Production line digital superposition control method, device and equipment and storage medium

Abstract

The invention relates to the technical field of production line control, and discloses a production line digital superposition control method, device and equipment and a storage medium, and the method comprises the steps: carrying out the parameter collection and data preprocessing of a production line, obtaining the state information of a production process, and determining a first control parameter of glue superposition positioning; creating a decision layer and an execution layer of production line superposition control, and determining a communication connection structure of each process control unit; based on the communication connection structure, superposition error information fusion processing is executed in the decision layer, and an optimal glue positioning control instruction is generated; the optimal glue positioning control instruction is transmitted to an execution layer, a glue spraying unit, a centering unit and a superposition unit are driven to execute glue superposition operation, and execution state data are collected; and according to the execution state data, the first control parameter is dynamically adjusted to obtain the second control parameter, so that the adaptation period during process switching is greatly shortened, and the flexibility and efficiency of the production line are improved.

Description

Technical Field

[0001] This application relates to the field of production line control technology, and in particular to a digital overlay control method, device, equipment and storage medium for production lines. Background Technology

[0002] Traditional production line control methods primarily rely on simple PID control or open-loop control strategies. When faced with complex factors such as multi-process coupling, time-varying delays, and environmental disturbances, these methods often struggle to meet the high-precision glue overlay positioning requirements. In particular, when production line speed changes or process parameters are adjusted, traditional control methods require a long adaptation period, leading to product quality fluctuations and increased material waste. Furthermore, in existing technologies, each process control unit typically operates independently, lacking effective information sharing and collaborative decision-making mechanisms, making it difficult to establish a unified overlay control strategy for the entire production line.

[0003] Existing digital control systems for production lines generally employ fixed delay compensation schemes to address inter-process time delays, which are unable to adapt to dynamic changes in the production process. When switching processes, the system requires manual recalibration of parameters, increasing debugging time and relying heavily on the operator's experience. Furthermore, existing systems primarily handle glue positioning errors through reactive compensation rather than predictive control, resulting in lag in system response and making it difficult to meet the real-time control requirements of high-speed production lines. Particularly in multi-layer material stacking processes, the cumulative effect of errors is significant, and traditional single-process control methods cannot effectively address the coordinated optimization of stacking position accuracy and overlap. Summary of the Invention

[0004] This application provides a digital overlay control method, device, equipment, and storage medium for production lines, which significantly shortens the adaptation cycle during process changeovers and improves the flexibility and efficiency of the production line.

[0005] The first aspect of this application provides a digital overlay control method for a production line, the method comprising: The production line is sampled and preprocessed to obtain production process status information, and the first control parameter for glue superposition is determined based on the production process status information. The decision-making and execution layers of the production line overlay control are created using the first control parameters, and the communication connection structure of each process control unit is established. Based on the aforementioned communication connection structure, superimposed error information fusion processing is performed in the decision layer to generate the optimal glue positioning control command. The optimal glue positioning control command is transmitted to the execution layer to drive the glue spraying unit, centering unit and lamination unit to perform glue superposition operation, and to collect execution status data; Based on the execution status data, the first control parameter is dynamically adjusted to obtain the second control parameter.

[0006] A second aspect of this application provides a digital overlay control device for a production line, the digital overlay control device comprising: The acquisition module is used to acquire parameters and preprocess data of the production line to obtain production process status information, and determine the first control parameter for glue superposition based on the production process status information. A module is created to use the first control parameters to create the decision layer and execution layer of the production line superimposed control, and to establish the communication connection structure of each process control unit. The fusion module is used to perform superimposed error information fusion processing in the decision layer based on the communication connection structure to generate the optimal glue positioning control command; The operation module is used to transmit the optimal glue positioning control command to the execution layer, drive the glue spraying unit, the centering unit and the overlapping unit to perform glue superposition operation, and collect execution status data; The adjustment module is used to dynamically adjust the first control parameter based on the execution status data to obtain the second control parameter.

[0007] A third aspect of this application provides an electronic device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the electronic device to execute the above-described production line digital overlay control method.

[0008] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described production line digital overlay control method.

[0009] Compared with existing technologies, this application has the following advantages: By constructing an adjustable Lyapunov stability matrix and variable enhancement matrix method for process intervals, the nonlinear delay control problem is solved, significantly improving glue positioning accuracy and ensuring product quality and functional stability. Based on a double-layer nested digital superposition control framework, the separation and collaboration of the decision-making layer and the execution layer are realized, enabling each process control unit to share information and coordinate actions, overcoming the limitations of independent operation of each process in traditional control methods. Through fixed-time sliding mode control law and event triggering mechanism, the system can converge within a predetermined time, significantly shortening the adaptation cycle during process switching and improving the flexibility and efficiency of the production line. By adopting a distributed optimization algorithm and dynamic learning rate mechanism, the computational burden of the system is reduced, enabling control resources to be intelligently allocated according to process requirements, avoiding resource waste and system overload. Through a fuzzy logic system and a fast fixed-time filter, interference from unknown production lines and discontinuity of execution signals are effectively compensated, enhancing the robustness of the system in complex process environments. Based on a multi-objective optimization algorithm and a three-stage verification and testing mechanism, the system can simultaneously optimize two key indicators: glue positioning accuracy and superposition overlap, overcoming the limitations of single-parameter optimization in traditional methods. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] The structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0012] Figure 1 This is a flowchart illustrating the digital overlay control method for production lines provided in an embodiment of the present invention; Figure 2 This is a schematic block diagram of the structure of the digital overlay control device for production lines provided in an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of the electronic device provided in the embodiment of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0015] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0016] It should also be further understood that the term "and / or" as used in this application specification and the appended claims refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes such combinations. See also Figure 1 One embodiment of the digital overlay control method for production lines in this application includes: Step 100: Collect parameters and preprocess data for the production line to obtain production process status information, and determine the first control parameter for glue superposition based on the production process status information; It is understood that the executing entity of this application can be a digital overlay control device for a production line, or it can be a terminal or a server; no specific limitation is made here. This application's embodiment uses a server as an example for illustration.

[0017] Specifically, multiple sensor nodes are deployed on the production line to ensure full coverage of key processes. These sensor nodes are arranged sequentially according to the process flow, covering all critical control points from the front end of the glue spraying machine, the upstream and downstream of the centering mechanism, the front and rear ends of the drying section, to the stacking and pressing mechanism, forming a sensor network system for continuously collecting status data from each process node throughout the entire production process. Each sensor node collects the current spatial position information of the workpiece, the production line speed data, and contact pressure data related to glue application and stacking, forming a raw dataset covering position, speed, and mechanical state. The raw dataset undergoes linear calibration to correct for biases, nonlinear drift, and environmental disturbances in various sensors, ensuring consistency in numerical scale and physical meaning of all sensor output data, thereby improving data comparability and subsequent usability. After calibration, the workpiece transfer process between different nodes is analyzed. Based on the actual measured positional relationships, the current production line speed, and the influence of temperature in the workshop environment on the movement process, the time required for each workpiece to move from one process node to another is calculated. By organizing the transmission times between all nodes into a matrix, a time delay structure between processes is constructed. This structure characterizes the flow rhythm and cycle control requirements of materials throughout the production line. Based on this time delay structure, key time delay parameters are extracted, such as the maximum delay, minimum delay, and delay fluctuation range between different nodes, forming an understanding of the overall dynamic characteristics of the system. Simultaneously, regarding process characteristics, key parameters of the adhesive are integrated, including its flow viscosity, curing time, coating thickness, and temperature sensitivity. These parameters, along with the transmission status and time delay parameters, affect the accuracy and stability of the adhesive layering process. These processed data are then comprehensively integrated to construct a global dataset describing the current production process status. This dataset covers multiple dimensions, including dynamic positional relationships, time synchronization mechanisms, and material response behavior. Based on this, the first control parameters required for controlling the adhesive layering positioning are extracted, including the target positioning position of the adhesive, the spraying start and end time windows, the synchronization and coordination relationships between processes, and the error tolerance boundary.

[0018] In this embodiment, production process state information is input into the system modeling module. Using system state-space modeling, a production line superimposed control model covering multiple processes such as glue spraying, alignment, drying, and pressing is established. This model uses key variables such as position, speed, and pressure as state dimensions, comprehensively considering the dynamic response characteristics and time-series dependency structure of each process control node to form a dynamic control system framework. To improve the accuracy of process delay characterization, the delay intervals between processes in the model are divided, and the range of delay variation, rate boundaries, and dynamic fluctuation characteristics are extracted to constitute a delay characteristic description of the system, ensuring that the modeling process has good adaptability and predictive ability to actual production fluctuations. Based on the above delay characteristic description, an adjustable process interval stability analysis mechanism is introduced. Within the control theory framework, a Lyapunov stability matrix structure is constructed to support dynamic scheduling and state coordination. By setting reasonable state dependency functions and energy decay mechanisms, the entire system can still maintain convergence and stability under conditions of uncertainty and time-varying disturbances. The stability structure is solved using a constrained optimization method to obtain a combination of control matrices that satisfy the stability conditions. This combination includes gain parameters for state feedback, dynamic adjustment coefficients related to delay compensation, and response adjustment factors under control action, forming a set of stable control matrices with physical constraints. The stability control matrix set is then transformed into control reference quantities that can be directly used in actual glue spraying operations. Parameterization is performed to give the matrix elements clearly defined process meanings and operable numerical ranges, resulting in a set of glue positioning parameters covering spraying position, angle, start-up time, nozzle pressure, and dynamic tracking response speed. To ensure the practicality and precision of the extracted parameters in actual control processes, the glue positioning parameter set is combined with the previously constructed inter-process time delay matrix, and a mapping relationship between glue superposition accuracy and control parameters is established. This forms a parameter sensitivity model to describe the impact of changes in control variables on spraying accuracy. This model can effectively quantify the impact of any control parameter on superposition error, time drift, and spraying consistency under different conditions. Based on this, and combined with actual process requirements, target values ​​for glue positioning accuracy and superposition overlap are set as constraints for control system optimization. These target parameters are solved in conjunction with the sensitivity model. Through inverse adjustment and iterative approximation, the first set of control parameters that can achieve stable and efficient glue spraying under the premise of meeting all performance indicators is calculated.

[0019] Step 200: Use the first control parameters to create the decision layer and execution layer of the production line overlay control, and establish the communication connection structure of each process control unit; Specifically, based on the first control parameter, the entire production line is logically divided and its control architecture is planned. In this process, according to sensor layout, control response frequency, process complexity, and accuracy requirements, the entire production line is divided into several control unit nodes. Each node represents a specific control entity or control area, and is assigned different control functions according to its functional positioning within its section, constructing a logically meaningful set of control nodes that collectively constitute the decision layer for the overlay control of the production line. This decision layer is primarily responsible for generating control commands, fusing state estimates, and calculating command optimization strategies, possessing high decision weight and system coordination capabilities. Simultaneously, all actual physical actuators on the production line are functionally grouped, divided into four functional modules based on their execution tasks: a glue spraying unit, a conveying unit, a centering unit, and a lamination unit. The actuator types and controlled objects included in each module are further refined. For example, the glue spraying unit includes a nozzle array, a pressure control module, and a flow regulation mechanism; the conveying unit includes a belt speed regulating motor and a trajectory correction device; the centering unit is responsible for positioning calibration; and the lamination unit performs multi-layer material pressing operations. These actual controlled objects collectively constitute the execution layer of the production line. A mapping relationship between the decision-making layer and the execution layer is established. Each physical execution unit is matched with its corresponding control command generation unit, that is, a mapping function is established in the logical control space. This ensures that the commands output by each node in the decision-making layer can be accurately transmitted to its corresponding execution device, and the update frequency, priority, and granularity of the commands can be dynamically adjusted according to the current operating conditions. This establishes an inter-layer information transmission channel between decision output and physical response in the overall system. This channel has command delivery function and also undertakes closed-loop control tasks such as status feedback, data return, and strategy correction. Due to differences in geographical location, transmission medium, and control frequency between different control units, information transmission delays and buffer fluctuations will occur during inter-layer information transmission. To address this, delay compensation and information buffering mechanisms are introduced. Buffers and delay compensation modules are introduced into the communication path to automatically calibrate the command delivery timing and receiving response window based on the historical delay characteristics of each channel, ensuring synchronous control and coordination consistency under multi-channel collaboration. Ultimately, a digital overlay control framework with a two-layer structure, clear division of labor, precise mapping, and efficient communication capabilities is formed, driven by the first control parameter.

[0020] Step 300: Based on the communication connection structure, perform superimposed error information fusion processing in the decision layer to generate the optimal glue positioning control command; It should be noted that a variable enhancement matrix containing the current state, historical delayed states, and state accumulation is constructed at each control node of the decision-making layer. This matrix integrates information from three dimensions—real-time perception, time lag, and state accumulation—into a unified expression structure, comprehensively representing the current operating status and dynamic trends of the process unit. Based on this enhancement matrix, a zero-sum relationship setting method is introduced. By setting a constraint that the sum of the state changes of each control unit is zero, the originally complex system dynamic relationship containing nonlinear delay derivatives is transformed into a simple and highly controllable linearized control model. Based on the linearized control model, the state estimators distributed at each control node are updated. Each control unit interacts and merges its own state with the information of neighboring units, dynamically adjusting its own state estimate to form a preliminary set of control parameters reflecting the current glue error state. To accelerate this estimation convergence process and improve the response capability of the control system under high-frequency scheduling, a momentum term mechanism is introduced into the update algorithm. By incorporating the state change trend of the previous moment into the current update, the directionality and inertia of the control parameter adjustment are enhanced, thereby suppressing estimation oscillations. Meanwhile, to adapt to different process cycles and error fluctuations, an adaptive learning rate mechanism is introduced. The learning rate is dynamically adjusted based on the changing trend of the estimated gradient, accelerating parameter updates when errors are large and gradually slowing down as the system approaches a stable state. This improves convergence speed while avoiding overshoot or instability. The control value, after joint adjustment of momentum and learning rate, can quickly approach the optimal solution and maintain stable updates in dynamic environments. As the control value iterates and converges, it is compared with the set error threshold and process change threshold to determine whether the current glue positioning error exceeds the allowable range or whether the process conditions have changed significantly. If any judgment result exceeds the threshold, the optimized information interaction strategy is activated, triggering the retransmission of the target state and a global information sharing mechanism in the control network. This prevents the system from falling into a state of misjudgment or control failure due to local anomalies. After the information interaction strategy is activated, each control node synchronously updates its own state estimate and re-integrates the collaborative target. Finally, the system generates a set of globally consistent optimal glue positioning control instructions, which include refined parameters such as the target spraying position, start time, and intensity of action for each execution unit under the current production cycle.

[0021] Step 400: Transmit the optimal glue positioning control command to the execution layer to drive the glue spraying unit, centering unit and overlapping unit to perform glue superposition operation, and collect execution status data; Specifically, a unified dynamic control model is constructed for the glue spraying unit, centering unit, and overlapping unit of the execution layer. Based on the physical process, each unit is modeled as a second-order differential control system with position and velocity states. This modeling method can comprehensively describe the dynamic response characteristics exhibited by the equipment during execution, including the accuracy of motion positioning, the delay of action switching, and the sensitivity to disturbances. On this basis, taking the deviation between the desired trajectory and the current state as the core, the current position error and velocity error are combined, and design parameters such as adjustment coefficients are introduced to construct a system trajectory convergence standard with dual constraints of convergence and stability. This standard ensures that the response curve of each actuator in the dynamic process can stably and predictably approach the target trajectory within a limited time. Around this convergence standard, a precise control law is constructed, which combines the proportional adjustment term with a power function term with nonlinear characteristics. An adaptive gain mechanism set for different execution objects is introduced, enabling the controller to maintain high-precision adjustment capability under conditions of load changes, velocity disturbances, or environmental fluctuations. This control law possesses sufficient response strength, and its nonlinear structure endows it with strong sensitivity, enabling it to quickly capture and correct minute errors in a timely manner. Before generating the control command, a third-order low-pass filter is applied to it. This process suppresses abrupt changes in the control signal caused by high-frequency interference in the state feedback, ensuring good continuity and smoothness in the control action output process and avoiding severe jitter or overload of actuators. The filtered control action sequence is distributed to each execution unit, and through a drive signal adaptation mechanism, the logic control command is converted into a signal format compatible with the physical interfaces of the glue spraying valve, centering motor, and pressing cylinder. During this process, the system considers the communication protocols, voltage requirements, pulse structures, and response delay characteristics of different actuators, ensuring that the control signal can be correctly decoded at the physical layer and trigger the corresponding action command. The glue spraying unit can thus precisely control the nozzle pressure, opening duration, and spraying position; the centering unit can adjust the translation and rotation of the workpiece on the production line accordingly; and the lamination unit precisely controls the pressing sequence, cylinder stroke, and application force. While each unit completes its actions according to the control instructions, the key execution status data of the entire glue layering process are collected in real time through the sensor network deployed in the early stage. This includes the spraying position and spraying trajectory of the glue spraying unit, the positioning accuracy of the centering unit, and the pressure changes and operational stability of the lamination unit at different pressing stages. These data constitute the execution status feedback set.

[0022] Step 500: Based on the execution status data, dynamically adjust the first control parameter to obtain the second control parameter.

[0023] Specifically, a comprehensive evaluation standard system covering multiple dimensions of system performance is constructed. This system includes measurement methods and target value settings for key indicators during adhesive spraying. Its main components encompass four aspects: positioning error, superposition overlap, control response time, and resource utilization. Positioning error measures the spatial deviation between the actual adhesive spraying position and the target position. Superposition overlap reflects the spatial overlap ratio of different adhesive layers. Control response time assesses the total time required for the system to complete a physical action from receiving a control command. Resource utilization examines the efficiency of the entire system in utilizing resources such as energy, processor power, and communication bandwidth during execution. Using the state data collected during execution, the key values ​​required for each evaluation are extracted sequentially. By comparing the actual adhesive spraying position with the position coordinates set in the control command, the spatial deviation value is calculated as a positioning accuracy indicator. Simultaneously, image recognition, edge matching, or sensor feedback technologies are used to calculate the ratio between the effective overlap area and the theoretical overlap area between different material layers, obtaining a quantitative result for superposition overlap. These two results, along with execution timing data and resource scheduling logs, are input into the performance analysis module to construct a performance evaluation vector reflecting the current operating conditions. This vector encompasses the system's real-time operating status, and its difference from the preset ideal performance target value during system initialization is calculated item by item. Then, weight coefficients are assigned according to the importance of each evaluation indicator, and a weighted summary of multiple indicators is completed to obtain a unified system performance deviation index, which quantifies the degree of deviation between the current control strategy and the optimal target. Based on this performance deviation index, a multi-objective optimization mechanism is initiated. Multiple adjustment dimensions are defined based on the first control parameter, and a candidate control parameter set is generated. This process employs an improved particle swarm optimization algorithm, differential evolution algorithm, or other problem solvers suitable for multivariate complex optimization, using minimizing performance deviation as the objective function to search for a new optimal solution in the parameter space. After obtaining the candidate control parameter set, unit-level testing, system-level testing, and production-level testing are carried out sequentially according to a bottom-up verification path. In the unit-level testing phase, the performance of each parameter combination in a single functional module (such as the glue spraying unit or the lamination unit) is verified to meet the basic responsiveness and stability requirements. In the system-level testing phase, the consistency and real-time response of the parameter combinations under multi-module collaborative operation are examined. In the production-level testing phase, the system runs the selected parameters under real production conditions, continuously records its performance indicators, and compares them with a preset baseline to ensure that it maintains the target performance during long-term operation. The test results are combined with the original performance deviation index for analysis to identify the parameter combination that performs optimally in terms of error, efficiency, and stability. Then, the initially set first control parameters are structurally modified and numerically updated to generate the final version of the second control parameters.

[0024] In this embodiment, the nonlinear delay control problem is solved by constructing an adjustable Lyapunov stability matrix and variable enhancement matrix method, significantly improving glue positioning accuracy and ensuring product quality and functional stability. Based on a double-layer nested digital overlay control framework, the separation and collaboration between the decision-making and execution layers are achieved, enabling control units of each process to share information and coordinate actions, overcoming the limitations of independent operation of each process in traditional control methods. Through fixed-time sliding mode control and event triggering mechanisms, the system can converge within a predetermined time, significantly shortening the adaptation cycle during process switching and improving the flexibility and efficiency of the production line. The use of distributed optimization algorithms and dynamic learning rate mechanisms reduces the computational burden of the system, allowing control resources to be intelligently allocated according to process requirements, avoiding resource waste and system overload. Through fuzzy logic systems and fast fixed-time filters, interference from unknown production lines and discontinuities in execution signals are effectively compensated, enhancing the system's robustness in complex process environments. Based on multi-objective optimization algorithms and a three-stage verification and testing mechanism, the system can simultaneously optimize two key indicators: glue positioning accuracy and overlay overlap, overcoming the limitations of single-parameter optimization in traditional methods.

[0025] In one specific embodiment, the process of performing step 100 may specifically include the following steps: Multiple sensor nodes are arranged on the production line, and the layout of the sensor nodes is configured to obtain a sensor network covering the front of the glue spraying machine, the centering mechanism, the front and rear of the dryer, and the superimposed pressing mechanism. The raw dataset of each sensor node is collected through a sensor network. The raw dataset includes workpiece position data, production line speed data, and actuator pressure data. The original dataset is linearly calibrated to obtain calibrated process data; The material transfer time between each sensor node is calculated based on the calibrated process data, and an inter-process time delay matrix is ​​constructed based on physical distance, production line reference speed, and temperature compensation coefficient. Delay analysis is performed on the time delay matrix between processes to obtain system time delay characteristic parameters. The calibrated process data is then combined with the system time delay characteristic parameters, and the glue characteristic parameter matrix is ​​integrated to obtain production process status information. Based on the production process status information, determine the glue positioning parameters and the inter-process delay relationship, and establish the first control parameter for glue superposition positioning.

[0026] Specifically, a multi-sensor layout is implemented around the key process nodes of the entire production line. During the layout design, the front section of the glue spraying machine, the inlet and outlet of the centering mechanism, the front and rear ends of the dryer, and the stacking and pressing mechanism are designated as the main control sections. Position sensors, speed detection devices, and pressure sensors are deployed in these areas to comprehensively perceive and capture the spatiotemporal dynamic characteristics of the entire production line. By planning the placement, coverage, sampling frequency, and data interfaces of these sensors, a multi-source sensing network is established with nodes as the core and processes as the chain. Once the production line is running, each sensor node collects raw data in real time, forming a raw dataset indexed by timestamps. This dataset covers three core types of information: the two-dimensional position coordinates of the workpiece at each control node, used to record the dynamic trajectory and spatial offset of the workpiece; instantaneous speed data of the production line, used to characterize the rhythm and overall flow rate characteristics of material movement; and pressure data connected to each actuator, reflecting the contact strength and feedback between the workpiece and the control device. A unified linear calibration mechanism is introduced to standardize and correct the scale of all collected data. By setting the gain coefficients and bias parameters for various sensors, a consistent numerical benchmark is achieved for similar data across different work sections, resulting in calibrated process data. Production cycle time modeling is then performed based on this calibrated process data. The material flow paths between sensor nodes are identified, and the material transfer time between any two nodes is calculated. Dynamic derivation is performed based on the physical distance between nodes, the current production line's baseline operating speed, and temperature variations in the workshop environment. The temperature compensation factor directly affects the viscosity change of the adhesive, the thermal expansion coefficient of the conveyor belt, and the thermal drift of the drive mechanism, thus indirectly affecting the actual arrival time of the material. The transfer times between nodes are arranged according to the process sequence to form a symmetric or asymmetric time delay matrix. This matrix records the cycle time coordination relationship and process sequence within the production line, serving as the core data structure describing time characteristics and control scheduling constraints. Based on the time delay matrix, delay analysis is performed. By calculating indicators such as maximum delay, minimum delay, rate of change of delay, and average cycle time difference, the overall time delay characteristic parameters of the system are obtained, reflecting the responsiveness of the control system under different operating conditions and demonstrating the compactness and stability of inter-process collaborative control. On this basis, the calibrated process data and time delay characteristic parameters are combined in multiple dimensions to form a structured data vector describing the three-dimensional state of the production process in space, time, and control. To more comprehensively reflect the impact of material properties and environmental coupling on the control strategy during production, an adhesive characteristic parameter matrix is ​​introduced. This matrix includes four core indicators: initial viscosity, curing time, coating thickness, and temperature sensitivity, effectively characterizing the adhesive's behavioral response under different temperature and humidity conditions and operating intensities. After completing the above data integration, a set of production process state information covering motion behavior, time characteristics, and material properties is obtained.Based on this information set, control parameters for precise glue positioning and timing coordination are derived. Glue positioning parameters are extracted from the deviation between the workpiece trajectory and the desired trajectory. These parameters include the target value of the spraying position coordinates, the maximum allowable error limit, and the corresponding control gain factor. Then, based on the time delay matrix and time delay characteristics, the synchronization trigger points and optimal timing ratios between each process are extracted, forming a stable and efficient inter-process delay relationship. These spatial coordinate, time synchronization, and control response parameters are integrated into a set of control inputs with real-time, adaptive, and controllable characteristics. This parameter set constitutes the first control parameter for glue overlay positioning.

[0027] In one specific embodiment, the process of determining the glue positioning parameters and inter-process delay relationships based on the production process status information, and establishing the first control parameters for glue overlay positioning, can specifically include the following steps: A system state-space model is performed on the production process state information to obtain a superimposed control model for the production line; The delay intervals between processes in the superimposed control model of the production line are divided to obtain a description of the delay characteristics; Based on the description of delay characteristics, an adjustable Lyapunov stability matrix is ​​constructed, and a set of stability control matrices is obtained by solving linear matrix inequalities. The stability control matrix group is parameterized to obtain the glue positioning parameter set; Based on the glue positioning parameter set and the inter-process time delay matrix, a mapping relationship between glue superposition accuracy and control parameters is established, and a parameter sensitivity model is obtained. Based on the parameter sensitivity model, the target values ​​for glue positioning accuracy and superposition overlap are set as constraints, and the first control parameter for glue superposition positioning is calculated.

[0028] Specifically, multi-dimensional process data, encompassing spatial location, velocity state, pressure feedback, time delay, and material properties, are integrated into a set of dynamic state variables. The interactions between these variables are clearly defined, forming a system state-space representation structure suitable for industrial control scenarios. During modeling, the states of each control node on the production line are represented as time functions. The physical state changes of the workpiece at each stage, such as adhesive spraying, alignment, drying, and lamination, are linked to control behavior. Structurally, a coupling relationship is introduced between state vectors, input vectors, and output vectors, allowing the dynamic behavior of the entire production line to be mapped to continuous changes in state variables and input-driven effects. This state-space model reflects the direct relationships between physical quantities and also includes factors such as control delay, disturbance response, and state feedback delay. The inter-process delay intervals in the production line superimposed control model are divided. Because the time it takes for materials to transition from one state to another between different process nodes is fluctuating and uncertain, this time lag phenomenon is divided into time delay intervals with upper and lower limits and changing trends during the modeling phase. By statistically fitting and dynamically analyzing historical production data, the maximum transmission time, minimum response time, and average cycle time difference between each process pair are extracted and further summarized into a delay interval feature vector, which is then incorporated into the model as a description of delay characteristics. Based on this delay characteristic description, an adjustable Lyapunov stability matrix for process intervals is constructed. This structure uses the rate of change of process state variables and the rate of decay of the energy function as core parameters. By constructing a stability function that monotonically decreases in any non-zero state, it ensures that the system remains asymptotically stable even under time lag conditions. To make this stability structure practically feasible, it is transformed into a set of linear matrix inequalities and solved using numerical optimization methods. Through iterative approximation, constraint convergence, and matrix expansion, a set of feasible stability control matrices is obtained. This set of matrices includes various parameters such as feedback gain, modulation factor, and dynamic adjustment coefficient, used to control the response behavior of each execution unit under input disturbances and feedback delay conditions. The stability control matrix group undergoes parameterization, which involves restoring its dimensions, normalizing it, and mapping its physical meaning to give it practical control significance that can be directly used for spraying task scheduling and action execution. The processed result is the glue positioning parameter set, which specifies the target parameters for the actual drive controller, including spraying position correction, time response advance, centering offset tolerance, and lamination window delay. These parameters serve as core inputs for state adjustment and action excitation during control system operation, determining whether the system can achieve accurate glue spraying and stable lamination under high dynamic conditions.To analyze and optimize the sensitivity of each control parameter to the glue overlay result, a mathematical mapping relationship is established by combining the aforementioned parameter set with the previously constructed inter-process time delay matrix. This relationship constitutes a functional correlation between the glue overlay accuracy and the control input, thus yielding a parameter sensitivity model. This model takes the changes in glue overlap, spraying deviation, and overlay stability caused by small variations in different control parameters as its response output. Structurally, it possesses characteristics such as differentiable mapping function, adjustable parameter space, and controllable error response, enabling the system to clearly identify the influence weight of each parameter under different operating conditions, thereby providing a basis for constructing the optimization objective function. Based on the parameter sensitivity model, target values ​​for glue positioning accuracy and overlay overlap are set as constraints for global optimization. These target values ​​are set according to process standards and actual operational requirements, such as requiring spraying deviation to be less than a certain distance and overlapping area to reach a minimum percentage, ensuring that the glue spraying quality is within a controllable range in the final product. Based on this, and combining the response characteristics of the sensitivity model, an optimization algorithm calculates a set of first control parameters for glue overlay positioning that satisfy all performance indicators.

[0029] In one specific embodiment, the process of performing step 200 may specifically include the following steps: Based on the first control parameter, the set of control unit nodes for superimposed control of the production line is divided and control functions are assigned to obtain the decision layer; The actuators of the production line superimposed control are divided into glue spraying unit, conveying unit, centering unit and superposition unit according to function to obtain the execution layer; Establish a mapping relationship between the decision-making layer and the execution layer, and map each physical execution unit to the corresponding control instruction generation unit to obtain the inter-layer information transmission channel; Based on the inter-layer information transmission channel, information transmission delay compensation and buffering are performed on the control units of each process to obtain the communication connection structure.

[0030] Specifically, the control objectives and execution constraints in the first control parameter are structurally analyzed. Based on the control objects, control cycles, data interaction frequencies, and positioning accuracy requirements involved in each process on the production line, the overlay control system of the entire production line is divided into several logically independent and functionally defined sets of control unit nodes. These node sets are constructed based on process links as basic units. Each control node corresponds to a local control area, such as a glue spraying control node, a centering calibration node, a conveyor speed regulation node, and a lamination and pressing node. Each node is spatially attached to physical equipment, but logically, it is uniformly scheduled and data fused by the central coordination module in the system. After the division is completed, according to the various constraints set in the first control parameter, such as the target spraying position, maximum allowable delay, centering accuracy tolerance, and pressing window timing, corresponding functional modules are assigned to each control node, including target tracking, state estimation, motion planning, instruction execution, and feedback acquisition functions. This forms a distributed control structure with intelligent response and local decision-making capabilities. This structure constitutes the decision layer of the digital production line overlay control. Simultaneously, to construct an execution layer corresponding to the decision-making layer, all physical devices actually involved in control actions on the production line are functionally categorized and modularized. This includes grouping all nozzle arrays, pressure valves, and adhesive application guides into the adhesive application unit; material conveying tracks, motor drives, and synchronous speed controllers into the conveying unit; high-precision positioning devices, vision calibration modules, and edge alignment mechanisms into the centering unit; and cylinder groups, hydraulic devices, and pressure head feedback devices that perform pressing actions into the stacking unit. All modules in the execution layer communicate with upper-level control commands using a unified interface standard and provide feedback based on actual operating status, forming a highly controllable and highly collaborative physical execution network. A mapping relationship between the decision-making layer and the execution layer is established. Based on the unique equipment number, a one-to-one correspondence structure from control nodes to execution units is formed through the binding between physical addresses and logical instructions. For example, the glue application coordinate command output by control node A will be mapped to nozzle M1 in the glue spraying unit, and the speed adjustment signal of control node B will correspond to the drive motor M2 in the transmission unit, and so on. The system completes the logical binding process of all execution units and generates a unified control mapping table based on this. This mapping table contains command routing paths, as well as communication protocol types, data packet formats, response delay ranges, and priority allocation, and is the basic channel structure for realizing inter-layer collaborative control. Through the control structure established by the mapping relationship, an inter-layer information transmission channel is built between the decision-making layer and the execution layer. This channel is responsible for issuing commands and handling tasks such as execution feedback, state synchronization, and fault tolerance verification. Because the physical locations of the control units in the control network are different, the communication path lengths are different, and the device drive response characteristics are different, problems such as transmission delay, bandwidth bottlenecks, and signal conflicts will occur during information transmission.To address this, information transmission delay compensation and buffering are implemented for each process control unit based on inter-layer information transmission channels to eliminate the impact of time lag on control effectiveness. A dynamic delay compensation mechanism is configured for each channel, estimating the current communication delay based on the historical transmission duration of each link and automatically adjusting its scheduling time before instruction issuance, ensuring that all instructions maintain synchronization and coordination upon arrival at the execution layer. To cope with sudden communication congestion and node blockage, a data buffer module is configured before critical channels, automatically adjusting the data push frequency and buffer release rate according to the data queue length to avoid execution delays or signal loss due to data accumulation. Through the continuous operation of these mechanisms, a stable, efficient, and clearly layered communication connection structure is formed.

[0031] In one specific embodiment, the process of performing step 300 may specifically include the following steps: Construct a variable enhancement matrix based on current state data, delayed state data, and cumulative state data; A zero-sum relationship is introduced based on the variable enhancement matrix, and the nonlinear delay derivative term is transformed into a linearized control model; Based on the linear control model, the state estimates of each process control unit are updated to obtain preliminary control parameters. Then, a momentum term and an adaptive learning rate mechanism are introduced into the preliminary control parameters to obtain control values ​​that accelerate convergence. Based on the control value for accelerated convergence, it is determined whether the positioning error or process parameter change exceeds the preset threshold, thus obtaining an optimized information interaction strategy. Based on the optimized information interaction strategy, the optimal glue positioning control command is generated.

[0032] Specifically, real-time status data, delayed status data, and cumulative status data are collected during operation. Current status data includes sensor inputs, positioning results, and feedback responses from each process control unit at the current time step, reflecting the system's immediate operating status. Delayed status data reflects time-lag information caused by communication delays, execution lags, or feedback synchronization mismatches, extracted from the cache in the form of time window backtracking to recover the system's historical dynamic trajectory. Cumulative status data is the statistical result of the total state over a certain time period, including average error, cumulative response offset, and parameter drift trends, representing the system's long-term behavioral characteristics and stability indicators. These three types of data are integrated into a set of high-dimensional state vectors after time alignment, dimension normalization, and structural splicing, and organized into a variable enhancement matrix with time-dependent characteristics. Based on the variable enhancement matrix, a zero-sum relationship is introduced as a structural constraint. This relationship originates from the dynamic balance mechanisms in multi-process control systems, such as total energy conservation, total error balance, and total offset constraints. That is, for multiple control nodes, the overall sum of their state changes should ideally be approximately zero. If the system deviates from this balance, it indicates that a certain part has an abnormal state or control imbalance. Therefore, this zero-sum constraint is introduced into the structural transformation of the variable enhancement matrix, and based on this, the nonlinear delay derivative term in the control model is algebraically transformed. Using variable substitution, differential constraints, and boundary conditions, the originally difficult-to-analyze nonlinear delay differential equation is transformed into an approximately linear state update model, resulting in a linearized control model. Based on the linearized control model, the state estimates of each process control unit are updated. The state estimation process, based on the principle of minimum mean square error or optimal observation, combines the current data and delay data in the enhancement matrix to predict the expected value and confidence interval of the target state variable, generating preliminary control parameters. These parameters include the spraying position offset, alignment correction angle, pressing delay adjustment, and control feedback gain, serving as the initial action reference for real-time system adjustment. A momentum mechanism is introduced into the preliminary control parameters, retaining some influence of the previous state adjustment direction and magnitude during each parameter update, forming an update path with inertial trends. This strategy effectively alleviates the parameter oscillation problem caused by noise or error gradient oscillations. Simultaneously, an adaptive learning rate strategy is superimposed on the momentum mechanism, allowing the step size of each parameter update to be dynamically adjusted according to the error change rate. When the error is large and the state fluctuates drastically, the learning rate will automatically increase to accelerate the adjustment speed; while when the error is close to convergence and the system tends to stabilize, the learning rate will gradually decrease to improve stability and avoid over-adjustment or oscillation, resulting in a set of control values ​​with fast convergence capability and high disturbance resistance. The control values ​​that accelerate convergence are compared and analyzed with the preset error upper limit, centering accuracy limit, and material overlap tolerance. Combined with real-time data collected during execution, such as glue spraying trajectory, material edge recognition accuracy, and pressing force feedback, it is determined whether the current system operating state triggers the control threshold condition.If the positioning error exceeds the threshold or the process state deviation exceeds the set range, it indicates that the system needs to adjust the current control rhythm and information interaction method to avoid uncontrolled spread or error accumulation. Therefore, the system initiates an optimized information interaction strategy, which includes increasing the state synchronization frequency between control nodes, increasing the scheduling priority of the master node to the slave node, compressing the information transmission cycle, and expanding the instruction confirmation mechanism, forming a more efficient and refined control information propagation path. Based on the optimized information interaction strategy, the updated control values ​​with convergence characteristics are sent to each actuator through the optimized network channel, and the instruction format and execution time window are reconfigured according to their respective dynamic characteristics and response capabilities to generate the optimal glue positioning control instruction for the current moment.

[0033] In one specific embodiment, the process of performing step 400 may specifically include the following steps: Second-order differential equations containing position and velocity states are constructed for the glue spraying unit, centering unit, and overlapping unit of the execution layer; Based on the combination of position and velocity errors using second-order differential equations and by introducing design parameters, the system trajectory convergence criterion is obtained. Based on the convergence standard of the system trajectory, the proportional term and the nonlinear power function term are combined and an adaptive gain is introduced to obtain the control command of the execution unit. The control command of the execution unit is then subjected to a third-order low-pass filter to obtain a smooth control action sequence. The smooth control action sequence is transmitted to the glue spraying valve, centering motor and pressing cylinder, and the glue superposition action is obtained by adapting the drive signal to the interface requirements of each actuator. The execution status data is obtained by collecting the glue spraying position, alignment accuracy, and pressing force during the glue layering action through a sensor network.

[0034] Specifically, second-order differential equations incorporating position and velocity states are constructed for the glue spraying unit, centering unit, and overlapping unit of the execution layer. Taking the glue spraying unit as an example, its nozzle movement trajectory is not only driven by the target position but also limited by factors such as nozzle mechanical damping, response inertia, and system feedback delay. Similarly, when the calibration platform in the centering unit performs translation or rotation operations, the effects of speed rise during motor startup, inertial backlash during braking, and trajectory correction on stability are considered. The pressing mechanism in the overlapping unit behaves as a typical dynamic control system, with its response speed, pressing stroke, and action duration all depending on the mechanical behavior of the cylinder or hydraulic cylinder. Therefore, using a unified second-order differential expression for modeling can comprehensively characterize the state evolution of each sub-module in the execution layer. After completing the dynamic modeling of each execution unit, the state difference between the desired trajectory and the actual trajectory is introduced as an error signal, specifically including the current position deviation and the current velocity error. By combining these two error terms, a composite error structure is constructed, comprising a target-oriented term and a dynamic response term. Design parameters, including error weighting coefficients, response inertia coefficients, and expected response time windows, are introduced as adjustment factors to control the system's response speed and adjustment intensity to different types of deviations. Under this adjustment structure, a trajectory convergence criterion is derived: the system must converge the actual trajectory to the target trajectory within a given response period and compress the residual error to within a tolerable range. This serves as the basic criterion for determining the effectiveness of the control strategy. Based on the trajectory convergence criterion, an actual control law is constructed, using the proportional term in error regulation as the basic control structure. This ensures that the system can immediately provide a linear response when significant deviations occur. A nonlinear power function term is introduced on top of the proportional term, enabling fine control even with small error amplitudes and rapid amplification of response intensity as the error increases, thereby enhancing the system's robustness and sensitivity. To avoid the limitations imposed by fixed parameters on system adaptability, an adaptive gain mechanism is introduced. This allows the control gain to dynamically adjust with error fluctuations, changes in its own state, and external disturbances, achieving adaptive matching of control intensity under different operating conditions and obtaining control commands for the execution unit. The control commands from the execution unit undergo third-order low-pass filtering, introducing three progressive filtering structures. Each filtering stage is based on the control output of the previous stage and sets different cutoff frequencies to reduce high-frequency interference, resulting in a smooth control action sequence. This smooth control action sequence is then transmitted to the specific physical execution interface, and signal conversion and adaptation are performed based on the structural characteristics of each execution unit and the control interface standard.For the glue spraying unit, the control sequence is converted into continuous pressure adjustment commands, nozzle displacement control commands, and start-up timing signals. For the centering unit, it is parsed into servo motor speed commands and position correction values, and embedded in the position closed loop for fine-tuning. For the lamination unit, the control signals are translated into the action start signal of the pressing cylinder, the stroke control solenoid valve signal, and the pressure regulating proportional valve signal, ensuring synchronous control of the pressing action in three dimensions: time, force, and position. The entire command transmission process is completed through standard bus protocols (such as CAN, EtherCAT, or Modbus), and a timestamp and acknowledgment mechanism is configured on the control terminal to ensure the integrity of command transmission and the consistency of response. While the action is being executed, operational feedback data is collected in real time through a sensor network pre-deployed in each unit. The glue spraying unit records the actual spray position and coverage shape of the glue using a spray trajectory detection sensor; the centering unit uses image recognition or photoelectric edge sensors to capture the offset distance and angle error after centering; the lamination unit uses pressure sensors, stroke encoders, and air pressure feedback modules to detect the pressing force, pressure head position, and action consistency. These data together constitute the execution status dataset.

[0035] In one specific embodiment, the process of performing step 500 may specifically include the following steps: Define system evaluation criteria that include positioning error, overlap ratio, control response time, and resource utilization. Based on the execution status data, the deviation between the actual glue spray position and the target position and the material superposition area ratio are calculated to obtain the positioning accuracy and superposition overlap values. Based on the system evaluation criteria and the values ​​of positioning accuracy and superposition overlap, a performance evaluation vector is constructed, and the performance evaluation vector is calculated and weighted to obtain the system performance deviation index by calculating the difference between the performance evaluation vector and the preset target value. Based on the system performance deviation index, multi-objective optimization is performed to obtain a set of candidate control parameters. The candidate control parameter set is then subjected to unit-level testing, system-level testing, and production-level testing to obtain parameter verification results. Based on the parameter verification results and the system performance deviation index, the first control parameter is adjusted to obtain the second control parameter.

[0036] Specifically, a multi-dimensional, quantifiable, and traceable system evaluation standard is constructed. This standard measures the difference between the overall performance of current control parameters in actual execution and the expected target, providing a theoretical basis for subsequent parameter tuning and control strategy evolution. The evaluation standard consists of four core dimensions: positioning error, superposition overlap, control response time, and resource utilization. Positioning error measures the spatial offset between the glue spraying position and the preset trajectory, directly reflecting the glue spraying accuracy. Superposition overlap assesses the boundary matching degree and effective overlap area ratio between multiple layers of materials, a key parameter for judging the consistency of quality during the pressing stage. Control response time reflects the time consumed by the system from receiving control commands to completing actual physical actions, an indicator for evaluating the real-time performance and coordination of the control system. Resource utilization measures the actual consumption level of energy, electricity, communication bandwidth, and computing resources during system operation, an important basis for reflecting the system's economy and operational efficiency. After the system formally enters the closed-loop control operation phase, data from all executed actions are collected in real time and integrated into an execution status dataset. Using the high-precision position feedback and image recognition data in this dataset, the difference between the actual landing point of each glue spray and the target spray trajectory is calculated. Combined with the boundary coordinates of the glue trajectory, the shape of the adhesive tape produced by each spray is matched with the target shape to determine the positioning error. Based on this, the pixels or physical coverage area of ​​the overlapping region are extracted, and the area ratio between the actual overlapping area and the theoretically expected overlapping area is calculated to obtain a quantitative index of material overlap. Simultaneously, time-series statistics are performed on the time period from control signal generation to execution completion, extracting the shortest response time, average response delay, and response jitter amplitude to form control response time analysis data. Regarding resource utilization, statistical analysis of energy consumption, current load rate, communication occupancy time, and computational load ratio in the execution layer is used to determine the total system resources consumed per unit of production task under the current operating conditions, and this is normalized into comparable economic indicators. The four dimensions of positioning error, overlap, control response time, and resource utilization are combined into a unified performance evaluation vector. Based on the target performance range or standard template set during system initialization, this evaluation vector is subtracted from the target vector term by term to obtain a performance deviation vector. Based on the importance and constraint level of each indicator in different application scenarios, weighting coefficients are set to weight and sum the deviations to calculate a performance deviation index that uniformly expresses the overall performance deviation of the system. A larger index value indicates that the current control parameters deviate more from the target control performance range, requiring prompt parameter reconstruction and optimization intervention. Once the performance deviation index exceeds a set threshold, the system initiates a multi-objective control parameter optimization module. Using the current first control parameter as the initial solution, a constrained optimization space is constructed, and minimizing the performance deviation index is introduced as the objective function to perform a multi-objective collaborative search.The search process employs adaptive algorithms based on genetic evolution, particle swarm optimization, or Bayesian optimization to dynamically generate multiple candidate control parameter sets in the parameter space. Each set of candidate parameters possesses independent combination characteristics in dimensions such as control strength, response rate, error correction gain, and delay compensation window. To ensure the engineering feasibility and operational reliability of the optimization results, each set of candidate parameters is implemented in a three-stage verification path: in unit-level testing, test commands are applied to the glue spraying, centering, and stacking modules to verify their response stability and parameter adaptability under unit actions; in system-level testing, the entire execution chain is run in tandem to evaluate the stability, synchronization, and collaborative accuracy of the parameter sets under multi-module collaborative scheduling; and in production-level testing under simulation or actual production cycles, control accuracy, stacking quality, and system load are continuously monitored and error convergence is evaluated to determine whether each set of candidate parameters can maintain controllable performance under high-intensity operating conditions. After testing, the validation results of each candidate parameter set are cross-compared with the original performance deviation index to identify the parameter combination with the best performance, fastest deviation convergence speed, and lowest resource consumption. This combination is then used as the basis to replace the corresponding parameter dimensions in the original first control parameter structure, achieving parameter-level structural adjustment and optimized convergence. This yields the second control parameters.

[0037] The above describes the production line digital overlay control method in the embodiments of this application. The following describes the production line digital overlay control device 10 in the embodiments of this application. Please refer to [link / reference]. Figure 2 One embodiment of the production line digital overlay control device 10 in this application includes: The acquisition module 11 is used to acquire parameters and preprocess data of the production line to obtain production process status information, and determine the first control parameter for glue superposition based on the production process status information. Create module 12 to create the decision layer and execution layer of the production line superimposed control using the first control parameters, and establish the communication connection structure of each process control unit; The fusion module 13 is used to perform superimposed error information fusion processing in the decision layer based on the communication connection structure to generate the optimal glue positioning control command; Operation module 14 is used to transmit the optimal glue positioning control command to the execution layer, drive the glue spraying unit, centering unit and overlapping unit to perform glue superposition operation, and collect execution status data; The adjustment module 15 is used to dynamically adjust the first control parameter based on the execution status data to obtain the second control parameter.

[0038] Through the collaborative efforts of the aforementioned components, and by constructing an adjustable Lyapunov stability matrix and variable enhancement matrix method for process intervals, the nonlinear delay control problem was solved, significantly improving glue positioning accuracy and ensuring product quality and functional stability. Based on a double-layer nested digital overlay control framework, the separation and collaboration between the decision-making and execution layers were achieved, enabling each process control unit to share information and coordinate actions, overcoming the limitations of independent operation of each process in traditional control methods. Through fixed-time sliding mode control laws and event-triggered mechanisms, the system can converge within a predetermined time, significantly shortening the adaptation cycle during process switching and improving the flexibility and efficiency of the production line. The use of distributed optimization algorithms and dynamic learning rate mechanisms reduced the system's computational burden, allowing control resources to be intelligently allocated according to process requirements, avoiding resource waste and system overload. Through fuzzy logic systems and fast fixed-time filters, interference from unknown production lines and discontinuities in execution signals were effectively compensated, enhancing the system's robustness in complex process environments. Based on multi-objective optimization algorithms and a three-stage verification and testing mechanism, the system can simultaneously optimize two key indicators: glue positioning accuracy and overlay overlap, overcoming the limitations of single-parameter optimization in traditional methods.

[0039] Please see Figure 3 , Figure 3 The present invention provides a schematic block diagram of the structure of an electronic device 300. The electronic device 300 includes a processor 301 and a memory 302, which are connected via a device bus 303. The memory 302 may include a non-volatile storage medium and internal memory.

[0040] The non-volatile storage medium can store a computer program. The computer program includes program instructions, which, when executed by the processor 301, cause the processor 301 to perform any of the aforementioned production line digital overlay control methods.

[0041] The processor 301 provides computing and control capabilities to support the operation of the entire electronic device 300.

[0042] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor 301, the processor 301 can execute any of the above-mentioned digital overlay control methods for production lines.

[0043] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device 300 involved in the present application. The specific electronic device 300 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0044] It should be understood that processor 301 can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0045] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the electronic device 300 described above can be referred to the corresponding process of the aforementioned production line digital overlay control method, and will not be repeated here.

[0046] This application also provides a computer-readable storage medium storing a computer program that, when executed by one or more processors, causes the one or more processors to implement the production line digital overlay control method provided in this application.

[0047] The computer-readable storage medium can be an internal storage unit of the electronic device 300 described in the foregoing embodiments, such as a hard disk or memory of the electronic device 300. The computer-readable storage medium can also be an external storage device of the electronic device 300, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., provided by the electronic device 300.

[0048] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0049] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0050] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A digital overlay control method for a production line, characterized in that, include: The production line is sampled and preprocessed to obtain production process status information, and the first control parameter for glue superposition is determined based on the production process status information. The decision-making and execution layers of the production line overlay control are created using the first control parameters, and the communication connection structure of each process control unit is established. Based on the aforementioned communication connection structure, superimposed error information fusion processing is performed in the decision layer to generate the optimal glue positioning control command. The optimal glue positioning control command is transmitted to the execution layer to drive the glue spraying unit, centering unit and lamination unit to perform glue superposition operation, and to collect execution status data; Based on the execution status data, the first control parameter is dynamically adjusted to obtain the second control parameter.

2. The production line digital overlay control method according to claim 1, characterized in that, The process of collecting parameters and preprocessing data on the production line to obtain production process status information, and determining the first control parameter for glue overlay positioning based on the production process status information, includes: Multiple sensor nodes are arranged on the production line, and the layout of the sensor nodes is configured to obtain a sensor network covering the front of the glue spraying machine, the centering mechanism, the front and rear of the dryer, and the superimposed pressing mechanism. The sensor network collects raw datasets from each sensor node, including workpiece position data, production line speed data, and actuator pressure data. The original dataset is linearly calibrated to obtain calibrated process data; The material transfer time between each sensor node is calculated based on the calibrated process data, and an inter-process time delay matrix is ​​constructed based on physical distance, production line reference speed, and temperature compensation coefficient. Delay analysis is performed on the inter-process time delay matrix to obtain system time delay characteristic parameters. The calibrated process data is then combined with the system time delay characteristic parameters, and the glue characteristic parameter matrix is ​​integrated to obtain production process status information. Based on the production process status information, the glue positioning parameters and inter-process delay relationships are determined, and the first control parameter for glue superposition positioning is established.

3. The production line digital overlay control method according to claim 2, characterized in that, The step of determining the glue positioning parameters and inter-process delay relationships based on the production process status information, and establishing the first control parameters for glue overlay positioning, includes: The system state space model is performed on the production process state information to obtain the production line superimposed control model; The inter-process delay intervals in the superimposed control model of the production line are divided to obtain a description of the delay characteristics. Based on the aforementioned delay characteristic description, an adjustable Lyapunov stability matrix is ​​constructed, and a set of stability control matrices is obtained by solving linear matrix inequalities. The stability control matrix group is parameterized to obtain the glue positioning parameter set; Based on the glue positioning parameter set and the inter-process time delay matrix, a mapping relationship between glue superposition accuracy and control parameters is established to obtain a parameter sensitivity model. Based on the parameter sensitivity model, the target values ​​for glue positioning accuracy and superposition overlap are set as constraints, and the first control parameter for glue superposition positioning is calculated.

4. The production line digital overlay control method according to claim 1, characterized in that, The process of creating a decision-making layer and an execution layer for overlay control of the production line using the first control parameters, and establishing the communication connection structure for each process control unit, includes: Based on the first control parameters, the set of control unit nodes for superimposed control of the production line is divided and control functions are assigned to obtain the decision layer; The actuators of the production line superimposed control are divided into glue spraying unit, conveying unit, centering unit and superposition unit according to function to obtain the execution layer; Establish a mapping relationship between the decision layer and the execution layer, and map each physical execution unit to the corresponding control instruction generation unit to obtain an inter-layer information transmission channel; Based on the inter-layer information transmission channel, information transmission delay compensation and buffering are performed on the control units of each process to obtain the communication connection structure.

5. The production line digital overlay control method according to claim 1, characterized in that, Based on the communication connection structure, the superimposed error information fusion processing is performed in the decision layer to generate the optimal glue positioning control command, including: Construct a variable enhancement matrix based on current state data, delayed state data, and cumulative state data; Based on the aforementioned variable enhancement matrix, a zero-sum relationship is introduced, and the nonlinear delay derivative term is converted into a linearized control model; Based on the linearized control model, the state estimates of each process control unit are updated to obtain preliminary control parameters. A momentum term and an adaptive learning rate mechanism are introduced into the preliminary control parameters to obtain control values ​​that accelerate convergence. Based on the control value for accelerated convergence, it is determined whether the positioning error or process parameter change exceeds a preset threshold, thereby obtaining an optimized information interaction strategy, and generating the optimal glue positioning control command based on the optimized information interaction strategy.

6. The production line digital overlay control method according to claim 1, characterized in that, The optimal glue positioning control command is transmitted to the execution layer to drive the glue spraying unit, centering unit, and lamination unit to perform glue layering operations, and execution status data is collected, including: A second-order differential equation containing position and velocity states is constructed for the glue spraying unit, centering unit, and overlapping unit of the execution layer; Based on the second-order differential equation, the position error and velocity error are combined, and design parameters are introduced to obtain the system trajectory convergence criterion; Based on the system trajectory convergence standard combination of proportional term and nonlinear power function term, and by introducing adaptive gain, the execution unit control command is obtained, and the execution unit control command is subjected to third-order low-pass filtering to obtain a smooth control action sequence. The smooth control action sequence is transmitted to the glue spraying valve, centering motor and pressing cylinder, and the glue superposition action is obtained by adapting the interface requirements of each actuator through drive signal conversion. The execution status data is obtained by collecting the glue spraying position, alignment accuracy, and pressing force of the glue superposition action through a sensor network.

7. The production line digital overlay control method according to claim 1, characterized in that, The step of dynamically adjusting the first control parameter based on the execution status data to obtain the second control parameter includes: Define system evaluation criteria that include positioning error, overlap ratio, control response time, and resource utilization. Based on the execution status data, the deviation between the actual glue spray position and the target position and the material superposition area ratio are calculated to obtain the positioning accuracy and superposition overlap values. Based on the system evaluation criteria and the positioning accuracy and superposition overlap values, a performance evaluation vector is constructed, and the performance evaluation vector is compared with the preset target value by difference calculation and weighted summation to obtain the system performance deviation index. Based on the system performance deviation index, multi-objective optimization is performed to obtain a set of candidate control parameters. The candidate control parameter set is then subjected to unit-level testing, system-level testing, and production-level testing to obtain parameter verification results. Based on the parameter verification results and the system performance deviation index, the first control parameter is adjusted to obtain the second control parameter.

8. A digital overlay control device for a production line, characterized in that, For performing the production line digital overlay control method as described in any one of claims 1-7, the production line digital overlay control device comprises: The acquisition module is used to acquire parameters and preprocess data of the production line to obtain production process status information, and determine the first control parameter for glue superposition based on the production process status information. A module is created to use the first control parameters to create the decision layer and execution layer of the production line superimposed control, and to establish the communication connection structure of each process control unit. The fusion module is used to perform superimposed error information fusion processing in the decision layer based on the communication connection structure to generate the optimal glue positioning control command; The operation module is used to transmit the optimal glue positioning control command to the execution layer, drive the glue spraying unit, the centering unit and the overlapping unit to perform glue superposition operation, and collect execution status data; The adjustment module is used to dynamically adjust the first control parameter based on the execution status data to obtain the second control parameter.

9. An electronic device, characterized in that, The electronic device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the electronic device to execute the production line digital overlay control method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, the production line digital overlay control method as described in any one of claims 1-7 is implemented.

Images