline behavior data to obtain the troubleshooting operation results for each target student, including: Extract the simulated operating parameter sequence of each target component under different operating stages in each target troubleshooting procedure from each updated status data, and extract the standard operating parameter sequence of each target component under different operating stages in the preset standard control procedure from the benchmark behavior data; The difference between the simulated operating parameter sequence and the standard operating parameter sequence of the same target component at the same operating stage is calculated to obtain the parameter deviation sequence of each target component at different operating stages; Based on the multiple parameter deviation sequences and corresponding preset weights of each target component under each target troubleshooting procedure, the troubleshooting score corresponding to each target student is calculated. Combined with the preset score threshold, the troubleshooting operation result corresponding to each target student is generated. The preset weights are set based on the fault type and fault parameters of each target component corresponding to the current teaching and training process task.

8. A data processing system based on digital twin intelligent production line training, characterized in that, include: The acquisition module is used to acquire multi-source datasets of the intelligent production line training box during the current teaching and training process, as well as student operation datasets of multiple historical target students during the historical training process. The adjustment module is used to generate a position compensation coefficient based on the physical position of each target component in the multi-source dataset and the corresponding virtual expected position in the first digital twin model of the intelligent production line training box, so as to adjust the first digital twin model and obtain the second digital twin model. The simulation module is used to acquire the baseline behavior data between each target component in the second digital twin model under a preset standard control program, perform fault simulation processing on the second digital twin model, and obtain a third digital twin model. The determination module is used to determine the target operation mode of each target student's student terminal based on the student operation dataset, so as to obtain the target troubleshooting program, number of modifications and troubleshooting time corresponding to each target student. The update module is used to obtain the actual operating data of each target component under each target troubleshooting procedure, and update the third digital twin model according to each actual operating data to obtain multiple corresponding update status data. The generation module is used to compare each updated status data with the baseline behavior data to obtain the troubleshooting operation results corresponding to each target student. Combining the corresponding number of modifications and troubleshooting time, the module generates the training troubleshooting data for each target student and sends the training troubleshooting data to the teacher terminal.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the data processing method based on digital twin intelligent production line training as described in any one of claims 1-7.

10. A computing device, characterized in that, The computing device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the at least one processor to perform a data processing method based on digital twin intelligent production line training as described in any one of claims 1-7.

Images