Modularized spinning equipment dynamic scheduling system

Abstract

The invention relates to the technical field of spinning equipment scheduling, in particular to a modular spinning equipment dynamic scheduling system which comprises a data processing unit, a production modeling unit, a collaborative scheduling unit, a process instruction generation unit and a quality regulation and control unit. The method comprises the following steps: standardizing production data of a spinning factory to obtain standard production data, and constructing an operation digital twinborn model; generating a candidate collaborative scheduling scheme set based on the operation digital twin model and different execution paths and resource combinations of the global production task among the plurality of factories, and selecting a global production scheduling plan from the candidate collaborative scheduling scheme set; according to the local production task allocated to the target factory and the corresponding preset product quality characteristic requirement, in combination with the operation digital twin model, generating a process parameter instruction for executing the local production task; and acquiring actual product quality data of the target factory, comparing the actual product quality data with the preset product quality characteristic requirement, and if a deviation exists, correcting the process parameter instruction according to a preset adjustment algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of spinning equipment scheduling, in particular to a modularized spinning equipment dynamic scheduling system. Background Art

[0002] With the rapid development of the textile industry, traditional spinning equipment scheduling systems are no longer able to meet modern production demands. Currently, most widely used spinning equipment scheduling systems utilize fixed production plans, relying on experience or simple information systems for scheduling and dispatching. These systems struggle to efficiently allocate production resources when handling flexible production of multiple varieties and small batches, leading to low equipment utilization, long production cycles, and high energy consumption. Existing systems lack the ability to dynamically adjust in real time to cope with unexpected order changes, equipment failures, or fluctuations in raw material supply, often resulting in production interruptions or significant delays, increasing operating costs.

[0003] Amidst the increasing complexity and volatility of global supply chains, existing spinning equipment scheduling systems struggle to achieve cross-regional, multi-factory collaborative production and optimal resource allocation. This is particularly true for textile companies with multiple production bases. When a factory faces raw material shortages, energy constraints, or production bottlenecks, existing systems are unable to intelligently divert production tasks and complement resources. This results in an inability to flexibly adjust the company's overall production capacity, increasing logistics costs and carbon emissions while also extending order delivery cycles. Furthermore, information such as equipment status, process parameters, and quality standards cannot be shared and centrally managed in real time across factories, hindering standardized production and quality consistency control. Developing a modular, dynamic spinning equipment scheduling system that supports cross-factory collaboration and resource sharing to achieve optimal manufacturing process management has become a key technical challenge.

[0004] Therefore, a modular dynamic scheduling system for spinning equipment is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a modular dynamic scheduling system for spinning equipment, which realizes the global optimal production resource configuration through a data processing unit, a production modeling unit, a collaborative scheduling unit, a process instruction generation unit and a quality control unit. The system includes: standardizing the production data of the spinning plant to obtain standard production data and constructing an operation digital twin model; based on the operation digital twin model and the different execution paths and resource combinations of global production tasks among multiple factories, generating a set of candidate collaborative scheduling schemes and selecting a global production scheduling plan from them; for the local production tasks assigned to the target factory and the corresponding preset product quality characteristic requirements, combining the operation digital twin model, generating process parameter instructions for executing the local production tasks; obtaining the actual product quality data of the target factory, comparing it with the preset product quality characteristic requirements, and if there is a deviation, correcting the process parameter instructions according to a preset adjustment algorithm.

[0006] To achieve the above object, the present invention provides the following technical solutions: A modular spinning equipment dynamic scheduling system, comprising: The data processing unit standardizes the production data of the spinning plant to obtain standard production data; a production modeling unit, which builds operational digital twin models for multiple spinning plants based on the standard production data; A collaborative scheduling unit, based on the operational digital twin model, generates a set of candidate collaborative scheduling solutions by simulating different execution paths and resource combinations of global production tasks across multiple factories, and selects a global production scheduling plan from the set of candidate collaborative scheduling solutions based on preset management objectives; a process instruction generation unit, receiving the global production scheduling plan, and generating, based on the local production tasks assigned to the target factory and the corresponding preset product quality characteristic requirements, a process parameter instruction for the target factory to execute the local production tasks in combination with the operation digital twin model of the target factory; The quality control unit obtains the actual product quality data of the target factory when the target factory performs the local production task, compares it with the preset product quality characteristic requirements, and corrects the process parameter instructions according to the preset adjustment algorithm if there is a deviation.

[0007] Preferably, the operational digital twin model is constructed for multiple spinning plants, and the specific process includes: Based on the standard production data, identify the key spinning equipment list and main production process nodes in each spinning plant; For key spinning equipment and main production process nodes, extract corresponding historical operating parameters, process parameter setting records, material consumption data and equipment maintenance information from the standard production data; Based on the extracted data, a performance characterization model is established for each key spinning equipment, and an input-output relationship model is established for each major production process node; The performance characterization model and the input-output relationship model are integrated and combined with the fixed asset information and layout information of each factory to construct the operation digital twin model.

[0008] Preferably, based on the operational digital twin model, a candidate collaborative scheduling solution set is generated by simulating different execution paths and resource combinations of global production tasks across multiple factories. The specific process includes: Receive the global production task and, based on the product specifications, quantity, and delivery requirements contained therein and in combination with the operational digital twin model of the spinning plant, determine a set of execution paths for completing the global production task, wherein each execution path includes the subtasks and processing sequences into which the global production task is decomposed in different plants; For each execution path, a set of resource allocation plans is matched based on the operational digital twin model. Each resource allocation plan includes the specific equipment model and quantity required for the execution path, as well as the corresponding material supply batch; Each execution path and its corresponding resource allocation plan form a collaborative scheduling combination. The operational digital twin model is used to perform simulation operations to predict the total time required for the collaborative scheduling combination to complete the global production task, the average equipment load of each related factory, the main material consumption indicators, and quantitative data on cross-factory material transfer. The prediction results are summarized to form comprehensive evaluation data for the corresponding collaborative scheduling combination. All the coordinated scheduling combinations and the corresponding comprehensive evaluation data are collected to form the candidate coordinated scheduling solution set.

[0009] Preferably, the global production scheduling plan includes local production tasks assigned to each target factory, as well as the planned start time, planned completion time, required key equipment and main material information of each local production task at the corresponding target factory.

[0010] Preferably, the specific process of generating the process parameter instruction for the target factory to execute the local production task includes: Based on the characteristics of the local production task, key process parameters are extracted according to corresponding successful process cases in the historical database, and the key process parameters are adaptively adjusted in combination with the current working conditions of the target factory to obtain the process parameter instructions; If there is no corresponding successful process case in the historical database, a parameter calculation optimization model is trained through machine learning methods, and the preset product quality characteristic requirements and the operating digital twin model parameters of the target factory are input to generate the process parameter instructions.

[0011] Preferably, the specific process of correcting the process parameter instruction according to the preset adjustment algorithm includes: If the deviation of any product quality indicator exceeds the preset allowable fluctuation range, the process parameter instruction correction process will be activated; The process parameter instruction correction process identifies key process parameters that need to be adjusted and calculates recommended adjustment values ​​for the key process parameters based on the sensitivity analysis results. The sensitivity analysis results describe the degree of impact of changes in key process parameters on product quality indicators. The process parameter instructions are modified according to the suggested adjustment values.

[0012] Preferably, the quality control unit also includes: during the execution of the global production scheduling plan, obtaining the updated standard production data from the data processing unit, monitoring the actual production progress and resource availability changes of multiple spinning plants; when the deviation between the monitored real-time status of the resource availability and the planned resource quantity required to execute the global production scheduling plan exceeds a preset availability deviation threshold, it is automatically triggered to regenerate the global production scheduling plan based on the updated standard production data and the operation digital twin model.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention standardizes the production data of multiple spinning plants through a data processing unit, laying a unified data foundation for subsequent precise modeling. The production modeling unit builds an operational digital twin model for each spinning plant based on standard production data. The operational digital twin model is not only a static description, but can also simulate the performance boundaries of the equipment and the material and information flow laws of the process flow, so that the production capacity, process constraints, resource status and even operational bottlenecks of each plant can be accurately mapped and dynamically presented in a digital way. It greatly improves the transparency and predictability of the production status of multiple plants, allowing managers to clearly grasp the real-time operating status and potential production capacity of each plant, and provides data insights and accurate simulation foundations for cross-plant collaborative decision-making.

[0014] 2. The collaborative scheduling unit proposed in the present invention uses the digital twin model of each factory operation to realize the intelligent planning and resource optimization of global production tasks across multiple factories, breaking through the limitations of traditional single-factory scheduling. By combining the execution path and resource allocation scheme into a collaborative scheduling combination, and using the digital twin model for simulation calculations, it is possible to accurately predict the total duration of each combination, the equipment load of each factory, material consumption, and quantitative data of cross-factory material transfer to form a comprehensive evaluation. Finally, the evaluation data of all combinations are collected to form a set of candidate collaborative scheduling schemes, and the global production scheduling plan is selected from them based on the preset management objectives. This mechanism can dynamically evaluate and select the optimal cross-factory production and resource allocation strategy from the perspective of the entire enterprise, realize intelligent diversion of production tasks and resource complementarity, significantly improve the overall operational efficiency and resource utilization of the enterprise, and effectively solve the problem of cross-regional collaborative production and resource optimization allocation.

[0015] 3. The collaborative work of the process instruction generation unit and the quality control unit of the present invention realizes a complete mechanism from global planning to local precise execution, and combines quality feedback for closed-loop control and dynamic adaptation. The process instruction generation unit can generate specific process parameter instructions for local production tasks assigned to specific target factories, making full use of historical success cases or parameter optimization models obtained through machine learning to ensure the initial rationality of the process parameter instructions and their adaptability to the characteristics of the target factory. The quality control unit monitors during the execution of production tasks, which not only ensures the quality consistency and process accuracy of each local production task in the context of multi-factory collaborative production, but also gives the entire scheduling system the ability to dynamically adapt to actual production fluctuations and resource changes, ensuring the continued feasibility and effectiveness of the plan, and improving the overall flexibility of the company's production capacity allocation and the reliability of order delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic structural diagram of a modularized spinning equipment dynamic scheduling system provided by an embodiment of the present invention; Figure 2 A schematic diagram of the process of generating a candidate collaborative scheduling solution set provided by an embodiment of the present invention; Figure 3 A schematic diagram of the implementation flow of a modularized spinning equipment dynamic scheduling system provided by an embodiment of the present invention. DETAILED DESCRIPTION

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

[0018] See also Figures 1 to 3 The present invention provides a modular spinning equipment dynamic scheduling system, the technical solution is as follows: Example 1: A modular spinning equipment dynamic scheduling system, such as Figure 1 Shown, including: The data processing unit standardizes the production data of the spinning plant to obtain standard production data; a production modeling unit, which builds operational digital twin models for multiple spinning plants based on the standard production data; A collaborative scheduling unit, based on the operational digital twin model, generates a set of candidate collaborative scheduling solutions by simulating different execution paths and resource combinations of global production tasks across multiple factories, and selects a global production scheduling plan from the set of candidate collaborative scheduling solutions based on preset management objectives; a process instruction generation unit, receiving the global production scheduling plan, and generating, based on the local production tasks assigned to the target factory and the corresponding preset product quality characteristic requirements, a process parameter instruction for the target factory to execute the local production tasks in combination with the operation digital twin model of the target factory; The quality control unit obtains the actual product quality data of the target factory when the target factory performs the local production task, compares it with the preset product quality characteristic requirements, and corrects the process parameter instructions according to the preset adjustment algorithm if there is a deviation.

[0019] In this example, a textile enterprise has three geographically dispersed spinning plants (referred to as Plant A, Plant B, and Plant C). Each plant has different models and quantities of spinning equipment, producing a variety of yarn specifications. Through the integrated systems of each spinning plant, various production data are aggregated, including: order information (order number, product code, product specifications, order quantity, customer delivery deadline, priority, etc.); equipment operating conditions (equipment number, equipment model, operating status, spindle speed, output count, energy consumption readings, sensor data such as equipment temperature); material information (raw material batch number, raw material inventory, work-in-progress inventory, finished product inventory, material location, supplier information, expected arrival time, etc.); process settings (process card number used for the current production task, key process parameter settings for each process step); and quality results. Standardization includes unified data formats and encoding, timestamp synchronization and calibration, and data cleansing.

[0020] Furthermore, the operational digital twin model is constructed for multiple spinning plants, and the specific process includes: Based on the standard production data, identify the key spinning equipment list and main production process nodes in each spinning plant; For key spinning equipment and main production process nodes, extract corresponding historical operating parameters, process parameter setting records, material consumption data and equipment maintenance information from the standard production data; Based on the extracted data, a performance characterization model is established for each key spinning equipment, and an input-output relationship model is established for each major production process node; The performance characterization model and the input-output relationship model are integrated and combined with the fixed asset information and layout information of each factory to construct the operation digital twin model.

[0021] The construction of the operational digital twin model involves identifying key elements, extracting multi-dimensional historical data, and establishing and integrating performance and relationship models. This ensures that the constructed digital twin model can highly realistically reflect the actual operating status and production capacity of each spinning plant. This provides an accurate and reliable model foundation for subsequent collaborative scheduling and process instruction generation, improving the accuracy and effectiveness of system decision-making.

[0022] Specifically, each plant has its own independent operational digital twin model, which includes not only static descriptions but, more importantly, dynamic behavior and capability characterization. In the spinning process, key spinning equipment, such as the blowroom, carding, draw frame, roving frame, spinning frame, and winding frame, are identified. The blowroom, carding, draw frame, roving, spinning, and winding processes are identified as key production process nodes. A performance characterization model is developed for each key spinning equipment. For example, a model for the spinning frame uses input parameters such as raw material type, yarn count, twist setting, spindle speed setting, and workshop temperature and humidity. The outputs are predicted output, predicted end-down rate, and predicted energy consumption. This performance characterization model can be developed based on historical data using data-driven methods such as regression analysis and neural networks. An input-output relationship model is developed for each key production process node. For example, a model for the draw frame uses input parameters such as sliver weight unevenness and short fiber percentage from the preceding process (carding) and outputs the improvement in sliver weight unevenness after the draw frame. Integrate the performance characterization model of each device and the input-output relationship model of the process node with the fixed asset information of each factory (such as equipment ledger, specification parameters), factory layout information (such as workshop area, equipment layout diagram, logistics path), and real-time equipment status, work-in-process inventory and other information, and form an operational digital twin model through a unified platform or framework that can simulate and predict the overall operational behavior of the factory.

[0023] Furthermore, based on the operational digital twin model, a set of candidate collaborative scheduling solutions is generated by simulating different execution paths and resource combinations of global production tasks across multiple factories. The specific process includes: Receive the global production task and, based on the product specifications, quantity, and delivery requirements contained therein and in combination with the operational digital twin model of the spinning plant, determine a set of execution paths for completing the global production task, wherein each execution path includes the subtasks and processing sequences into which the global production task is decomposed in different plants; For each execution path, a set of resource allocation plans is matched based on the operational digital twin model. Each resource allocation plan includes the specific equipment model and quantity required for the execution path, as well as the corresponding material supply batch; Each execution path and its corresponding resource allocation plan form a collaborative scheduling combination. The operational digital twin model is used to perform simulation operations to predict the total time required for the collaborative scheduling combination to complete the global production task, the average equipment load of each related factory, the main material consumption indicators, and quantitative data on cross-factory material transfer. The prediction results are summarized to form comprehensive evaluation data for the corresponding collaborative scheduling combination. All the coordinated scheduling combinations and the corresponding comprehensive evaluation data are collected to form the candidate coordinated scheduling solution set.

[0024] The schematic diagram of the generation process of candidate collaborative scheduling solution set is shown in Figure 2 By determining execution paths, matching resource allocations, simulating and predicting key indicators, and conducting comprehensive evaluations, it is possible to systematically explore and compare various cross-factory collaborative production possibilities. This ensures that the final global production scheduling plan is an optimized decision based on a comprehensive consideration of various complex factors, thereby maximizing the company's overall operational benefits, such as reducing total costs, shortening delivery cycles, or improving resource utilization.

[0025] Specifically, when a company receives a large order containing multiple yarn products, a large total volume, and a tight delivery deadline, this order becomes a global production task. A rule-based reasoning engine can be used to generate a preliminary set of feasible paths; alternatively, a heuristic search algorithm can be used, such as one based on the expansion and combination of product process routes, to explore and prune diverse execution paths across a multi-factory network. The resource allocation plan matching process queries the real-time equipment availability, material inventory, and batch information in the digital twin models of each relevant factory's operations. Furthermore, the resource allocation plan matches the requirements of each subtask in the execution path for equipment type, processing capacity, and material specifications. Constraint satisfaction or optimization algorithms (such as variations of linear programming or assignment problem algorithms) are used to screen or combine specific equipment instances, quantities, time periods, and available material batches that meet the conditions, thereby forming a resource allocation plan that matches the execution path.

[0026] When selecting a global production scheduling plan, based on the preset management objectives, for example, setting the management objective to "lowest total cost", the candidate collaborative scheduling scheme set is evaluated and ranked, and the scheme that best meets the preset management objectives is selected as the global production scheduling plan based on the comprehensive evaluation data.

[0027] Furthermore, the global production scheduling plan includes local production tasks assigned to each target factory, as well as the planned start time, planned completion time, required key equipment and main material information of each local production task in the corresponding target factory.

[0028] The macro-global plan can be effectively broken down and delivered to each execution level, providing clear and actionable guidance for the precise generation of subsequent process instructions and the orderly execution of the production process, ensuring the feasibility of the plan and the coordination and consistency among factories.

[0029] Furthermore, the specific process of generating the process parameter instruction for the target factory to execute the local production task includes: Based on the characteristics of the local production task, key process parameters are extracted according to corresponding successful process cases in the historical database, and the key process parameters are adaptively adjusted in combination with the current working conditions of the target factory to obtain the process parameter instructions; If there is no corresponding successful process case in the historical database, a parameter calculation optimization model is trained through machine learning methods, and the preset product quality characteristic requirements and the operating digital twin model parameters of the target factory are input to generate the process parameter instructions.

[0030] Combining the advantages of experience reuse and intelligent computing, the generated process parameter instructions can not only draw on proven mature experience, but also make intelligent recommendations for new situations, thereby improving the accuracy and efficiency of process parameter settings and helping to ensure product quality and production stability.

[0031] Furthermore, the specific process of correcting the process parameter instruction according to the preset adjustment algorithm includes: If the deviation of any product quality indicator exceeds the preset allowable fluctuation range, the process parameter instruction correction process will be activated; The process parameter instruction correction process identifies key process parameters that need to be adjusted and calculates recommended adjustment values ​​for the key process parameters based on the sensitivity analysis results. The sensitivity analysis results describe the degree of impact of changes in key process parameters on product quality indicators. The process parameter instructions are modified according to the suggested adjustment values.

[0032] A dynamic adjustment mechanism based on actual product quality feedback has been established, which can promptly correct process deviations in the production process and continuously optimize process parameters, thereby effectively ensuring and improving the quality consistency of the final product and reducing the defective rate.

[0033] Specifically, if the system's preset allowable fluctuation range is ±3% of the preset product quality characteristic requirements, during a certain period, the yarn unevenness of products produced by Factory A was measured three times online, yielding results of 14.5%, 14.6%, and 14.4%. These fluctuations exceeded the upper limit of the preset product quality characteristic requirement (strip unevenness ≤ 14.0%), triggering the process parameter instruction correction process. A pre-stored sensitivity analysis model (which describes the degree and direction of the impact of changes in various process parameters on yarn unevenness) was called. Based on the current deviation and model analysis, it was determined that adjusting the twist coefficient and front roller gauge were effective means of improving yarn unevenness. Considering the equipment's constraints (e.g., the spindle speed is nearing its upper limit, making it unsuitable to increase the speed to indirectly affect splicing quality), a preset adjustment algorithm (e.g., rule-based reasoning or simple PID control logic) calculated the recommended adjustment values.

[0034] Furthermore, the quality control unit also includes: during the execution of the global production scheduling plan, obtaining the updated standard production data from the data processing unit, monitoring the actual production progress and resource availability changes of multiple spinning plants; when the deviation between the monitored real-time status of the resource availability and the planned resource quantity required to execute the global production scheduling plan exceeds a preset availability deviation threshold, it is automatically triggered to regenerate the global production scheduling plan based on the updated standard production data and the operation digital twin model.

[0035] The system monitors changes in production progress and resource availability, triggering the regeneration of the global production schedule when resource availability deviations exceed a threshold. This enables the dynamic scheduling system to rapidly respond to and adapt to changes in the actual production environment, enhancing its robustness and flexibility in the face of emergencies such as raw material shortages and equipment failures, and ensuring the continued viability of production plans.

[0036] Specifically, the preset availability deviation threshold is a critical value or a set of rules set for each key resource based on factors such as production experience, material supply cycle, and equipment reliability. When the gap between actual availability and planned demand reaches or exceeds this critical point, a resource availability change requiring intervention is considered to have occurred.

[0037] The present invention realizes transparent management and accurate prediction of the production capacity and status of each factory by constructing a refined operational digital twin model for each factory. On this basis, the collaborative scheduling unit can carry out global task allocation and resource optimization across factories, intelligently generate a global production scheduling plan, effectively deal with problems such as raw material shortages and production bottlenecks, and realize intelligent diversion of production tasks and resource complementarity. The process instruction generation and quality control unit ensures the accurate implementation of the global plan in each factory and the consistency of product quality, and improves the system's adaptability to actual production fluctuations through closed-loop feedback and dynamic adjustment mechanisms. The present invention can significantly improve the collaborative production efficiency and resource allocation level of textile enterprises across multiple regions and factories, reduce operating costs, and enhance the overall production flexibility and market competitiveness of the enterprise.

[0038] Example 2: In this embodiment, the embodiment 1 is further described, and the specific implementation process of a modular spinning equipment dynamic scheduling system is shown in FIG. Figure 3 ,include: Standardize the production data of the spinning plant to obtain standard production data; Building operational digital twin models for multiple spinning plants based on the standard production data; Furthermore, the operational digital twin model is constructed for multiple spinning plants, and the specific process includes: Based on the standard production data, identify the key spinning equipment list and main production process nodes in each spinning plant; For key spinning equipment and main production process nodes, extract corresponding historical operating parameters, process parameter setting records, material consumption data and equipment maintenance information from the standard production data; Based on the extracted data, a performance characterization model is established for each key spinning equipment, and an input-output relationship model is established for each major production process node; The performance characterization model and the input-output relationship model are integrated and combined with the fixed asset information and layout information of each factory to construct the operation digital twin model.

[0039] Table 1 shows some key parameters and descriptions included in the digital twin model of the spinning section operation of Factory A. The applicable yarn count range is 20S-60S, where S is the unit of British cotton spinning count, which is used to represent the fineness of the yarn. The larger the value, the finer the yarn.

[0040] Table 1 Examples of key parameters of the operational digital twin model Based on the operational digital twin model, a set of candidate collaborative scheduling solutions is generated by simulating different execution paths and resource combinations of global production tasks across multiple factories. A global production scheduling plan is selected from the set of candidate collaborative scheduling solutions based on preset management objectives. Furthermore, based on the operational digital twin model, a set of candidate collaborative scheduling solutions is generated by simulating different execution paths and resource combinations of global production tasks across multiple factories. The specific process includes: Receive the global production task and, based on the product specifications, quantity, and delivery requirements contained therein and in combination with the operational digital twin model of the spinning plant, determine a set of execution paths for completing the global production task, wherein each execution path includes the subtasks and processing sequences into which the global production task is decomposed in different plants; For each execution path, a set of resource allocation plans is matched based on the operational digital twin model. Each resource allocation plan includes the specific equipment model and quantity required for the execution path, as well as the corresponding material supply batch; Each execution path and its corresponding resource allocation plan form a collaborative scheduling combination. The operational digital twin model is used to perform simulation operations to predict the total time required for the collaborative scheduling combination to complete the global production task, the average equipment load of each related factory, the main material consumption indicators, and quantitative data on cross-factory material transfer. The prediction results are summarized to form comprehensive evaluation data for the corresponding collaborative scheduling combination. All the coordinated scheduling combinations and the corresponding comprehensive evaluation data are collected to form the candidate coordinated scheduling solution set.

[0041] receiving the global production scheduling plan, and generating, based on the local production tasks assigned to the target factory and the corresponding preset product quality characteristic requirements, a process parameter instruction for the target factory to execute the local production tasks in combination with the operational digital twin model of the target factory; Furthermore, the global production scheduling plan includes the local production tasks assigned to each target factory, as well as the planned start time, planned completion time, required key equipment and main material information of each local production task at the corresponding target factory; Furthermore, the specific process of generating the process parameter instruction for the target factory to execute the local production task includes: Based on the characteristics of the local production task, key process parameters are extracted according to corresponding successful process cases in the historical database, and the key process parameters are adaptively adjusted in combination with the current working conditions of the target factory to obtain the process parameter instructions; If there is no corresponding successful process case in the historical database, a parameter calculation optimization model is trained through machine learning methods, and the preset product quality characteristic requirements and the operating digital twin model parameters of the target factory are input to generate the process parameter instructions.

[0042] When the target factory performs the local production task, actual product quality data of the target factory is obtained and compared with the preset product quality characteristic requirements. If there is a deviation, the process parameter instruction is modified according to a preset adjustment algorithm; Furthermore, the specific process of correcting the process parameter instruction according to the preset adjustment algorithm includes: If the deviation of any product quality indicator exceeds the preset allowable fluctuation range, the process parameter instruction correction process will be activated; The process parameter instruction correction process identifies key process parameters that need to be adjusted and calculates recommended adjustment values ​​for the key process parameters based on the sensitivity analysis results. The sensitivity analysis results describe the degree of impact of changes in key process parameters on product quality indicators. Modifying the process parameter instruction according to the suggested adjustment value; Furthermore, it also includes: during the execution of the global production scheduling plan, obtaining the updated standard production data, monitoring the actual production progress and resource availability changes of multiple spinning plants; when the deviation between the real-time status of the monitored resource availability and the planned resource quantity required to execute the global production scheduling plan exceeds a preset availability deviation threshold, automatically triggering the regeneration of the global production scheduling plan based on the updated standard production data and the operation digital twin model.

[0043] Table 2 shows the results of comparing key indicators before and after the implementation of the present invention through simulation data. The specific indicators are monthly average values.

[0044] Table 2 Comparison of key indicators before and after implementation While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A modular spinning equipment dynamic scheduling system, characterized in that: include: The data processing unit standardizes the production data of the spinning plant to obtain standard production data; a production modeling unit, which builds operational digital twin models for multiple spinning plants based on the standard production data; A collaborative scheduling unit, based on the operational digital twin model, generates a set of candidate collaborative scheduling solutions by simulating different execution paths and resource combinations of global production tasks across multiple factories, and selects a global production scheduling plan from the set of candidate collaborative scheduling solutions based on preset management objectives; a process instruction generation unit, receiving the global production scheduling plan, and generating, based on the local production tasks assigned to the target factory and the corresponding preset product quality characteristic requirements, a process parameter instruction for the target factory to execute the local production tasks in combination with the operation digital twin model of the target factory; The quality control unit obtains the actual product quality data of the target factory when the target factory performs the local production task, compares it with the preset product quality characteristic requirements, and corrects the process parameter instructions according to the preset adjustment algorithm if there is a deviation.

2. A modular spinning equipment dynamic scheduling system according to claim 1, characterized in that: The operational digital twin models were built for multiple spinning plants. The specific process included: Based on the standard production data, identify the key spinning equipment list and main production process nodes in each spinning plant; For key spinning equipment and main production process nodes, extract corresponding historical operating parameters, process parameter setting records, material consumption data and equipment maintenance information from the standard production data; Based on the extracted data, a performance characterization model is established for each key spinning equipment, and an input-output relationship model is established for each major production process node; The performance characterization model and the input-output relationship model are integrated and combined with the fixed asset information and layout information of each factory to construct the operation digital twin model.

3. A modular spinning equipment dynamic scheduling system according to claim 1, characterized in that: Based on the operational digital twin model, a set of candidate collaborative scheduling solutions is generated by simulating different execution paths and resource combinations of global production tasks across multiple factories. The specific process includes: Receive the global production task and, based on the product specifications, quantity, and delivery requirements contained therein and in combination with the operational digital twin model of the spinning plant, determine a set of execution paths for completing the global production task, wherein each execution path includes the subtasks and processing sequences into which the global production task is decomposed in different plants; For each execution path, a set of resource allocation plans is matched based on the operational digital twin model. Each resource allocation plan includes the specific equipment model and quantity required for the execution path, as well as the corresponding material supply batch; Each execution path and its corresponding resource allocation plan form a collaborative scheduling combination. The operational digital twin model is used to perform simulation operations to predict the total time required for the collaborative scheduling combination to complete the global production task, the average equipment load of each related factory, the main material consumption indicators, and quantitative data on cross-factory material transfer. The prediction results are summarized to form comprehensive evaluation data for the corresponding collaborative scheduling combination. All the coordinated scheduling combinations and the corresponding comprehensive evaluation data are collected to form the candidate coordinated scheduling solution set.

4. A modular spinning equipment dynamic scheduling system according to claim 1, characterized in that: The global production scheduling plan includes the local production tasks assigned to each target factory, as well as the planned start time, planned completion time, required key equipment and main material information of each local production task at the corresponding target factory.

5. A modular spinning equipment dynamic scheduling system according to claim 1, characterized in that: The specific process of generating the process parameter instructions for the target factory to execute the local production task includes: Based on the characteristics of the local production task, key process parameters are extracted according to corresponding successful process cases in the historical database, and the key process parameters are adaptively adjusted in combination with the current working conditions of the target factory to obtain the process parameter instructions; If there is no corresponding successful process case in the historical database, a parameter calculation optimization model is trained through machine learning methods, and the preset product quality characteristic requirements and the operating digital twin model parameters of the target factory are input to generate the process parameter instructions.

6. A modular spinning equipment dynamic scheduling system according to claim 1, characterized in that: The specific process of correcting the process parameter instructions according to the preset adjustment algorithm includes: If the deviation of any product quality indicator exceeds the preset allowable fluctuation range, the process parameter instruction correction process will be activated; The process parameter instruction correction process identifies key process parameters that need to be adjusted and calculates recommended adjustment values ​​for the key process parameters based on the sensitivity analysis results. The sensitivity analysis results describe the degree of impact of changes in key process parameters on product quality indicators. The process parameter instructions are modified according to the suggested adjustment values.

7. A modular spinning equipment dynamic scheduling system according to claim 1, characterized in that: The quality control unit also includes: during the execution of the global production scheduling plan, obtaining the updated standard production data from the data processing unit, monitoring the actual production progress and resource availability changes of multiple spinning plants; when the deviation between the monitored real-time status of the resource availability and the planned resource quantity required to execute the global production scheduling plan exceeds the preset availability deviation threshold, it is automatically triggered to regenerate the global production scheduling plan based on the updated standard production data and the operation digital twin model.

Images