Self-adaptive production control system of discrete workshop

Abstract

The invention discloses a self-adaptive production control system of a discrete workshop. The invention provides a discrete workshop self-adaptive production control system based on a control theory. The problem that an existing production control system is insufficient in adaptability in a dynamic and changeable workshop environment is solved. A traditional modeling method cannot effectively analyze dynamic behaviors of a system and is difficult to deal with transient changes caused by disturbance such as emergency order insertion and equipment faults. By introducing an event-driven mechanism and a multi-target coordination control strategy, dynamic optimization and self-adaptive control of a workshop production system are realized. According to the system, work-in-process and overstock tasks are used as control targets, and dynamic adjustment is achieved by controlling the order input rate and the actual productivity. And the control strategy planning layer and the workshop operation decision-making layer are linked by adopting a centralized event-driven mechanism, and meanwhile, a distributed event-driven mechanism is used in the workshop operation decision-making layer to construct an intelligent unit autonomous decision-making mechanism. The method is suitable for dynamic production control of discrete manufacturing workshops, overcomes the defects in the prior art, and meets the requirements of complex, dynamic and changeable workshop operation environments.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, specifically to an adaptive production control system suitable for discrete manufacturing workshops, which can dynamically adjust production control strategies to adapt to the complexity and variability of the workshop environment, thereby improving production efficiency and system stability. Background Technology

[0002] In modern discrete manufacturing workshops, modeling and optimizing production control systems are crucial for achieving efficient production. Conventional modeling methods for production control systems, such as discrete event simulation, system dynamics, and Petri nets, cannot directly analyze the fundamental dynamic characteristics of the manufacturing system under transient conditions. For example, how do we estimate the system's settling time and overshoot when the system transitions from a steady state to an unstable state due to the insertion of urgent orders or equipment failures? How do we determine the system's damping characteristics based on the initial oscillation trend under different production lead times? How do we determine the range of disturbances that cause significant fluctuations in system performance due to internal and external disturbances in the workshop manufacturing system? Traditional modeling methods are mostly limited to assuming certain key parameters are constant, analyzing problems from a static system perspective, and neglecting the system's dynamic characteristics.

[0003] Modeling of production control systems based on control theory: Control theory is a powerful tool for analyzing dynamic systems. Evaluating the "intelligence" of production control systems by assessing the dynamic response and robustness of the system is an effective method for the operation control and evaluation analysis of workshop manufacturing systems.

[0004] Figure 1 The diagram shows the production planning and control system architecture established using this method. This model integrates the production planning layer and the shop floor control layer. The designed logic decision unit is equivalent to a prototype of modern ERP, mainly used for capacity planning, demand forecasting, and finished goods inventory control; by changing the control gain (G in the diagram)... c and G I This allows for adjustments to production capacity and inventory levels. Meanwhile, the workshop level maintains stable work-in-process levels by adjusting productivity.

[0005] The design philosophy of the above model is to ensure the synchronization of production capacity and production, focusing on upper-level planning and not mentioning the execution issues at the workshop level; the model uses work-in-process, productivity and inventory levels as control targets, and the logic judgment unit realizes the function of switching control strategies for different production control targets. Figure 1 The proposed control scheme is a centralized control architecture that does not take into account the "loose coupling" characteristics of the manufacturing units within the discrete workshop manufacturing system, making it difficult to directly apply to the production control of the discrete workshop.

[0006] Furthermore, in actual production, the production capacity of a workshop manufacturing system is limited, and the production capacity of bottleneck equipment determines the overall system capacity. For systems already in operation, their production capacity limits are fixed; that is, new processing equipment cannot be added based on fluctuations in production demand. Instead, the demand for production capacity can only be adjusted by extending working hours, increasing operator proficiency, or changing logistics routes. Secondly, in actual production control, the objectives of production control change with different production tasks. Therefore, the modeling of the production control system should revolve around the production control objectives and facilitate dynamic analysis.

[0007] Discrete manufacturing workshops contain a large number of system features, and it is impossible to use all features to effectively describe the operating state of the system. Therefore, feature selection is necessary to choose a small number of important features as key parameters of the production control system. Summary of the Invention

[0008] The purpose of this invention is to provide an adaptive production control system for discrete workshops to solve the problems mentioned in the background art above:

[0009] (1) Framework of an adaptive production control system for work-in-process and backlog based on control theory. Based on the key parameters of the discrete manufacturing workshop production control system: delivery rate, work-in-process inventory, equipment utilization rate and production cycle, control variables and control objectives are defined from the perspective of control theory. The control of work-in-process and backlog tasks is achieved by controlling the order input rate and actual production rate.

[0010] To achieve the above objectives, the present invention provides the following technical solution:

[0011] An adaptive production control system for discrete workshops;

[0012] It includes a workshop manufacturing system, a controller module, a performance analysis and strategy optimization module, and a workshop operation decision-making module;

[0013] The workshop manufacturing system includes a production planning layer, a task generation layer, a task input layer, and a task execution layer;

[0014] The controller module and the performance analysis and strategy optimization module are responsible for production control strategy planning, while the workshop operation decision module is embedded in the distributed intelligent manufacturing resources for distributed execution of control strategies.

[0015] The controller module includes a backlog task controller and a work-in-process controller. First, by periodically sampling and measuring the key parameters of each intelligent manufacturing resource in the workshop manufacturing system, the overall performance index of the workshop manufacturing system can be calculated. The actual calculation results are compared with the original planned performance index. If the actual performance index is lower than the expected index, the corresponding control strategy is selected to adjust the production capacity and order input rate.

[0016] The performance analysis and strategy optimization module is mainly used to perform simulation analysis and evaluate the potential performance of a production control system model before using a certain production control strategy. If the expected requirements of the workshop manufacturing system are met, the current control strategy is determined, and the parameters of the workshop manufacturing system are modified at an appropriate time to trigger the backlog task controller and work-in-process controller. If the expected performance is not met, the parameters of the backlog task controller and work-in-process controller need to be optimized again until the performance requirements of the workshop manufacturing system are met.

[0017] The workshop operation decision module decomposes control strategies, and each intelligent unit in the workshop organizes and implements or improves the production decision process. Based on the unit's own processing capacity and current status, it makes decisions on the tasks released to the unit, and the decision results include acceptance, pending, and rejection.

[0018] The workflow of the adaptive production control system in this discrete workshop is as follows: First, the task generation layer receives the order information issued by the production planning layer and completes the production planning and scheduling of the processing tasks; after generating the processing task information, the task is simultaneously sent to the task execution layer and the controller module; based on the task information, the current production control target is generated and sent to the controller module as a given value or planned value; under stable conditions, the workshop manufacturing system controls the workshop operation through the task execution layer; after a certain sampling period, the status information of the actual production process is updated, and the actual production information of the workshop is fed back to the production controller module.

[0019] Based on the above technical solution, the present invention can be further improved as follows.

[0020] Furthermore, the controller module internally calculates the error or deviation between the planned value and the actual production value. If the deviation is within the planned threshold range, the current production process remains unchanged. If the deviation exceeds the planned threshold range, the controller module generates a corresponding control strategy based on the corresponding production control objectives. The newly generated control strategy is first evaluated by the performance analysis module. If the updated control strategy meets expectations, adjustment measures are generated and implemented at the workshop level in the next adjustment cycle. If the expected performance is not met, the strategy is optimized again until the desired value is achieved.

[0021] The controller module needs to determine the optimal control decision point and make timely decisions for the subsequent workshop operation at that point in time. That is, it adopts appropriate control measures based on the operating status of the workshop manufacturing system. In the discrete workshop production process, the data acquisition terminal acquires a large amount of data from intelligent equipment, intelligent machine tools and other devices in the workshop through periodic sampling. The system characteristics are obtained through data preprocessing methods and time series analysis of the data.

[0022] Based on the above technical solution, the present invention can be further improved as follows.

[0023] Furthermore, the adaptive production control strategy of this discrete workshop's adaptive production control system, with the driving error as a function, is as follows:

[0024] u(t k )=f(e(t))e(t),t∈[t k ,t k+1 )

[0025] In the above formula, t k Let e(t) represent the driving time point, e(t) be the driving error, and f(e(t)) be the variable control gain as a function of the driving error.

[0026] To better understand the technical content of this invention, the adaptive production control system of this discrete workshop will be referred to as "this system" below.

[0027] The beneficial effects of this system are:

[0028] (1) A production control system architecture capable of adapting to the dynamic and ever-changing workshop operating environment was designed. The system architecture integrates centralized and decentralized event-driven mechanisms, enabling global "control" of production control strategies and autonomous execution at the workshop's underlying levels. This design allows the production control system to meet the stringent requirements of complex and dynamic workshop operating environments.

[0029] (2) At the same time, an adaptive control strategy with the driving error as a function was proposed, which improved the performance of the workshop system and reduced the frequency of the controller's operation, thus avoiding excessive adjustments by the workshop operation decision-making level. Attached Figure Description

[0030] Figure 1 This is a production planning and control system architecture diagram of a conventional production control system in existing technology.

[0031] Figure 2 This is a block diagram of the discrete manufacturing workshop production control system in an embodiment of the adaptive production control system for this discrete workshop.

[0032] Figure 3 This is a system technical architecture diagram of an embodiment of the adaptive production control system for this discrete workshop. Detailed Implementation

[0033] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0034] The terms “vertical,” “horizontal,” “left,” “right,” and similar expressions used in this document are for illustrative purposes only and do not represent the only possible implementation.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0036] The control objective model for production control systems can be divided into two dimensions: the operational performance of the workshop manufacturing system and the external demand response of the workshop manufacturing system. The objectives of production control are defined as: work-in-process and backlog tasks. The reasons for defining these two objectives are explained below.

[0037] (1) If the input rate of the order task flow remains constant, an excessively high level of work-in-process inventory will increase the production cycle. The drawback of the "push" production method is that there is a large amount of work-in-process in the conventional system, which leads to excessively high inventory costs and large fluctuations in production or processing takt time. The workshop has to extend the production cycle to compensate for the fluctuations in production takt time. On the other hand, in the conventional workshop manufacturing system of "pull", the number of work-in-process is relatively small.

[0038] (2) Work-in-process is easy to measure and can be calculated by measuring order quantity, while production cycle is difficult to measure and the actual production process is difficult to statistically analyze.

[0039] (3) The actual production cycle has certain limits, that is, the production cycle cannot be infinitely large, and its adjustment range is limited when analyzing the conventional workshop manufacturing system from the time domain perspective. Therefore, it is more practical to indirectly control the production cycle by controlling work-in-process.

[0040] (4) It helps improve equipment utilization. A certain amount of work-in-process is a necessary condition to ensure the continuity and stability of production. If the level of work-in-process is too low, it is easy to cause equipment to run out of materials and be in an "idle" state. At the same time, if the production rhythm between adjacent processes is inconsistent, interference will occur if there is no work-in-process. This will cause some equipment to be blocked by materials, while other equipment will be in a waiting state for materials, reducing equipment utilization and causing productivity loss.

[0041] (5) Excessive work-in-process inventory leads to capital accumulation, such as increased capital occupation time and storage site costs, which is not conducive to capital turnover.

[0042] (6) A reliable delivery probability is an effective measure to ensure customer satisfaction. By controlling backlog tasks, fast delivery can be guaranteed, thereby improving the ability to deliver on time.

[0043] Control strategies or methods are crucial in determining the operational mode of production control in a manufacturing system and serve as the starting point for production control system modeling and analysis. Based on the production control objectives of a discrete manufacturing workshop, the basic production control strategies are determined: order release strategy and production capacity adjustment strategy.

[0044] (1) Order release refers to controlling the actual flow of orders into the shop floor. For discrete manufacturing shops, in a conventional shop floor manufacturing system, for each order type, the parts pass through the shop floor according to a predefined processing sequence or process route, including the material flow path and processing information of the parts. To simplify the complexity of the problem, the model treats the flow of parts in the shop floor manufacturing system as a continuous flow model, that is, the problem of controlling order release evolves into controlling the size of the order flow. For example... Figure 2 As shown, by controlling the actual order input and the actual output of the workshop manufacturing system, it is possible to control the work-in-process inventory, production cycle, and equipment utilization in the workshop manufacturing system. Commonly used order release control strategies generally fall into the following categories:

[0045] ① Order release strategy based on delivery date;

[0046] ② Order release strategy based on work-in-process.

[0047] From a control theory perspective, a delivery-based order release strategy is an open-loop system. Order release begins on the planned start date, producing by "pushing" the order flow. This often results in a large amount of work-in-process (WIP) in conventional manufacturing systems. As mentioned earlier, if the system's order flow input rate remains constant, excessively high WIP levels will increase production cycle time. Therefore, a delivery-based order release strategy results in large production cycle fluctuations and is difficult to control. A WIP-based order release strategy, on the other hand, is a closed-loop system that uses WIP as a feedback loop, controlling the order flow input by monitoring the WIP inventory level.

[0048] (2) Production capacity control strategies mainly involve the adjustment of a company's production capacity over a future period, such as increasing work shifts and providing intensive training for on-the-job personnel to improve production capacity. Considering actual working conditions, the workshop manufacturing system generally requires a response (or delay) time when adjusting production capacity, and the response time varies depending on the method of production capacity adjustment. Therefore, the length of the workshop manufacturing system's production capacity adjustment response time reflects the flexibility of the workshop manufacturing system's production capacity. Production capacity control is generally divided into the following two types:

[0049] ① Production capacity control based on backlogged tasks;

[0050] ② Production capacity control oriented towards planning.

[0051] Ideally, the planned output of a shop floor manufacturing system matches its actual output. However, due to internal and external disturbances such as urgent orders and equipment failures, the system may fail to complete its tasks on time, resulting in actual output falling short of planned output and creating a backlog. Conversely, overproduction occurs when the actual output exceeds the planned output. Backlog-based capacity control adjusts production capacity by monitoring the amount of backlog to ensure the shop floor manufacturing system completes its tasks within the specified timeframe. By controlling backlog, delivery time accuracy is indirectly guaranteed, improving customer satisfaction.

[0052] Plan-oriented capacity control is a medium- to long-term capacity planning approach, involving manufacturing resources that require a considerable period to acquire. However, because shop floor manufacturing systems involve multiple production stages, capacity planning often exhibits dynamic uncertainty, making it difficult to balance predicted demand with actual capacity in plan-oriented capacity control.

[0053] The adaptive production control system of this discrete workshop includes a workshop manufacturing system, a controller module, a performance analysis and strategy optimization module, and a workshop operation decision module.

[0054] The workshop manufacturing system includes a production planning layer, a task generation layer, a task input layer, and a task execution layer;

[0055] The controller module and performance analysis and strategy optimization module are located at the upper layer, responsible for production control strategy planning. The lower-level workshop operation decision module is embedded in distributed intelligent manufacturing resources for distributed execution of the upper-level control strategy. The architecture of this discrete workshop adaptive production control system is as follows: Figure 3 As shown.

[0056] The controller module includes a backlog task controller and a work-in-process controller. First, by periodically sampling and measuring the key parameters of each smart manufacturing resource in the workshop manufacturing system, the overall performance index of the system can be calculated. The actual calculation results are compared with the original planned performance index. If the actual output is lower than expected (e.g., low actual output, backlog of tasks, which may lead to delays in future delivery), the corresponding control strategy is selected. In actual operation, the controller module will implement different control strategies according to different production control objectives, such as adjusting production capacity and order input rate.

[0057] The performance analysis and strategy optimization module is mainly used to perform simulation analysis and evaluate the potential performance of a given production control strategy using the adaptive production control system model of the discrete workshop before its implementation. If the expected requirements of the system are met, the current control strategy is determined, and the workshop manufacturing system parameters are modified at an appropriate time to trigger the backlog task controller and work-in-process controller. If the expected performance is not met, the parameters of the backlog task controller and work-in-process controller need to be optimized again until the system performance requirements are met.

[0058] The workshop operation decision module is equivalent to the executor of the adaptive production control system in this discrete workshop. It refers to the decision-making process by which intelligent units within the workshop organize, implement, or improve production by decomposing control strategies. For example, based on the unit's own processing capacity and current state, it makes decisions about the tasks released to the unit, and the decision results can include acceptance, pending, and rejection.

[0059] The workflow of the adaptive production control system in this discrete workshop is as follows: First, the task generation layer receives order information from the production planning layer, such as the ERP system, and completes the production planning and scheduling of processing tasks. After generating processing task information, the task is simultaneously sent to the task execution layer, such as the MES system, and the controller module. Based on the task information, the current production control target is generated and sent to the controller module as a given value or planned value. Under stable conditions, the workshop manufacturing system controls the workshop operation through the task execution layer (MES). After a certain sampling period, the status information of the actual production process is updated, and the actual production information of the workshop is fed back to the production controller module.

[0060] The controller module internally calculates the error or deviation between the planned value and the actual production value. If the deviation is within the planned threshold range, the current production process remains unchanged. If the deviation exceeds the threshold range planned by the upper layer, the controller module generates a corresponding control strategy based on the corresponding production control objectives. The newly generated control strategy is first evaluated by the performance analysis module. If the updated control strategy meets the expectations, adjustment measures are generated and deployed to the workshop level in the next adjustment cycle (or event trigger point). If the expected performance is not met, the strategy is optimized again until the expected value is achieved.

[0061] The workshop's operational decision-making is the executor of the control system. By establishing decision-making mechanisms for manufacturing and logistics units at all levels of the workshop and embedding them into industrial control computers, intelligent units are constructed to match equipment processing capacity with logistics processes, optimize workshop operation processes, and complete control commands issued by the controller module.

[0062] To ensure optimal performance of the workshop manufacturing system, the controller module needs to determine the optimal control decision point and make timely decisions for subsequent workshop operations at that point. This means adopting appropriate control measures based on the operating status of the workshop manufacturing system, such as work-in-process level and load rate. In discrete workshop production, data acquisition terminals periodically sample large amounts of data from intelligent equipment and machine tools within the workshop. System characteristics are obtained through data preprocessing and time-series analysis. System characteristics refer to the state variables that characterize the operation of the workshop manufacturing system, and their values ​​must be easily measurable. This research focuses more on the steady-state mean of the system characteristics within the control cycle, rather than the change process of the state characteristics throughout the entire control cycle.

[0063] For workshop manufacturing systems, if the system stabilizes after a period of operation, the data acquisition terminal continues to transmit signals, and the upper-level controller module still performs periodic drives, leading to frequent changes in the workshop operation decision-making layer. This inevitably wastes computing space and overloads communication resources. In reality, the manufacturing units, logistics units, and other intelligent agents within the workshop layer are mostly embedded industrial control computers. Frequent data reading and writing inevitably keep these agents in a busy state. To avoid unnecessary adjustments by the workshop operation decision-making layer and reduce the system's operating frequency, this system adopts an event-driven control mechanism to trigger control strategies from the control strategy layer to the workshop decision-making layer.

[0064] The system controller module updates due to event-driven mechanisms. An event refers to the difference between the measured value and the ideal value of a characteristic of the workshop manufacturing system reaching a boundary. Currently existing event-driven mechanisms include centralized and decentralized mechanisms, and can be further classified into time-dependent and state-dependent types based on threshold values.

[0065] A centralized event-driven mechanism refers to a system where all intelligent units within the workshop manufacturing system only synchronize and update when the global performance indicator measurement error is greater than or equal to the threshold set by the control strategy. The global performance indicator error can be defined as e(t) = x(t). k )-x(t),t∈[t k ,t k+1 ), where t k This indicates the driving time point. Its control strategy can be expressed as:

[0066] u(t k )=ke(t),t∈[tk ,t k+1 (1)

[0067] As the event occurs—that is, the deviation from the production control target exceeds the threshold—the adaptive production control system of this discrete workshop generates a series of control action points (in actual production, these points are defined as workdays or production cycles). The system controller's input is updated at each control action point. Within the action cycle, the controller's input remains unchanged.

[0068] u(t)=u(t k ),t∈[t k ,t k+1 (2)

[0069] Among them, t k k = 0, 1, 2, L represents a series of control action time points;

[0070] In contrast, the decentralized event-driven mechanism does not require the system controller to update only when the deviation of the control target of the workshop manufacturing system is greater than or equal to the threshold. It only needs to drive the update based on local information, that is, based on the relevant information of the relevant intelligent units. The control time point and the frequency of action of each intelligent unit are different.

[0071] If we define the driving error e of the intelligent unit at the workshop level as changing over time... i (t)=x i (t k )-x i (t), t∈[t k ,t k+1 When considering the workshop manufacturing system as a linear time-definite system:

[0072]

[0073] Where, x i (t)∈R n u i (t)∈R n These represent the state of each intelligent unit and the controller input, respectively; A∈R n×n , B∈R n×p The state matrix and input matrix are represented in separate tables; the control strategy can be:

[0074]

[0075] In the formula, K1,K2∈R n×n This is the corresponding production control target gain matrix.

[0076] In actual production, manufacturing system performance optimization places greater emphasis on economical and stable operation, focusing on overall optimal performance. Upper-level control strategies often adopt a global approach to manage the manufacturing system. Therefore, this system employs a centralized event-driven mechanism to link the control strategy planning layer and the shop floor operation decision-making layer; the shop floor decision-making layer uses a decentralized event control mechanism to construct an autonomous decision-making mechanism for intelligent units within the shop floor.

[0077] When introducing event-driven mechanisms into a workshop manufacturing system composed of multiple intelligent units, the first thing to consider is the threshold issue of the control action point. The wider the range of the threshold setting, the fewer times the controller updates, and the lower the control frequency; conversely, the narrower the threshold setting, the higher the control frequency.

[0078] Regarding the impact of control frequency on the dynamic behavior of manufacturing systems, excessively low control frequencies are detrimental to the stability of the manufacturing system. Conversely, when the control frequency reaches its maximum value, the manufacturing system changes from discrete control to continuous control, which improves the performance of the manufacturing system but increases the system burden.

[0079] The adaptive production control strategy of this discrete workshop, which uses the driving error as a function, is as follows:

[0080] u(t k )=f(e(t))e(t),t∈[t k ,t k+1 (5)

[0081] In the above formula, e(t) is the driving error; f(e(t)) is the variable control gain as a function of the driving error.

[0082] The above description is only one embodiment of the present invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the principle of the present invention, and these should also be considered to fall within the protection scope of the present invention.

Claims

1. An adaptive production control system for a discrete workshop, characterized by: It includes a workshop manufacturing system, a controller module, a performance analysis and strategy optimization module, and a workshop operation decision-making module; The workshop manufacturing system includes a production planning layer, a task generation layer, a task input layer, and a task execution layer; The controller module and the performance analysis and strategy optimization module are responsible for production control strategy planning, while the workshop operation decision module is embedded in the distributed intelligent manufacturing resources for distributed execution of control strategies. The controller module includes a backlog task controller and a work-in-process controller. First, by periodically sampling and measuring the key parameters of each intelligent manufacturing resource in the workshop manufacturing system, the overall performance index of the workshop manufacturing system can be calculated. The actual calculation results are compared with the original planned performance index. If the actual performance index is lower than the expected index, the corresponding control strategy is selected to adjust the production capacity and order input rate. The performance analysis and strategy optimization module is mainly used to perform simulation analysis and evaluate the potential performance of a production control system model before using a certain production control strategy. If the expected requirements of the workshop manufacturing system are met, the current control strategy is determined, and the parameters of the workshop manufacturing system are modified at an appropriate time to trigger the backlog task controller and work-in-process controller. If the expected performance is not met, the parameters of the backlog task controller and work-in-process controller need to be optimized again until the performance requirements of the workshop manufacturing system are met. The workshop operation decision module decomposes control strategies, and each intelligent unit in the workshop organizes and implements or improves the production decision process. Based on the unit's own processing capacity and current status, it makes decisions on the tasks released to the unit, and the decision results include acceptance, pending, and rejection. The workflow of the adaptive production control system in this discrete workshop is as follows: First, the task generation layer receives the order information issued by the production planning layer and completes the production planning and scheduling of the processing tasks; after generating the processing task information, the task is simultaneously sent to the task execution layer and the controller module; based on the task information, the current production control target is generated and sent to the controller module as a given value or planned value; under stable conditions, the workshop manufacturing system controls the workshop operation through the task execution layer; after a certain sampling period, the status information of the actual production process is updated, and the actual production information of the workshop is fed back to the production controller module.

2. The adaptive production control system for discrete workshops according to claim 1, characterized in that: The controller module internally calculates the error or deviation between the planned value and the actual production value. If the deviation is within the planned threshold range, the current production process remains unchanged. If the deviation exceeds the planned threshold range, the controller module generates a corresponding control strategy based on the corresponding production control objectives. The newly generated control strategy is first evaluated by the performance analysis module. If the updated control strategy meets the expectations, adjustment measures are generated and implemented at the workshop level in the next adjustment cycle. If the expected performance is not met, the strategy is optimized again until the expected value is achieved. The controller module needs to determine the optimal control decision point and make timely decisions for the subsequent workshop operation at that point in time. That is, it adopts appropriate control measures based on the operating status of the workshop manufacturing system. In the discrete workshop production process, the data acquisition terminal acquires a large amount of data from intelligent equipment, intelligent machine tools and other devices in the workshop through periodic sampling. The system characteristics are obtained through data preprocessing methods and time series analysis of the data.

3. The adaptive production control system for discrete workshops according to claim 1, characterized in that: The adaptive production control strategy of this discrete workshop, which uses the driving error as a function, is as follows: u(t k )=f(e(t))e(t),t∈[t k ,t k+1 ) In the above formula, t k Let e(t) represent the driving time point, e(t) be the driving error, and f(e(t)) be the variable control gain as a function of the driving error.

Images