An intelligent factory production and training system based on digital twins

Abstract

The present invention relates to the technical field of integrated manufacturing systems, and specifically to an intelligent factory production and training system based on digital twins. The system includes: a dynamic production scheduling module, which extracts task priority, resource allocation status and task execution time based on production task information, compares task priority and resource utilization rate, and calculates time difference and resource occupancy rate. In the present invention, by dynamically adjusting the resource allocation order and task priority, analyzing production task information in real time, identifying key turning points and adjusting resource allocation, production scheduling is made more flexible and efficient. The coordinated optimization of task switching reduces time waste during task switching and ensures smooth task connection by accurately calculating time differences. The real-time feedback mechanism timely adjusts the execution path and task priority by comparing execution data and predetermined parameters, thereby improving the accuracy of task execution and resource utilization.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated manufacturing systems, and in particular to an intelligent factory production and training system based on digital twins. Background Art

[0002] Integrated Manufacturing System (IMS) refers to a manufacturing system that uses information technology to highly integrate and optimize all aspects of the manufacturing process, from design, production, management to service. The goal of IMS is to achieve the optimal configuration and coordination of manufacturing resources, making the entire manufacturing process more efficient, flexible, precise and intelligent. This technical field includes the integrated application of multiple advanced technologies such as automation control, computer integrated manufacturing (CIM), Internet of Things, big data analysis, artificial intelligence, etc. Integrated manufacturing systems can not only improve production efficiency, but also significantly reduce production costs and shorten product development cycles, while improving product quality and market response speed.

[0003] The intelligent factory production and training system belongs to the field of integrated manufacturing system technology. It aims to optimize and manage the factory's production process through intelligent means and provide training functions. The purpose is to realize the automation and intelligent management of factory production, improve production efficiency, and provide operators with an efficient training environment through a combination of simulation and actual operation, helping them to better master production equipment and processes, ensure the smooth progress of the production process and improve overall productivity.

[0004] However, the existing technology has certain shortcomings in realizing the integration and coordination of manufacturing resources. First, the task priority and resource allocation in the existing technology are relatively fixed, which makes it difficult to cope with the dynamically changing production environment, resulting in low resource utilization and difficulty in adapting to fluctuations in production demand. Second, the task switching process lacks precise time management and coordination, which is prone to poor task connection and time waste, affecting overall production efficiency. In addition, the feedback adjustment mechanism in the existing technology is relatively lagging and cannot respond to changes in the production process in a timely manner, resulting in execution deviations that are difficult to correct in a timely manner, affecting product quality and production efficiency. Finally, the training systems in the existing technology are mostly static training, which is difficult to combine with the dynamic changes in actual production, the training effect is limited, and it is difficult for operators to quickly adapt to changes in the production environment. These shortcomings make it difficult for existing technologies to fully realize their potential in efficient production and intelligent management. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose an intelligent factory production and training system based on digital twins.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solution: A digital twin-based intelligent factory production and training system includes:

[0007] The dynamic production scheduling module extracts task priority, resource allocation status and task execution time based on production task information, compares task priority and resource utilization rate, calculates time difference and resource occupancy rate, identifies key turning points and adjusts the resource allocation order to generate a dynamic scheduling plan;

[0008] The task switching coordination module extracts the task connection sequence and switching time nodes based on the dynamic scheduling scheme, analyzes the time distribution of the connection points and the task execution sequence, calculates the time difference during the switching period and adjusts the time parameters and the connection sequence to generate a coordinated switching scheme;

[0009] A real-time feedback adjustment module, based on the coordinated switching scheme, extracts execution data and predetermined parameters, compares the execution data with the predetermined parameters, calculates deviation values, adjusts execution paths and task priorities, and generates feedback adjustment results;

[0010] The production task optimization module analyzes the task execution time and resource consumption data based on the feedback adjustment results, calculates the difference between time and resource allocation, readjusts the task priority and resource allocation order, optimizes the matching degree between tasks and resources, and generates a production optimization plan;

[0011] The training simulation module extracts the change trend of operating parameters and the task execution path based on the production optimization plan, simulates the optimized production process, analyzes the changes in operating parameters, adjusts the training plan, and generates a training plan.

[0012] As a further solution of the present invention, the dynamic scheduling scheme includes adjusted task priorities, optimized resource allocation strategies and updated time node arrangements; the coordinated switching scheme includes adjusted switching times, optimized task connection sequences and updated time control parameters; the feedback adjustment results include revised execution paths, reallocated task priorities and adjusted deviation control parameters; the production optimization scheme includes optimized resource allocation strategies, adjusted task priorities and improved matching parameters; and the training scheme includes updated operator training plans, simulated production scenarios and adjusted parameter change guidance.

[0013] As a further solution of the present invention, the dynamic production scheduling module includes:

[0014] The task priority submodule, based on production task information, uses bifurcation theory and singular perturbation method to extract task priority and related data, organize and classify task data, analyze the importance and urgency of tasks, prioritize them, and compare the relevance between tasks to generate priority ranking results;

[0015] The resource allocation submodule extracts the resource allocation status and related data based on the priority sorting result, analyzes the resource allocation and usage, calculates the current resource usage rate, adjusts the resource allocation order, and reallocates resources according to the priority to generate a resource allocation adjustment result;

[0016] The time calculation submodule extracts the task execution time and related data based on the resource allocation adjustment results, segments and compares the task time, calculates the time difference between tasks, identifies the key time nodes in the task execution process, adjusts the time arrangement, and generates a dynamic scheduling plan.

[0017] As a further solution of the present invention, the task switching coordination module includes:

[0018] The task connection submodule extracts the task connection sequence and related data based on the dynamic scheduling scheme, analyzes the dependency relationship between multiple tasks, adjusts the task connection sequence, and generates a preliminary connection sequence;

[0019] The time node submodule extracts the switching time nodes and related data based on the preliminary connection sequence, calculates and adjusts the time nodes of task switching, optimizes the arrangement of the time nodes, and generates adjusted time nodes;

[0020] The switching sequence submodule extracts the time distribution and associated data of the connection points based on the adjusted time nodes, optimizes the task switching sequence, adjusts the sequence and time parameters of the task connection, and generates a coordinated switching plan.

[0021] As a further solution of the present invention, the real-time feedback adjustment module includes:

[0022] An execution data extraction submodule extracts execution data and associated parameters based on the coordinated switching scheme, monitors and records the data during task execution in real time, compares the execution data with the predetermined parameters, analyzes the differences and summarizes the results, and generates deviation data;

[0023] The deviation calculation submodule extracts the deviation value and related parameters based on the deviation data, calculates the error amplitude during task execution, classifies and sorts the deviation values, determines the necessity of adjustment, and formulates an adjustment plan according to the deviation size to generate an adjustment demand result;

[0024] The path adjustment submodule extracts the execution path and task priority based on the adjustment requirement result, analyzes and adjusts the execution path, reorders the task priority, adjusts the execution order and path of the task, and generates feedback adjustment results.

[0025] As a further solution of the present invention, the production task optimization module includes:

[0026] The time analysis submodule extracts the task execution time and related data based on the feedback adjustment result, analyzes and compares the task execution time, calculates the difference value of the time allocation of multiple tasks, and generates the time difference analysis result;

[0027] The resource allocation optimization submodule extracts resource consumption data and related information based on the time difference analysis results, analyzes and optimizes the resource allocation situation, reallocates resources, calculates the efficiency after resource allocation, and generates resource optimization results;

[0028] The matching degree adjustment submodule extracts task and resource matching data and related information based on the resource optimization result, analyzes the matching degree of tasks and resources, adjusts the allocation sequence of tasks and resources, and generates a production optimization plan.

[0029] As a further solution of the present invention, the training simulation module includes:

[0030] The parameter extraction submodule, based on the production optimization scheme, extracts the change trend of the operating parameters and the associated data through Laplace transform and transient behavior analysis, analyzes the change pattern of the operating parameters, classifies and sorts the parameter change trend, and generates parameter classification records;

[0031] The production simulation submodule extracts the task execution path and related information based on the parameter classification records, simulates the optimized production process, analyzes the data changes during the simulation process, records and summarizes the simulation data, and generates simulation analysis results;

[0032] The training adjustment submodule extracts the changes in operation parameters and task execution paths based on the simulation analysis results, analyzes and adjusts the operation training plan, formulates new training content, updates the operation guidance of the training, and generates a training plan.

[0033] As a further solution of the present invention, the bifurcation theory and singular perturbation method are according to the formula:

[0034]

[0035] Calculate the task priority sorting results, where: is the task priority parameter, is the system disturbance parameter, is the task importance parameter, is the urgency parameter of the task, A comprehensive function that describes the stability, importance, and urgency of the mission system is a function of task relevance, represents the comprehensive rate at which task priorities change with system disturbances, and is the weight coefficient, which indicates the influence of different parameters in the calculation.

[0036] As a further solution of the present invention, the Laplace transform and transient behavior analysis are performed according to the formula:

[0037]

[0038] Calculate the trend value of the operating parameter change, where: For the improved representation of the operating parameters in the frequency domain, is the function of the operating parameter in the time domain, is the complex frequency parameter, is the time variable, is the weight coefficient, which indicates the reference amplitude adjustment of the operating parameter. is the frequency modulation parameter, which indicates the frequency of change of the operating parameter over time. is a phase offset parameter, indicating the initial phase offset of the operating parameter.

[0039] Compared with the prior art, the advantages and positive effects of the present invention are:

[0040] In the present invention, by dynamically adjusting the resource allocation order and task priority, analyzing the production task information in real time, identifying the key turning points and adjusting the resource allocation, the production scheduling is made more flexible and efficient. The coordinated optimization of task switching reduces the time waste during task switching by accurately calculating the time difference, ensuring smooth task connection. The real-time feedback mechanism timely adjusts the execution path and task priority by comparing the execution data and the predetermined parameters, thereby improving the accuracy of task execution and resource utilization. In terms of production task optimization, by analyzing the difference between task execution time and resource consumption, the resource allocation is further optimized, making the production process more efficient. At the same time, the training simulation function simulates the optimized production process and adjusts the training plan, thereby improving the training effect and operation proficiency of the operators. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0043] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0044] Example: See Figure 1 The present invention provides a technical solution: an intelligent factory production and training system based on digital twins includes:

[0045] The dynamic production scheduling module extracts task priority, resource allocation status and task execution time based on production task information, compares task priority and resource utilization rate, calculates time difference and resource occupancy rate, identifies key turning points and adjusts the resource allocation order to generate a dynamic scheduling plan;

[0046] The task switching coordination module extracts the task connection sequence and switching time nodes based on the dynamic scheduling scheme, analyzes the time distribution of the connection points and the task execution sequence, calculates the time difference during the switching period and adjusts the time parameters and connection sequence to generate a coordinated switching scheme;

[0047] The real-time feedback adjustment module extracts execution data and predetermined parameters based on the coordinated switching scheme, compares the execution data with the predetermined parameters, calculates the deviation value, adjusts the execution path and task priority, and generates feedback adjustment results;

[0048] The production task optimization module analyzes the task execution time and resource consumption data based on the feedback adjustment results, calculates the difference between time and resource allocation, readjusts the task priority and resource allocation order, optimizes the matching degree between tasks and resources, and generates a production optimization plan;

[0049] The training simulation module extracts the change trend of operating parameters and the task execution path based on the production optimization plan, simulates the optimized production process, analyzes the changes in operating parameters, adjusts the training plan, and generates a training plan.

[0050] The dynamic scheduling plan includes adjusted task priorities, optimized resource allocation strategies and updated time node arrangements; the coordinated switching plan includes adjusted switching times, optimized task connection sequences and updated time control parameters; the feedback adjustment results include revised execution paths, reallocated task priorities and adjusted deviation control parameters; the production optimization plan includes optimized resource allocation strategies, adjusted task priorities and improved matching parameters; the training plan includes updated operator training plans, simulated production scenarios and adjusted parameter change guidance.

[0051] In the dynamic production scheduling module, by extracting and analyzing the production task information, extracting the task priority, resource allocation status and task execution time, the system uses a linear programming algorithm to analyze the relationship between task priority and resource utilization. After comparing the two, the time series analysis method is used to calculate the difference in task execution time, and combined with the resource occupancy rate, the critical path method is used to identify the key turning points in the task execution process. At this time, the resource allocation order will be adjusted according to the results of the critical path method, and the greedy algorithm will be used to optimize the resource allocation strategy. In this way, the rationality of resource allocation can be ensured, and the time nodes can be dynamically adjusted through the regression analysis of the time series. The final generated dynamic scheduling plan includes the adjusted task priority, the optimized resource allocation strategy and the updated time node arrangement.

[0052] In the task switching coordination module, through the analysis and execution of the dynamic scheduling scheme, the task connection sequence and switching time nodes are extracted, the time distribution of the connection points is analyzed using the time window algorithm, and the time difference during the switching period is calculated by combining the task priority sorting and dynamic scheduling information with the backtracking algorithm. The system will adjust the time parameters and task connection sequence through the dynamic programming method to ensure seamless connection between tasks while minimizing the switching time. The time window algorithm helps to identify the best time point for task switching, while the dynamic programming method optimizes the task connection sequence, ensuring the efficiency and smoothness of the task switching process, and finally generates a coordinated switching plan, which includes the adjusted switching time, the optimized task connection sequence, and the updated time control parameters. This plan effectively improves the coordination and efficiency of task switching, reduces the waste of resources and time delays in the switching process, and the generated files help maintain the continuity and stability of production in actual production.

[0053] In the real-time feedback adjustment module, by extracting and comparing the execution data and predetermined parameters of the coordinated switching scheme, the system uses an error analysis algorithm to calculate the deviation value between the actual execution data and the predetermined parameters, and uses a fuzzy control algorithm to adjust the execution path and task priority through a feedback adjustment mechanism based on this deviation value. When processing feedback adjustments, the fuzzy control algorithm can dynamically adjust the execution path according to the size of the deviation to ensure the efficiency and accuracy of task execution during the production process. At the same time, the priority of the task is adjusted using the deviation sensitivity analysis method to ensure that the task can be executed as planned in the adjusted production environment. The feedback adjustment results finally generated include the revised execution path, the reallocated task priority, and the adjusted deviation control parameters. The result file is used to guide the task adjustment in real time in production to ensure that the production process can respond to changes in a timely manner, thereby improving the robustness and flexibility of the overall production system.

[0054] In the production task optimization module, by analyzing the feedback adjustment results, the system uses a multi-objective optimization algorithm to conduct a detailed analysis of the task execution time and resource consumption data, and combines the Lagrange multiplier method to calculate the difference between time and resource allocation, and then optimizes the task priority and resource allocation order through the genetic algorithm. In this process, the multi-objective optimization algorithm is used to balance the conflict between task execution time and resource consumption, the Lagrange multiplier method is used to find the optimal solution for time and resource allocation, and the genetic algorithm finds the optimal task and resource matching combination in a large-scale search space. The final production optimization solution includes an optimized resource allocation strategy, adjusted task priority, and improved matching parameters.

[0055] In the training simulation module, through the extraction and analysis of the production optimization plan, the system first extracts the change trend of the operating parameters and the task execution path, and uses the digital twin technology based on physical simulation to simulate the optimized production process. The system uses the Bayesian network to analyze the changes in the operating parameters to identify the causal relationship and potential risks between the parameters, and then uses the dynamic system modeling technology to adjust the training plan to ensure that the training content is consistent with the optimized production process. The Bayesian network algorithm provides probabilistic inference of parameter changes to help identify key operating parameters, and the dynamic system modeling technology matches the parameters with the actual production situation to ensure the scientific nature and pertinence of the training plan. The final generated training plan includes an updated operator training plan, simulated production scenarios, and adjusted parameter change guidance.

[0056] In this embodiment, the dynamic production scheduling module includes:

[0057] The task priority submodule, based on production task information, uses bifurcation theory and singular perturbation method to extract task priority and related data, organize and classify task data, analyze the importance and urgency of tasks, prioritize them, and compare the relevance between tasks to generate priority ranking results;

[0058] The resource allocation submodule extracts the resource allocation status and related data based on the priority sorting results, analyzes the resource allocation and usage, calculates the current resource utilization rate, adjusts the resource allocation order, and reallocates resources according to the priority to generate the resource allocation adjustment results;

[0059] The time calculation submodule extracts the task execution time and related data based on the resource allocation adjustment results, segments and compares the task time, calculates the time difference between tasks, identifies the key time nodes in the task execution process, adjusts the time arrangement, and generates a dynamic scheduling plan.

[0060] In the task priority submodule, by collecting and processing production task information, bifurcation theory is used to identify key points of task priority. Specifically, bifurcation theory is used to analyze the dynamic changes in production task scheduling to identify key turning points that cause drastic changes in priority within the system. Turning points are manifested as sudden urgency of tasks or temporary unavailability of resources. The system performs multi-dimensional analysis of the execution conditions, resource requirements and time constraints of each task, and formats the data into a multi-dimensional matrix, including task name, required resources, scheduled time, urgency, etc. Subsequently, the tasks are perturbed in a small range through perturbation parameters to simulate the changes in task response under different conditions, and the perturbation results are used to re-sort the task priorities. The sorting process takes into account the correlation between tasks, especially the dependencies between tasks and potential resource conflicts. Through a comprehensive comparison of the priorities of each task, the task priority sorting results are finally generated.

[0061] In the resource allocation submodule, based on the priority ranking results generated by the task priority submodule, the resource allocation status data associated with each task is first extracted, including the current number of available resources, resource occupancy, and expected availability of future resources. The data is organized into a resource status matrix, which contains information such as resource type, current occupancy ratio, and remaining availability. Subsequently, the resource utilization rate of each task is calculated by cross-analyzing the resource allocation status and task priority. Specifically, the system matches the required resources for each task with the available resources and adjusts the resource allocation order according to the task priority. In order to optimize the utilization efficiency of resources, the system also continuously monitors the real-time occupancy of resources through a dynamic resource adjustment mechanism, and automatically adjusts the allocation of resources when resource conflicts or unreasonable utilization are found to ensure that high-priority tasks obtain the necessary resource support. The generated resource allocation adjustment results are output in the form of a resource allocation table, which clearly lists the specific resources allocated to each task and their usage schedule.

[0062] In the time calculation submodule, based on the resource allocation adjustment results generated by the resource allocation submodule, the task execution time and related data, including the planned start time, estimated duration, and resource preparation time, are first extracted to form a task schedule. The system segments the execution time of each task through time series analysis, paying special attention to those tasks that are immediately after or before resource-intensive tasks, and identifies the key time nodes in the execution process by calculating the time differences between tasks. Key time nodes are those locations that have the greatest impact on the entire task chain and determine whether downstream tasks can start on time. After identifying the nodes, the system compares the time nodes with the task priorities and resource allocation, and adjusts the schedule when necessary to reduce potential delay risks and resource conflicts. The adjusted schedule forms the final dynamic scheduling plan, which is output in the form of a Gantt chart or other visual schedule, in which the specific execution time period and required resources of each task are marked.

[0063] In this embodiment, the task switching coordination module includes:

[0064] The task connection submodule extracts the task connection sequence and related data based on the dynamic scheduling scheme, analyzes the dependencies between multiple tasks, adjusts the task connection sequence, and generates a preliminary connection sequence;

[0065] The time node submodule extracts the switching time nodes and related data based on the preliminary connection sequence, calculates and adjusts the time nodes of task switching, optimizes the arrangement of time nodes, and generates adjusted time nodes;

[0066] The switching order submodule extracts the time distribution and related data of the connection points based on the adjusted time nodes, optimizes the task switching order, adjusts the order and time parameters of task connection, and generates a coordinated switching plan.

[0067] In the task connection submodule, by analyzing the dynamic scheduling scheme, the task connection order and related data are first extracted. The data exists in the form of a task dependency matrix and a task priority list. The task dependency matrix is ​​represented in the form of a two-dimensional array, in which each element corresponds to the dependency between two tasks (for example, task A must be completed before task B), and the task priority list contains the priority index of each task. By analyzing the data, the topological sorting algorithm and the critical path analysis method are used to analyze and sort the dependencies between tasks. First, the task dependency matrix is ​​matrix decomposed and weighted to decompose the matrix into multiple independent or dependent chains. Then, the topological sorting is used to determine the priority task order in the acyclic dependency chain, and a preliminary connection order list is generated. This list contains the execution order of each task and its dependent predecessor tasks. At the same time, the critical path analysis is used to identify the most time-consuming path in the task chain, optimize the connection between tasks to reduce the execution time of the overall task chain, and the generated preliminary connection order ensures the logic and order of the dependency between tasks when connecting tasks, and reduces the risk of execution conflicts.

[0068] In the time node submodule, based on the preliminary connection sequence, the task switching time nodes and related data are first extracted. The data exists in the form of task execution time distribution table and task connection schedule. The task execution time distribution table records the start time and end time of each task, while the task connection schedule describes the switching time between tasks. The time data is processed and analyzed using linear regression and time series analysis algorithms. First, the data in the task execution time distribution table is subjected to time series analysis to identify the distribution trend of time nodes and potential bottleneck periods. The linear regression model is used for data fitting and error analysis to predict the delay and early completion during task execution, and the arrangement of switching time nodes is adjusted by optimizing the parameters of the model. Through iterative optimization and adjustment, the time nodes of the connection tasks are fine-tuned forward or backward to maximize the time utilization, and finally an adjusted time node distribution table is generated. This table optimizes the key time nodes in task connection, ensures smooth and efficient switching between tasks, and reduces idle resources and task delays caused by improper time scheduling.

[0069] In the switching order submodule, based on the adjusted time nodes, the time distribution and associated data of the task connection points are extracted. The data includes the task switching schedule and the task execution priority table. The switching schedule shows the actual time distribution of each task connection point, while the priority table provides the priority weight of the task. The data is refined and optimized through the dynamic programming algorithm and constraint satisfaction problem solving technology. First, the dynamic programming algorithm is used to globally optimize the time distribution of the task connection points, and the global optimal task switching order is found in the multi-dimensional time space. By calculating the total execution time and resource occupancy rate for each possible switching sequence, the switching order with the shortest total execution time is selected. At the same time, the constraint satisfaction problem solving technology is used to verify the time constraints in the switching order to ensure that the time of each task switching meets the preset time node, and the time parameters of the connection point are adjusted to meet the minimum time difference and maximize the resource utilization efficiency. The generated coordinated switching scheme lists the optimal switching order and time parameters of all tasks in detail. This scheme effectively reduces the waiting time and resource conflicts in the task switching process, improves the overall task execution efficiency and optimizes the allocation and use of resources.

[0070] In this embodiment, the real-time feedback adjustment module includes:

[0071] The execution data extraction submodule extracts execution data and associated parameters based on the coordinated switching scheme, monitors and records the data during task execution in real time, compares the execution data with the predetermined parameters, analyzes the differences and summarizes the results, and generates deviation data;

[0072] The deviation calculation submodule extracts deviation values ​​and related parameters based on deviation data, calculates the error amplitude during task execution, classifies and sorts deviation values, determines the necessity of adjustment, and formulates adjustment plans based on the deviation size to generate adjustment demand results;

[0073] The path adjustment submodule extracts the execution path and task priority based on the adjustment requirement results, analyzes and adjusts the execution path, reorders the task priority, adjusts the execution order and path of the tasks, and generates feedback adjustment results.

[0074] In the execution data extraction submodule, by coordinating the data acquisition of the switching scheme, the system uses multi-sensor fusion technology to monitor and record various data in the task execution process in real time. The data includes parameters such as execution time, resource consumption, and task status. The data is stored in the database in the form of time series and compared with the predetermined parameters in real time. To ensure the accuracy and timeliness of the data, the system uses the Kalman filter algorithm to preprocess the real-time collected data to filter out noise and abnormal data. Then, the difference algorithm is used to calculate the difference between the execution data and the predetermined parameters, generate difference data, and summarize and summarize the differences through cluster analysis to identify the deviation patterns that occur in the execution of various tasks. Cluster analysis further helps the system to classify similar types of differences into the same category, and finally generates a deviation data file, which records in detail the specific differences from the predetermined parameters during the task execution process, including the amplitude, frequency, and type of the difference. The role of the data file is to provide basic data support for subsequent deviation calculation and path adjustment, ensuring that abnormal situations in task execution can be captured in real time during system operation, and timely correction and adjustment can be made.

[0075] In the deviation calculation submodule, by extracting the deviation data generated by the execution data extraction submodule, the system uses a regression analysis algorithm to calculate and analyze the deviation value in detail, including quantitative evaluation of the error amplitude, frequency and duration during the task execution process. The regression analysis method can not only identify the specific value of the deviation, but also predict the possible development trend of the deviation in the future execution process. According to the different deviation values, the system divides the deviation into three categories: slight, moderate and severe, and uses fuzzy logic to classify and organize the deviations. The fuzzy logic method makes the classification process more flexible and can handle fuzzy boundary conditions caused by data uncertainty. For different categories of deviations, the system uses the analytic hierarchy process (AHP) to determine whether it is necessary to adjust the task execution path and priority. The analytic hierarchy process assigns weights to various influencing factors by constructing a judgment matrix, thereby scientifically evaluating the necessity of adjustment. Based on this analysis result, the system generates an adjustment requirement result file, which contains the specific items, priorities and adjustment plans that need to be adjusted during the task execution process.

[0076] In the path adjustment submodule, by reading the adjustment requirement result file generated by the deviation calculation submodule, the system first extracts the current execution path and task priority, and uses graph theory algorithms to analyze and optimize the existing path. Specifically, the system uses Dijkstra's Algorithm to calculate the shortest path of the task execution path, and reorders the priority of the task in combination with the dynamic programming method. For the task order that needs to be adjusted, the system rearranges the execution order of the tasks through the topological sorting algorithm to ensure that the adjusted path achieves the optimal configuration in terms of time and resources. Topological sorting determines the dependency relationship between tasks by constructing a directed acyclic graph (DAG) of tasks, avoiding the circular dependency problem after the task order is adjusted. During the adjustment process, the system also uses a genetic algorithm to further optimize the adjusted task path. The genetic algorithm simulates the natural selection process and continuously iterates to generate a better path plan. The feedback adjustment result finally generated includes the adjusted execution path, the reordered task priority, and the optimized execution order. This result file can effectively reduce the total time of task execution, improve resource utilization, and ensure the coordination and consistency between tasks in the actual production process.

[0077] In this embodiment, the production task optimization module includes:

[0078] The time analysis submodule extracts the task execution time and related data based on the feedback adjustment results, analyzes and compares the task execution time, calculates the difference value of multiple task time allocations, and generates time difference analysis results;

[0079] The resource allocation optimization submodule extracts resource consumption data and related information based on the time difference analysis results, analyzes and optimizes the resource allocation situation, reallocates resources, calculates the efficiency after resource allocation, and generates resource optimization results;

[0080] The matching degree adjustment submodule extracts task and resource matching data and related information based on resource optimization results, analyzes the matching degree between tasks and resources, adjusts the allocation order of tasks and resources, and generates a production optimization plan.

[0081] In the time analysis submodule, based on the feedback adjustment results, the execution time data of each task is first extracted. The data includes the planned start time, actual start time, expected completion time, actual completion time, etc. of the task, and the data is stored in the form of time series. Then, the task time is analyzed and compared item by item. The system calculates the difference between the actual execution time and the planned time of each task to identify which tasks have time delays or are completed ahead of schedule. The system quantifies the time difference values ​​between tasks and represents them in the form of a matrix. The matrix includes the time deviation values ​​between tasks, the dependency of tasks on time, and the time difference on the critical path. In the process of difference calculation, the system uses cumulative deviation calculation to associate the time difference of each task with the time arrangement of its subsequent tasks to determine which time differences will have a significant impact on the entire production process. Through in-depth analysis of the differences, the system generates time difference analysis results, which include a detailed time deviation list, time adjustment suggestions for key tasks, and time node information that causes process bottlenecks. The time difference analysis results provide data support for subsequent resource allocation optimization and task adjustment, helping to identify time problem areas that need to be prioritized to reduce time waste and task conflicts in production.

[0082] In the resource allocation optimization submodule, the resource consumption data associated with each task is extracted based on the time difference analysis results. The data includes resource type, resource usage time period, actual resource consumption, planned resource consumption, etc. The data format is presented in the form of resource utilization matrix. The system combines the resource data with the time difference analysis results, and optimizes the consumption of each resource in different tasks through resource-time two-dimensional analysis to find out the imbalance or waste points in resource allocation. Then, the system reallocates resources and adjusts the allocation order of resources between tasks. The specific operations include reordering the use priority of resources, optimizing the use period of resources to reduce resource conflicts during peak periods, and reconfiguring resources to ensure that high-priority tasks obtain the required resources. The system also calculates the efficiency after resource allocation. The efficiency value is based on the comprehensive evaluation indicators of task completion time, resource consumption and resource utilization. The generated resource optimization results include the adjusted resource allocation table, resource utilization report and optimized task-resource correspondence.

[0083] In the matching degree adjustment submodule, based on the resource optimization results, the data of task and resource matching is first extracted. The data includes the task resource requirement list, the available time window of resources, the priority of task execution, the actual availability of resources, etc. The data is organized in the form of a task-resource matching matrix. The system calculates the matching degree between tasks and resources by analyzing the matching data. The matching degree is calculated based on the degree of task dependence on resources, the overlap between resource availability and task execution time, and the fit between task priority and resource allocation. The system identifies the task and resource combinations with low matching degree and adjusts the allocation order of the combination. The specific operations include rearranging the allocation order of resources, adjusting the execution order of tasks to better match the availability of resources, or changing the resource type used by the task when necessary. The system generates a production optimization plan based on the adjusted matching degree, which includes the optimal matching list of tasks and resources, the reordering suggestions for task execution, and the adjustment list of resource allocation. The production optimization plan ensures that the matching between tasks and resources is more reasonable during the execution process, reduces resource waste and conflicts in task execution, and makes the entire production process smoother and more efficient.

[0084] In this embodiment, the training simulation module includes:

[0085] The parameter extraction submodule, based on the production optimization scheme, extracts the change trend of operating parameters and related data through Laplace transform and transient behavior analysis, analyzes the change pattern of operating parameters, classifies and organizes the parameter change trend, and generates parameter classification records;

[0086] The production simulation submodule extracts the task execution path and related information based on parameter classification records, simulates the optimized production process, analyzes data changes during the simulation process, records and summarizes simulation data, and generates simulation analysis results;

[0087] The training adjustment submodule extracts the changes in operation parameters and task execution paths based on the simulation analysis results, analyzes and adjusts the operation training plan, formulates new training content, updates the operation instructions for training, and generates training plans.

[0088] In the parameter extraction submodule, the Laplace transform and transient behavior analysis are used to deeply analyze the change trend of the operating parameters in the production optimization scheme. First, the original operating parameter data is extracted. The data records the parameter changes in each operation step in the form of a time series, including temperature, pressure, speed and other parameters. The data set of each parameter consists of a time point and a value pair. The Laplace transform is used to convert the data in the time domain to the frequency domain, thereby revealing the high-frequency fluctuations and transient change characteristics hidden in the time series. By performing Fourier transform on the converted frequency domain data, the key frequency components and transient behavior patterns that affect the stability of the parameters are further identified. Based on the analysis results, the k-means clustering algorithm is used to classify and organize the parameter change trends. The clustering process is based on the spectral characteristics and classifies the parameters with similar transient behaviors into the same category. Each category represents a specific parameter change pattern. The generated parameter classification record lists in detail the characteristic frequency, transient behavior pattern and performance of each parameter category in different production scenarios.

[0089] In the production simulation submodule, based on the parameter classification record, the task execution path and its associated information are first extracted from it. The information includes the flowchart of the production task, the execution conditions of each task node, and the operation instructions corresponding to the parameter classification. When simulating the optimized production process, the simulation data is stored in the form of high-precision time steps, recording the parameter values ​​and task status at each moment. The entire production process is dynamically simulated by the method based on Discrete Event Simulation (DES). The simulation process adopts an event-driven model. The simulator adjusts the simulation path according to the current task status and the classification results of the operation parameters in each time step, and evaluates the completion time and resource occupancy of each task node. As the simulation proceeds, the system continuously collects and summarizes the data generated during the simulation process, including key indicators such as parameter fluctuations, task delays, and resource conflicts.

[0090] In the training adjustment submodule, based on the simulation analysis results, key data in the operation parameter changes and task execution paths are extracted. The data includes the actual execution time of each task node, resource utilization rate, and the related operation parameter change pattern. The operation training plan is carefully analyzed and adjusted. First, according to the bottlenecks and abnormal situations found in the simulation analysis, the focus of the training content is adjusted to ensure that operators can cope with various emergencies in production. The task sequence and operation parameters in the training plan are rearranged using the dynamic programming method. The optimized training content not only covers the standard operating procedures, but also strengthens the training of error-prone links. At the same time, special training modules for different operation parameter change patterns are introduced. The training program file finally generated lists the content, schedule and expected results of each training module in detail, and updates the operation instruction manual to ensure that the training process can better simulate the complexity of the actual production environment and improve the operator's coping ability and production efficiency.

[0091] In this embodiment, the bifurcation theory and singular perturbation method are based on the formula:

[0092]

[0093] Calculate the task priority sorting results, where: is the task priority parameter, is the system disturbance parameter, is the task importance parameter, is the urgency parameter of the task, A comprehensive function that describes the stability, importance, and urgency of the mission system is a function of task relevance, represents the comprehensive rate at which task priority changes with system disturbances, and is the weight coefficient, which indicates the influence of different parameters in the calculation.

[0094] Execution process:

[0095] First, according to the production task information, determine the task priority parameter and system disturbance parameter , and combined with the task importance parameter and the urgency parameter , construct a comprehensive function that describes the stability of the task system and its importance and urgency Functions associated with tasks ;

[0096] Next, calculate and , respectively, represent the sensitivity of task priority and relevance to these parameters;

[0097] Then, determine the weight coefficient and ,in is the importance weight of the priority and the stability of the task system, The weights of relevance and urgency are determined based on historical data or expert scoring. By comparing the actual sorting results with the expected sorting, the optimal weight value is determined using optimization methods such as the least squares method. Finally, the task priority change rate is calculated using the improved formula to obtain the final priority sorting result.

[0098] In this embodiment, Laplace transform and transient behavior analysis are performed according to the formula:

[0099]

[0100] Calculate the trend value of the operating parameter change, where: For the improved representation of the operating parameters in the frequency domain, is the function of the operating parameter in the time domain, is the complex frequency parameter, is the time variable, is the weight coefficient, which indicates the reference amplitude adjustment of the operating parameter. is the frequency modulation parameter, which indicates the frequency of change of the operating parameter over time. is a phase offset parameter, indicating the initial phase offset of the operating parameter.

[0101] Execution process:

[0102] First, determine the function of the operating parameters in the time domain and the complex frequency parameters , and then introduce new parameters , and ,in It indicates the adjustment of the base amplitude of the operating parameter, and its optimal value is determined by combining historical data or through statistical analysis. is the frequency modulation parameter, which indicates the frequency of change of the operating parameter over time and can be determined by frequency analysis of the system response. is the phase offset parameter, which indicates the initial phase offset of the operating parameter and can be set according to the initial conditions of the system

[0103] Next, these parameters are substituted into the improved Laplace transform formula, combined with the time domain function of the operating parameters and complex frequency parameters , perform integral calculations to obtain the representation of the operating parameters in the frequency domain ,Finally, through the improved formula, a more accurate value of ,operating parameter change trend is generated.

[0104] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0105] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0106] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent factory production and training system based on digital twins, characterized in that: The system comprises: The dynamic production scheduling module uses bifurcation theory and singular perturbation method based on production task information to extract task priority, resource allocation status and task execution time, compare task priority with resource utilization rate, calculate time difference and resource occupancy rate, identify key turning points and adjust resource allocation order to generate dynamic scheduling plan; The bifurcation theory and singular perturbation method are based on the formula: Calculate the task priority sorting results, where: is the task priority parameter, is the system disturbance parameter, is the task importance parameter, is the urgency parameter of the task, A comprehensive function that describes the stability, importance, and urgency of the mission system is a function of task relevance, represents the comprehensive rate at which task priorities change with system disturbances, and is the weight coefficient, which indicates the influence of different parameters in the calculation; The task switching coordination module extracts the task connection sequence and switching time nodes based on the dynamic scheduling scheme, analyzes the time distribution of the connection points and the task execution sequence, calculates the time difference during the switching period and adjusts the time parameters and the connection sequence to generate a coordinated switching scheme; A real-time feedback adjustment module, based on the coordinated switching scheme, extracts execution data and predetermined parameters, compares the execution data with the predetermined parameters, calculates deviation values, adjusts execution paths and task priorities, and generates feedback adjustment results; The production task optimization module analyzes the task execution time and resource consumption data based on the feedback adjustment results, calculates the difference between time and resource allocation, readjusts the task priority and resource allocation order, optimizes the matching degree between tasks and resources, and generates a production optimization plan; The training simulation module, based on the production optimization plan, extracts the change trend of operating parameters and the task execution path through Laplace transform and transient behavior analysis, simulates the optimized production process, analyzes the change of operating parameters, adjusts the training plan, and generates a training plan; The Laplace transform and transient behavior analysis are according to the formula: Calculate the trend value of the operating parameter change, where: For the improved representation of the operating parameters in the frequency domain, is the function of the operating parameter in the time domain, is the complex frequency parameter, is the time variable, is the weight coefficient, which indicates the reference amplitude adjustment of the operating parameter. is the frequency modulation parameter, which indicates the frequency of change of the operating parameter over time. is a phase offset parameter, indicating the initial phase offset of the operating parameter.

2. The intelligent factory production and training system based on digital twin according to claim 1 is characterized by: The dynamic scheduling scheme includes adjusted task priorities, optimized resource allocation strategies and updated time node arrangements; the coordinated switching scheme includes adjusted switching times, optimized task connection sequences and updated time control parameters; the feedback adjustment results include revised execution paths, reallocated task priorities and adjusted deviation control parameters; the production optimization scheme includes optimized resource allocation strategies, adjusted task priorities and improved matching parameters; the training scheme includes updated operator training plans, simulated production scenarios and adjusted parameter change guidance.

3. The intelligent factory production and training system based on digital twin according to claim 1 is characterized in that: The dynamic production scheduling module includes: The task priority submodule, based on production task information, uses bifurcation theory and singular perturbation method to extract task priority and related data, organize and classify task data, analyze the importance and urgency of tasks, prioritize them, and compare the relevance between tasks to generate priority ranking results; The resource allocation submodule extracts the resource allocation status and related data based on the priority sorting result, analyzes the resource allocation and usage, calculates the current resource usage rate, adjusts the resource allocation order, and reallocates resources according to the priority to generate a resource allocation adjustment result; The time calculation submodule extracts the task execution time and related data based on the resource allocation adjustment results, segments and compares the task time, calculates the time difference between tasks, identifies the key time nodes in the task execution process, adjusts the time arrangement, and generates a dynamic scheduling plan.

4. The intelligent factory production and training system based on digital twin according to claim 1 is characterized in that: The task switching coordination module includes: The task connection submodule extracts the task connection sequence and related data based on the dynamic scheduling scheme, analyzes the dependency relationship between multiple tasks, adjusts the task connection sequence, and generates a preliminary connection sequence; The time node submodule extracts the switching time nodes and related data based on the preliminary connection sequence, calculates and adjusts the time nodes of task switching, optimizes the arrangement of the time nodes, and generates adjusted time nodes; The switching sequence submodule extracts the time distribution and associated data of the connection points based on the adjusted time nodes, optimizes the task switching sequence, adjusts the sequence and time parameters of the task connection, and generates a coordinated switching plan.

5. The intelligent factory production and training system based on digital twin according to claim 1 is characterized in that: The real-time feedback adjustment module comprises: An execution data extraction submodule extracts execution data and associated parameters based on the coordinated switching scheme, monitors and records the data during task execution in real time, compares the execution data with the predetermined parameters, analyzes the differences and summarizes the results, and generates deviation data; The deviation calculation submodule extracts the deviation value and related parameters based on the deviation data, calculates the error amplitude during task execution, classifies and sorts the deviation values, determines the necessity of adjustment, and formulates an adjustment plan according to the deviation size to generate an adjustment demand result; The path adjustment submodule extracts the execution path and task priority based on the adjustment requirement result, analyzes and adjusts the execution path, reorders the priority of the tasks, adjusts the execution order and path of the tasks, and generates feedback adjustment results.

6. The intelligent factory production and training system based on digital twin according to claim 1 is characterized in that: The production task optimization module includes: The time analysis submodule extracts the task execution time and related data based on the feedback adjustment result, analyzes and compares the task execution time, calculates the difference value of the time allocation of multiple tasks, and generates the time difference analysis result; The resource allocation optimization submodule extracts resource consumption data and related information based on the time difference analysis results, analyzes and optimizes the resource allocation situation, reallocates resources, calculates the efficiency after resource allocation, and generates resource optimization results; The matching degree adjustment submodule extracts task and resource matching data and related information based on the resource optimization result, analyzes the matching degree of tasks and resources, adjusts the allocation sequence of tasks and resources, and generates a production optimization plan.

7. The intelligent factory production and training system based on digital twin according to claim 1 is characterized in that: The training simulation module includes: The parameter extraction submodule, based on the production optimization scheme, extracts the change trend of the operating parameters and the associated data through Laplace transform and transient behavior analysis, analyzes the change pattern of the operating parameters, classifies and sorts the parameter change trend, and generates parameter classification records; The production simulation submodule extracts the task execution path and related information based on the parameter classification records, simulates the optimized production process, analyzes the data changes during the simulation process, records and summarizes the simulation data, and generates simulation analysis results; The training adjustment submodule extracts the changes in operation parameters and task execution paths based on the simulation analysis results, analyzes and adjusts the operation training plan, formulates new training content, updates the operation guidance of the training, and generates a training plan.

Images