Food packaging paper coating precision compensation control system

Abstract

The present application relates to the technical field of food packaging control, and discloses a food packaging paper film coating precision compensation control system, which comprises a physical hardware support environment and a main controller, and the main controller comprises a data acquisition module, a thermodynamic calculation module, a time sequence alignment module and a multivariable control module. The data acquisition module acquires process parameters and real-time state variables; the thermodynamic calculation module calculates transient phase change vapor pressure barrier data; the time sequence alignment module calculates a delay period by using a running linear speed and performs alignment release; and the multivariable control module synchronously generates mechanical compensation, geometric wrap angle, tension speed and rheological opening degree instructions. The present application accurately quantifies water vapor physical resistance through a thermodynamic model, eliminates spatial misplacement deviation by using a time sequence alignment mechanism, cooperates with multivariable decoupling control to synchronously implement mechanical pressure increment and wrap angle exhaust compensation, offsets edge overflow and tension imbalance secondary interference, improves film coating adhesion, and maintains continuous dynamic balance of the production line.

Description

Technical Field

[0001] This invention relates to the field of food packaging control technology, specifically a food packaging paper lamination precision compensation control system. Background Technology

[0002] During the high-speed extrusion coating process of food packaging paper, residual moisture may remain within the pores of the paper base. When the high-temperature polymer melt comes into contact with the paper web, the moisture inside the paper base undergoes instantaneous vaporization and expansion due to heat, generating an outward normal reaction force and forming a phase change vapor pressure barrier at the interface. This vapor pressure barrier directly hinders the polymer melt from penetrating deeply into the underlying fibers, resulting in decreased adhesion of the coated product and making it prone to micro-delamination defects in subsequent processing.

[0003] To overcome the resistance caused by the expansion of moisture, existing control methods rely solely on increasing the pressure of the extrusion and calendering machinery. Simply increasing mechanical pressure not only fails to effectively expel the expanding gas accumulated at the interface, easily leading to the risk of paper structure collapse, but also forces the viscous polymer melt to spread excessively to both sides, causing resin overflow at the edges and contaminating the equipment. Some systems attempt to assist in venting by adjusting the spatial geometry of the paper web's feed pattern, but this alters the actual feed path length, leading to secondary physical disturbances such as tension imbalance and mechanical strip breakage. Due to the lack of decoupling mechanisms between multiple variables, the system struggles to maintain the physical balance of the entire production line while eliminating phase change vapor resistance.

[0004] When using front-end sensors for state detection and early intervention, the existing control system struggles to accurately quantify and align transient moisture resistance with the dynamic operating linear velocity due to the significant physical distance between the moisture detection location and the final extrusion calendering cut point. This results in a physical misalignment between the issuance time of the feedforward compensation command and the actual arrival time of the high-moisture area of ​​the paper web at the calendering cut point, affecting control accuracy. Furthermore, when abnormal conditions such as localized transient high moisture content cross the calendering cut point, the control system, in the process of withdrawing transient feedforward intervention and handing over control to the steady-state closed-loop algorithm, is prone to sudden changes in calculation deviations and rigid pullback impacts of the actuators due to the lack of smooth exit and state preset mechanisms. This makes it impossible to guarantee the operational safety and continuous stability of the production equipment when facing complex disturbances. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a food packaging paper coating precision compensation control system, which solves the problems of lag in feedforward compensation when overcoming moisture resistance, easy melt overflow and tension imbalance caused by single pressurization, and mechanical shock and instability caused by abrupt changes in control handover.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a food packaging paper lamination accuracy compensation control system, comprising a physical hardware support environment and a main controller. The physical hardware support environment includes a multi-source sensor array and an execution module; the main controller includes: The data acquisition module is used to acquire reference process parameters and read the measurement signals from the multi-source sensor array to generate a real-time state variable sequence that includes the running linear velocity. The thermodynamic calculation module is used to calculate transient phase change vapor pressure barrier data based on the real-time state variable sequence; The timing alignment module is used to calculate the delay period based on the running linear velocity, cache the transient phase change vapor pressure barrier data, and release it after the delay period is reached to generate aligned vapor pressure barrier data. The multivariable control module is used to receive aligned vapor pressure barrier data, synchronously generate mechanical compensation commands, geometric wrap angle commands, tension speed commands, and rheological opening commands, and send the mechanical compensation commands, geometric wrap angle commands, tension speed commands, and rheological opening commands to the execution module respectively.

[0007] Furthermore, the execution module includes an extrusion and calendering module, a rheology adjustment module, and a geometry and tension execution module; the extrusion and calendering module includes a servo actuator, the rheology adjustment module includes a servo mechanical deflector, and the geometry and tension execution module includes a servo electric cylinder and a servo traction roller; the multivariable control module sends mechanical compensation commands to the servo actuator, geometric wrap angle commands to the servo electric cylinder, tension and speed commands to the servo traction roller, and rheology opening commands to the servo mechanical deflector.

[0008] Furthermore, the multi-source sensor array includes an infrared moisture meter, a speed encoder, and thermocouples; the real-time state variable sequence includes absolute moisture content, running linear velocity, and polymer melt extrusion temperature; the data acquisition module reads the measurement signal from the infrared moisture meter to generate absolute moisture content, reads the measurement signal from the speed encoder to generate running linear velocity, and reads the measurement signal from the thermocouples to generate polymer melt extrusion temperature; the reference process parameters include basic mechanical linear pressure, reference pre-calendering wrap angle, reference rheological opening, and basic set tension.

[0009] Furthermore, the thermodynamic calculation module contains an algebraic calculation model and calls upon the internally stored effective transverse calendering width and the basis weight of food packaging paper. The real-time state variable sequence is substituted into the algebraic calculation model to perform the following calculations: The difference between the polymer melt extrusion temperature and the boiling point constant of moisture is calculated to obtain the enthalpy difference; the product of the basis weight of food packaging paper, absolute moisture content, and running linear velocity is calculated, and the calculated product is divided by the effective transverse calendering width to obtain the absolute moisture mass flux distribution density; the enthalpy difference, absolute moisture mass flux distribution density, and paper base porosity release coefficient are multiplied to generate transient phase change vapor pressure barrier data.

[0010] Furthermore, the timing alignment module has a shift register queue inside and performs the following operations: extracting the material path length from the infrared moisture meter to the calendering cut point of the extrusion calendering module; calculating the quotient of the material path length and the running linear speed to obtain the absolute lag time; calculating the quotient of the absolute lag time and the fixed control cycle, and performing a rounding mapping operation on the obtained quotient to generate a delay period; pushing the transient phase change vapor pressure barrier data into the first address of the shift register queue, and using the delay period as the offset index, performing reverse addressing in the shift register queue to extract the historical values ​​in the corresponding target storage address as the aligned vapor pressure barrier data.

[0011] Furthermore, the multivariable control module has a control matrix internally, and uses the control matrix to perform the following synchronous calculations: calculate the product of the aligned vapor pressure barrier data and the pressure compensation gain coefficient to obtain the pressure compensation increment, superimpose the pressure compensation increment onto the base mechanical linear pressure to generate the mechanical pressure target value, and encapsulate the mechanical pressure target value into a mechanical compensation command; calculate the ratio of the aligned vapor pressure barrier data to the reference normalized pressure, add the obtained ratio to the numerical value and perform a natural logarithmic operation, multiply the obtained logarithmic value by the wrap angle guidance sensitivity coefficient to obtain the wrap angle adjustment increment, superimpose the wrap angle adjustment increment onto the reference pre-coiling wrap angle to generate the geometric wrap angle target value, and encapsulate the geometric wrap angle target value into a geometric wrap angle command.

[0012] Furthermore, the multivariable control module is also used to perform the following compensation calculations: calculate the difference between the reference pre-coiling wrap angle and the target value of the geometric wrap angle, multiply the obtained difference by the tension coupling elastic coefficient to obtain the tension compensation increment; add the basic set tension and the tension compensation increment to generate the dynamic tension target value, and encapsulate the dynamic tension target value as a tension speed command; calculate the difference between the mechanical pressure target value and the basic mechanical linear pressure, multiply the obtained difference by the lateral spreading overflow coefficient to obtain the physical overflow amount, subtract the physical overflow amount from the reference rheological opening to generate the rheological opening target value, and encapsulate the rheological opening target value as a rheological opening command.

[0013] Furthermore, the multivariable control module internally includes a boundary safety limiting mechanism and a first-order inertial hysteresis model, and performs the following processing: bidirectional physical threshold clamping is applied to the target mechanical pressure value; when the target mechanical pressure value is greater than the upper limit safety pressure, the upper limit safety pressure is output; when the target mechanical pressure value is less than the lower limit safety pressure, the lower limit safety pressure is output; when the target mechanical pressure value is between the upper and lower limit safety pressures, the current target mechanical pressure value is output, generating the safe mechanical pressure; the first-order inertial hysteresis model is used to extract the final execution geometric envelope angle of the previous control cycle and multiply it by the dynamic matching filter coefficients to obtain the first product; the difference between the first value and the dynamic matching filter coefficients is calculated, the current geometric envelope angle target value is extracted, and multiplied by the difference to obtain the second product; the first product and the second product are added together to generate the final execution geometric envelope angle at the current moment.

[0014] Furthermore, the multivariable control module compares the aligned steam pressure barrier data with the preset steady-state reference threshold in each control cycle. When the aligned steam pressure barrier data is less than the steady-state reference threshold, it calls the exponential decay model to perform the following feedforward decay operation: calculates the quotient of the cumulative elapsed time of the aligned steam pressure barrier data falling below the steady-state reference threshold and the decay time constant, takes the negative sign of the quotient and uses it as the exponent of the natural constant to calculate the feedforward decay coefficient; and uses the feedforward decay coefficient as a global multiplication factor to synchronously apply to the internally generated pressure compensation increment, wrap angle adjustment increment and tension compensation increment.

[0015] Furthermore, the multivariable control module has a steady-state closed-loop algorithm. During the dynamic decay of the feedforward attenuation coefficient, the underlying dominant control is gradually returned to the steady-state closed-loop algorithm. At the initial moment when the steady-state closed-loop algorithm re-intervenes, a dynamic pre-loading operation is performed on the integral accumulator inside the steady-state closed-loop algorithm: the difference between the actual physical state value being executed at the current moment and the corresponding reference process parameter is mapped to the pre-loaded initial value of the integral accumulator.

[0016] This invention provides a precision compensation control system for food packaging paper lamination. It possesses at least one of the following beneficial effects: 1. This invention achieves precise quantification and time-domain alignment of transient vapor pressure resistance through the cooperation of a thermodynamic calculation module and a timing alignment module. The system uses a multi-source sensor array to acquire state variables such as absolute moisture content, substitutes them into an algebraic calculation model to obtain transient phase change vapor pressure barrier data, and uses a shift register queue to perform reverse addressing according to the running linear velocity to match the delay period. The timing alignment operation eliminates the spatial physical misalignment between the front-end detection position and the extrusion calendering tangent point, ensuring that the feedforward compensation command accurately corresponds to the actual high moisture content area of ​​the paper web on the time axis, thereby improving the control accuracy of the coating adhesion. 2. This invention employs a synergistic compensation strategy of mechanical pressure and spatial geometric dimension to solve the micro-delamination defect caused by water vapor expansion. The multivariable control module increases the mechanical line pressure to directly counteract the phase change steam resistance, while simultaneously calculating the target value of the geometric wrap angle based on the aligned steam pressure barrier data. It dynamically adjusts the position of the pre-compression tension guide roller. The dynamic adjustment of the geometric wrap angle creates tiny exhaust channels on the physical contact surface. While forcing the polymer melt to penetrate deeply into the underlying fibers, it also dredges the expansion gas accumulated at the interface, avoiding the risk of paper crushing caused by simply increasing the pressure. 3. This invention establishes a multivariable decoupling and disturbance-free control handover mechanism, ensuring the dynamic balance and operational safety of production equipment. The multivariable control module synchronously generates tension compensation increments and rheological opening target values ​​to offset the material tension fluctuations caused by changes in wrap angle and melt edge overflow caused by high pressure. After the local abnormal working conditions end, the system calls the exponential decay model to smooth and weaken the feedforward intervention action, and performs dynamic preloading of the actual state value on the integral accumulator of the steady-state closed-loop algorithm to avoid sudden changes in calculation deviation and mechanical rigidity impact during the control mode switching, thus maintaining the continuous stability of the entire production line. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the architecture of the food packaging paper coating accuracy compensation control system of the present invention; Figure 2 This is an overall flowchart of the food packaging paper lamination precision compensation control method of the present invention; Figure 3 This is a flowchart of the mathematical model construction and numerical calculation steps of the present invention; Figure 4 This is a schematic diagram of the shift register queue space and timing alignment steps of the present invention; Figure 5 This is a flowchart of the steps for generating mechanical pressure compensation and geometric wrap angle guidance instructions in this invention; Figure 6 This is a flowchart of the control handover steps after the abnormal state disappears according to the present invention; Figure 7 This is a diagram illustrating the calculation and timing alignment of the vapor pressure barrier in this invention. Figure 8 This is a diagram showing the dynamic compensation of mechanical pressure and the trajectory of flexible transition in this invention. Detailed Implementation

[0018] The technical solutions in 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 embodiments of the present invention, and not all embodiments. 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.

[0019] See attached document Figure 1 The present invention provides a food packaging paper coating accuracy compensation control system, which includes a physical hardware support environment and a main controller.

[0020] The physical hardware support environment includes a multi-source sensor array and execution modules, which include an extrusion and calendering module, a rheology adjustment module, and a geometry and tension execution module.

[0021] The extrusion calendering assembly includes an extruder, an extrusion die, a composite pressure roller, a rubber back roller, and a cooling roller. The composite pressure roller is connected to a servo motor or a high-frequency proportional valve that controls the linear pressure.

[0022] The rheology adjustment module includes servo-mechanical baffles located at both ends inside the extrusion die. These servo-mechanical baffles are used to adjust the physical dimensions of the die edge openings.

[0023] The geometry and tension execution module includes a pre-calendering tension guide roller, a servo electric cylinder, and a servo traction roller. The servo electric cylinder is connected to the bearing housing of the pre-calendering tension guide roller and is used to adjust the spatial position of the pre-calendering tension guide roller. The servo traction roller is located upstream of the pre-calendering tension guide roller.

[0024] The multi-source sensor array includes an infrared moisture meter, a speed encoder, and thermocouples. The infrared moisture meter is positioned upstream of the extrusion die in the material path. The speed encoder is connected to the drive spindle. The thermocouples are embedded within the flow channels of the extrusion die.

[0025] The main controller is internally configured with a data acquisition module, a thermodynamic calculation module, a timing alignment module, and a multivariable control module. These modules communicate with each other via an internal data bus.

[0026] The data acquisition module is connected to a multi-source sensor array. It is used to acquire baseline process parameters and read measurement values ​​from the infrared moisture analyzer, speed encoder, and thermocouples at a set frequency.

[0027] The thermodynamic calculation module receives data transmitted from the data acquisition module. Internally, the thermodynamic calculation module contains an algebraic calculation model. It is used to calculate the normal reaction force generated by the thermal vaporization of water.

[0028] The timing alignment module receives the calculation results from the thermodynamic solution module. Internally, the timing alignment module contains a shift register queue. This module is used for timing caching of the calculation results.

[0029] The multivariable control module is connected to the timing alignment module and simultaneously establishes control command communication with the extrusion and calendering module, the rheology adjustment module, and the geometry and tension execution module.

[0030] See attached document Figure 2 This invention provides a method for compensating and controlling the lamination accuracy of food packaging paper, comprising the following steps: S100, the data acquisition module acquires the absolute moisture content of food packaging paper, running line speed and polymer melt extrusion temperature, and generates a real-time state variable sequence; S200, the thermodynamic solution module substitutes the real-time state variable sequence into the algebraic calculation model to calculate transient phase change vapor pressure barrier data; S300, the timing alignment module calculates the delay period based on the physical installation position and operating linear velocity of the multi-source sensor array, stores the transient phase change vapor pressure barrier data into the shift register queue, and releases it to the multivariable control module after the delay period is reached; S400, the multivariable control module receives transient phase change vapor pressure barrier data and uses the internal control matrix to synchronously generate mechanical compensation commands, geometric wrap angle commands, tension speed commands and rheological opening commands; The S500 multivariable control module sends mechanical compensation commands to the servo actuator of the composite pressure roller to adjust the calendering line pressure, sends geometric wrap angle commands to the servo electric cylinder to change the position of the pre-calendering tension guide roller, sends tension speed commands to the servo traction roller to compensate for speed deviation, and sends rheological opening commands to the servo mechanical baffle block to change the edge melt extrusion amount.

[0031] The working principle of each part will be explained in detail below with specific steps and formulas.

[0032] During step S100, the data acquisition module is specifically responsible for the initialization and configuration of the system's baseline process parameters and the acquisition of real-time status variables. The initialization and configuration process of the baseline process parameters can be described in several sub-steps.

[0033] S101, the data acquisition module establishes communication with an external human-machine interface or host computer database through an internal data bus to read the process formula file of the current production batch. The process formula file contains the physical operating poles required for the extrusion calendering module to maintain standard coating quality under steady-state, undisturbed conditions. These steady-state operating poles are the absolute reference zero points for the subsequent multivariable control module to implement dynamic feedforward compensation and error calculation.

[0034] The data acquisition module extracts the basic mechanical linear pressure, reference pre-compression wrap angle, reference rheological opening, and basic set tension from the process recipe file. These static values ​​together constitute the reference process parameter set of this control system. After extraction, the data acquisition module writes the above reference process parameters into the shared memory area of ​​the main controller for direct access by the thermodynamic calculation module and the multivariable control module during the calculation cycle.

[0035] S102, the data acquisition module maps and defines the physical boundary properties of the extracted baseline mechanical parameters in memory. The basic mechanical linear pressure is defined as the constant pressure required by the composite pressure roller to force the polymer melt into the paper matrix pores to reach the target adhesion depth, without interference from the phase change vapor pressure resistance within the food packaging paper. The numerical range of this parameter is jointly determined by the fiber basis weight and density index of the food packaging paper and the melt index of the extruded polymer; the normal operating range is typically set between 1.5 MPa and 3.5 MPa.

[0036] The baseline pre-calender wrap angle is defined as the initial spatial geometric contact angle when the paper web cuts into the rubber back roll from the pre-calender tension guide roll. The purpose of setting this wrap angle is to provide a basic venting pressure angle to expel entrained air from the bottom of the paper web under normal operating conditions where moisture evaporates rapidly. The lower limit of this angle is the critical wrap angle that prevents the paper web from slipping on the roll surface by static friction, while the upper limit is limited by the mechanical interference stroke threshold of the pre-calender tension guide roll, typically ranging from π / 6 rad to π / 4 rad.

[0037] S103, the data acquisition module continues to complete the mapping definition between rheological parameters and tension reference. The reference rheological opening is defined as the initial physical distance between the servo mechanical baffles at both ends of the die head under steady-state operation. The value of this state is preset based on the effective transverse width of the current batch of food packaging paper and the natural die expansion rate of the polymer melt used inside the die head, providing a reference point for positional displacement for subsequent high-frequency edge interception actions.

[0038] The basic set tension is the fundamental mechanical tension required to maintain the flatness of the paper web during steady-state conveying in the entire food packaging paper laminating production line. This parameter is defined as the mechanical benchmark for the undisturbed conveying of the paper web in space. For the tension closed-loop control mechanism of the multi-drive point of the production line, those skilled in the art can use a general servo driver tension closed-loop architecture for deployment. The specific underlying driver hardware and tension closed-loop configuration are well-known technologies in the field and will not be described in detail here.

[0039] After completing the initial configuration of the baseline process parameters, the data acquisition module starts the high-frequency synchronous acquisition mechanism of multi-source state variables. This process is specifically executed through the following sub-steps.

[0040] S104, the data acquisition module reads the measurement signals from the infrared moisture meter in the multi-source sensor array in real time via the internal data bus according to the set sampling period. The infrared moisture meter emits an infrared beam of a specific wavelength and receives the reflected light intensity from the paper surface. Based on the absorption and attenuation characteristics of water molecules in a specific infrared frequency band, it determines the mass percentage of moisture inside the paper. These underlying electrical signals are then directly interpreted into the absolute moisture content value of the food packaging paper.

[0041] To capture the transient and drastic moisture fluctuations in the paper web during high-speed transport, the sampling period is set to the millisecond level, typically ranging from 1ms to 5ms. This high-frequency sampling mechanism ensures that the control system can promptly detect water spots or dry bands caused by uneven local moisture absorption, avoiding missed detections due to low-frequency sampling, thus providing high-resolution data support for subsequent feedforward control.

[0042] S105, within the same control cycle for reading the absolute moisture content, synchronously outputs measurement signals from the speed encoder connected to the drive spindle and the thermocouple embedded in the extrusion die flow channel. An internal high-speed counter calculates the pulse increment within a unit cycle and converts it into the current operating linear velocity. Simultaneously, the voltage signal generated by the thermocouple is also converted into the polymer melt extrusion temperature.

[0043] To ensure alignment of different physical quantities on the time axis, the data acquisition module relies on a hardware timer to establish a synchronization reference. When the timer triggers a hardware interrupt, the system forcibly latches the underlying register values ​​of the three sensors simultaneously. For the acquisition of speed encoder pulses and the analog-to-digital conversion and nonlinear compensation processing of thermocouple temperature signals, those skilled in the art can use the standard high-speed counting module and temperature input module of an industrial general-purpose programmable logic controller. The specific underlying circuitry and conversion algorithms are well-known technologies in the field and will not be elaborated here.

[0044] S106. Due to high-frequency electromagnetic interference from frequency converters and sensor reading errors caused by mechanical vibration in industrial settings, the absolute moisture content, operating line speed, and polymer melt extrusion temperature latched at the same time need to undergo digital filtering. The moving average filtering algorithm is used here to remove transient spikes and outliers from the data and smooth high-frequency noise, ensuring that the measurement data accurately reflects the actual physical state of the process.

[0045] After filtering, the valid values ​​at the current moment are packaged according to a unified timestamp to generate a real-time state variable sequence corresponding to the current control cycle. This sequence is then pushed to the shared memory of the main controller and triggers the thermodynamic calculation module to read it, using it as the dynamic input parameters for subsequent algebraic calculation models.

[0046] During step S200, the thermodynamic calculation module analyzes and prepares a model based on the collected real-time state variable sequence to analyze and model the physical mechanism of water vapor phase change in the porous structure of food packaging paper. The specific analysis logic of this process is divided into the following stages.

[0047] S201, the substrate of food packaging paper is composed of interwoven plant fibers with a microporous internal structure. During transport, a certain proportion of residual moisture is retained within the substrate, with these water molecules distributed on the fiber surface or in the capillary pores. When the paper web is conveyed to the calendering point between the composite pressure roll and the rubber backing roll, a high-temperature polymer melt is extruded and covers the surface of the paper web. The extrusion temperature of the polymer melt is much higher than the boiling point of water, and surface heat conduction occurs instantaneously upon contact.

[0048] In S202, water molecules inside the pores absorb heat and rapidly reach their boiling point, undergoing a physical phase transition from liquid to gas. This vaporization is accompanied by a rapid expansion in volume. Due to the enclosing effect of the high-viscosity polymer melt on the outside and the internal fibrous structure restricting gas flow, the expanding water vapor cannot quickly diffuse into the external space. This expansion process generates a normal reaction force perpendicular to the paper web surface and extending outwards within the micropores.

[0049] The normal reaction force forms a physical resistance layer at the interface between the paper and the polymer melt, directly hindering the polymer melt from penetrating deeply into the pores of the paper substrate. If the system maintains the basic mechanical linear pressure under normal conditions, the melt can only adhere to the surface of the paper web and cannot form a deep mechanical anchor with the underlying fibers. This shallow adhesion state makes the coated material prone to micro-delamination defects in the deep drawing and stretching forming process of downstream products.

[0050] S203, to quantitatively compensate for the aforementioned physical resistance phenomenon, the thermodynamic calculation module abstracts the process of water vaporization by heating into an algebraic calculation model. The core of the model construction lies in quantifying the thermodynamic driving source that generates the normal reaction force. Physical model analysis shows that this driving source is jointly determined by the enthalpy difference provided by the polymer melt and the total amount of water actually participating in the phase change at the current moment.

[0051] The enthalpy difference is characterized by the temperature difference between the polymer melt extrusion temperature and the boiling point of water. The total amount of water participating in the phase change is directly affected by the absolute moisture content of the food packaging paper, the running linear velocity, and the width of the physical coating. The thermodynamic calculation module extracts the corresponding variables mentioned above from the real-time state variable sequence, using them as the underlying logical basis and basic input parameters for quantifying the transient phase change vapor pressure barrier value.

[0052] See attached document Figure 3After the thermodynamic solution module clarified the physical resistance mechanism of the porous water vapor phase change, the specific construction and numerical calculation process of the mathematical model began. This process was carried out through the following sub-steps.

[0053] S204, After receiving the real-time state variable sequence transmitted by the data acquisition module, the thermodynamic calculation module extracts the current time... absolute moisture content Operating linear velocity and polymer melt extrusion temperature These three dynamic variables establish the transient thermodynamic boundary conditions for the extrusion and calendering process.

[0054] Meanwhile, the thermodynamic calculation module calls the pre-configured effective transverse rolling width in memory. and the standard basis weight for food packaging paper The effective transverse calendering width represents the physical span of the polymer melt that actually covers the paper web and makes contact; its value is determined by the baseline rheological opening in the baseline process settings. After the thermodynamic calculation module gathers the above input parameters, it starts the internal algebraic calculation model to perform data calculation.

[0055] S205, the algebraic calculation model transforms the extracted dynamic parameters into transient phase change vapor pressure barrier data characterizing the normal reaction force. The specific transformation calculation relies on the following algebraic equations: ; in, Indicates the current time Transient phase change vapor pressure barrier data; Indicates the porosity release coefficient of the paper base; Indicates the standard basis weight of food packaging paper; Indicates the current time The absolute moisture content; Indicates the current time The linear velocity of the operation; Indicates the effective transverse rolling width; Indicates the current time The polymer melt extrusion temperature; This represents the boiling point constant of water.

[0056] S206, the variables in the above algebraic equation directly correspond to the physical phase transition mechanism. Water boiling point constant. The value is fixed at 373.15 K under standard atmospheric pressure. Polymer melt extrusion temperature. With the boiling point constant of water The difference constitutes the enthalpy difference that drives the instantaneous gas phase transition of moisture. (Base weight of food packaging paper) Absolute moisture content With running linear velocity The product divided by the effective transverse rolling width It quantifies the distribution density of absolute water mass flux participating in phase transition per unit time.

[0057] Paper-based porosity release coefficient at the front end of the equation This coefficient characterizes the physical resistance properties of food packaging paper with a specific fiber structure that prevent the outward escape of internally expanding gases. It is obtained through experimental calibration based on the fiber porosity and air permeability parameters of the packaging paper at the time of manufacture. For conventional food-grade paper types with different basis weights and densities, the paper base porosity release coefficient... The calibration value range is set between 0.5 and 2.5.

[0058] After the thermodynamics solution module completes the floating-point calculations of the algebraic equations, it outputs the transient phase change vapor pressure barrier data. The result is written into the internal register. For the specific assembly execution logic of the floating-point multiplication and division instructions in the underlying hardware, those skilled in the art can use the instruction set of a general-purpose digital signal processor chip for programming. Its microinstruction architecture and register scheduling allocation are well-known technologies in the field and will not be elaborated upon here. The transient phase change vapor pressure barrier data obtained from the calculation... This forms a numerical benchmark for the subsequent quantitative application of external mechanical compensation to the system.

[0059] During the execution of step S300, the timing alignment module first addresses the control lag problem caused by spatial physical position misalignment. Its established dynamic calculation model is run through the following sub-steps.

[0060] S301, the timing alignment module reads the spatial layout parameters of the system hardware from the internal memory. These parameters mainly include the material path length from the center of the infrared moisture analyzer's detection spot to the calendering point between the composite pressure roller and the rubber back roller in the extrusion calendering module. Material path length This represents the actual physical distance that the food packaging paper travels forward after being scanned by an infrared moisture meter until it comes into contact with the polymer melt. This distance is determined as a known constant after the system's mechanical equipment is installed and fixed, and the paper feeding process is completed.

[0061] S302, the timing alignment module continuously receives the current time synchronously transmitted by the data acquisition module. running linear velocity Based on this, the dynamic calculation model within the timing alignment module establishes an algebraic functional relationship between physical distance and dynamic velocity. Since the transport of food packaging paper is considered a one-dimensional linear motion, the absolute lag time required for the detected moisture state to reach the calendering cut point with the actual paper web is considered. It is directly determined by the quotient of the material path length and the running linear speed. The specific calculation equation is as follows: ; in, Indicates the current time The absolute lag time; Indicates the length of the material feeding path; Indicates the current time The linear velocity of the operation.

[0062] To prevent the underlying microprocessor from triggering a division-by-zero exception when the device is stopped or operating at extremely low speeds, when the linear speed is... When the temperature drops to the minimum design linear velocity threshold set by the system, the above formula... The calculation is forced to use the minimum design linear speed threshold. The lower limit of this minimum design linear speed threshold is determined by the low-frequency paper crawling speed of the underlying main drive inverter, while its upper limit is limited by the maximum addressing depth of the shift register queue in subsequent steps to ensure that the calculated delay period does not cause a memory overflow. For conventional food packaging paper laminating production lines, the typical range for this minimum design linear speed threshold is set between 0.01 m / s and 0.05 m / s.

[0063] S303, In order to adapt the absolute lag time based on continuous physical time to digital logic operations, the timing alignment module further converts it into a delay period in the discrete control domain. The timing alignment module extracts the fixed control cycle set by the underlying real-time operating system of the main controller. The absolute time lag Divide by fixed control period The rounding algorithm is used to perform integer mapping, and the specific transformation equation is as follows: ; in, Indicates the current time The delay period; Indicates the current time The absolute lag time; Indicates a fixed control cycle; This represents the floor function.

[0064] S304, the calculated delay period It is a dynamically changing sequence of integers. This sequence essentially defines the control system at the current moment. The specific number of cycles for which control commands need to be suspended and delayed is required. In actual production conditions, even if the system sets a main reference linear speed, the operating linear speed... Minor fluctuations may still occur due to interference from mechanical transmission clearances and tension fine-tuning. The timing alignment module recalculates the delay period in real time for each control cycle. It can accurately compensate for transmission time deviations caused by speed fluctuations.

[0065] For fixed control cycle The underlying clock reference generation and hardware interrupt configuration can be implemented by those skilled in the art using high-precision hardware timers within industrial real-time processors. Timer interrupt scheduling and clock tree configuration are well-known technologies in the field and will not be elaborated upon here. (Generation of delay period) Subsequently, the timing alignment module uses it as the core index for subsequent deep addressing of the shift register queue.

[0066] See attached document Figure 4 After obtaining the dynamic delay period, the timing alignment module performs precise spatial and timing alignment of the feedforward data through the shift register queue. This execution mechanism is completed through the following sub-steps.

[0067] The S305 timing alignment module allocates a first-in, first-out (FIFO) shift register queue in the main controller's contiguous memory space. The physical addressing depth of this queue must cover the lowest operating line speed encountered during actual production line operation. The maximum queue depth threshold is calculated by dividing the material path length by the system's minimum design line speed, and then by the fixed control cycle. To ensure that memory space does not overflow while still meeting the requirements for long-distance lag compensation, the queue depth is typically set between 1024 and 4096 memory units.

[0068] S306, at the arrival of each fixed control cycle, the timing alignment module receives the current time output by the thermodynamic calculation module. Transient phase change vapor pressure barrier data The timing alignment module then pushes this data as the freshest state variable into the first address of the shift register queue.

[0069] At the same time, all existing historical data within the queue is synchronously shifted one storage bit width to the adjacent address behind, driven by the underlying clock. This continuous high-frequency push and shift mechanism forms a complete state data stream in the system memory that maps the actual physical motion trajectory of the paper.

[0070] S307, After the data translation and enqueueing actions are completed, the timing alignment module calls the dynamically calculated current time. Delay period The timing alignment module directly converts the delay period into an offset index of the memory pointer, and uses this index to perform reverse addressing in the shift register queue to accurately locate the corresponding target memory address.

[0071] The historical values ​​read from the target address by the timing alignment module represent those detected by the multi-source sensor array at a specific time in the past, and which happen to be detected at the current time. The target value for normal reaction force compensation is transmitted along with the physical entity of the paper web to the calendering cut point. To physically distinguish this extracted value from the raw calculated value acquired at the front end, the timing alignment module defines it as aligned vapor barrier data. The specific memory addressing and fetching logic is represented by the following algebraic relationship: ; in, Indicates the current time Alignment of vapor pressure barrier data; This represents the memory base address array of the shift register queue; Indicates the current time The delay period.

[0072] S308, Timing alignment module acquires alignment vapor barrier data. Subsequently, the data is released to the multivariable control module via the internal data bus. This data release action is constrained by the timing of the main controller's hardware interrupt, thereby ensuring that the underlying compensation signal issued by the controller coincides in the time domain with the instant the water vapor resistance layer is formed in the physical world.

[0073] For the dynamic allocation mechanism of the underlying memory pointers and the batch data transfer instructions under the bus architecture, those skilled in the art can use a general direct memory access controller hardware configuration to implement them. The underlying addressing mode and register-level data flow logic are well-known technologies in the field and will not be elaborated upon here. Successful data release signifies that the feedforward physical state is fully ready, and the system then transitions to the matrix generation stage of multivariable decoupling instructions.

[0074] See attached document Figure 5 After the multivariable control module receives the aligned vapor pressure barrier data released by the timing alignment module, the system enters the calculation phase of the multivariable decoupling command. The multivariable control module first generates mechanical pressure compensation and geometric wrap angle guidance commands based on the physical resistance model. This generation mechanism is specifically executed through the following sub-steps.

[0075] S401, the multivariable control module retrieves the previously initialized basic mechanical linear pressure and reference preload wrap angle from shared memory via the internal data bus. Simultaneously, the multivariable control module extracts the current time... Alignment of vapor barrier data The multivariable control module uses the mechanical solution unit of the internal control matrix to calculate the target mechanical pressure value required to overcome the phase change steam resistance.

[0076] Simply maintaining the basic mechanical pressure is insufficient to overcome the transiently generated high-pressure vapor barrier. The multivariable control module uses aligned vapor barrier data as a feedforward compensation component, superimposed on the basic mechanical pressure to generate a dynamic target mechanical pressure value. The specific algebraic equations are as follows: ; in, Indicates the current time The target value of mechanical pressure; Indicates the basic mechanical line pressure; This represents the pressure compensation gain coefficient; Indicates the current time Aligned vapor barrier data.

[0077] S402, in the above equation, the pressure compensation gain coefficient This coefficient is used to compensate for mechanical damping losses and elastic deformation absorption by the rubber back roller during the transmission of linear pressure in the mechanical actuator. The value of this coefficient is obtained through offline mechanical calibration based on the mechanical structural characteristics of the composite pressure roller and the Shore hardness of the rubber back roller, with a typical range of 1.1 to 1.5. The multivariable control module calculates the target mechanical pressure value. It is encapsulated as a mechanical compensation instruction.

[0078] In S403, under extreme transient high moisture content conditions, simply increasing mechanical pressure can crush the fiber structure of food packaging paper or cause the polymer melt to rupture under localized high pressure. To facilitate the drainage of phase-change moisture accumulated at the calendering cut point, the multivariable control module synchronously generates geometric instructions to change the spatial cutting shape of the paper web. By adjusting the spatial position of the pre-calendering tension guide roller and changing the wrap angle of the paper web cutting rubber back roller, a tiny wedge-shaped venting channel can be constructed on the physical contact surface, causing the high-temperature expanding moisture to overflow into the open space at the front end against the reverse material path.

[0079] S404, the multivariable control module uses the geometric solution unit of the internal control matrix to convert the aligned vapor barrier data into a geometric wrap angle target value. To prevent slippage and stripping caused by excessive wrap angle adjustment leading to a loss of static friction between the paper web and the rubber back roll, this conversion process employs a logarithmic decay model. The specific calculation equation is as follows: ; in, Indicates the current time The target value of the geometric containment angle; Indicates the reference pre-rolling wrap angle; This represents the sensitivity coefficient of the wrap angle conduction; Indicates the current time Alignment of vapor pressure barrier data; This represents the reference normalized pressure.

[0080] S405, in the logarithmic decay model, reference normalized pressure This is a constant used for dimensionless processing, typically taken as the standard atmospheric pressure value of 101.325 kPa. (The term "wrapping angle conduction sensitivity coefficient") The opening rate of the exhaust passage as the steam pressure changes is determined, and its value is set between 0.05 rad and 0.15 rad. The multivariable control module will calculate the target value of the geometric wrap angle. Package into a geometric corner wrapping instruction.

[0081] For the analog-to-digital signal conversion of the control commands for the underlying servo actuators and the message packaging and encapsulation of the fieldbus protocol, those skilled in the art can use a common motion controller standard protocol stack for deployment. The underlying communication handshake logic and message frame structure are well-known technologies in the field and will not be elaborated upon here. After the mechanical compensation command and geometric wrap angle command are generated, the system control matrix will continue to generate tension and rheological related commands that are strongly coupled with them.

[0082] After the multivariable control module outputs mechanical pressure compensation and geometric wrap angle guidance commands, the action of the physical actuator will induce secondary coupling interference within the system. In order to maintain the dynamic balance of overall production, the multivariable control module simultaneously generates tension anti-breakage and edge rheological cutoff commands. This mechanism is specifically executed through the following sub-steps.

[0083] S406, the tension decoupling unit inside the multivariable control module monitors the dynamic changes in the target value of the geometric wrap angle. The physical model shows that changing the wrap angle of the food packaging paper as it cuts into the rubber back roller directly alters the actual path length of the paper web in space. The instantaneous shortening or lengthening of this path disrupts the tension balance within the original conveying zone, causing additional stress concentration or relaxation within the paper.

[0084] If this accompanying tension disturbance is not addressed, the high-speed paper web is prone to lateral wrinkling or even mechanical breakage. The multivariate control module extracts the geometric wrap angle target value calculated at the current moment and combines it with the baseline pre-calendering wrap angle during system initialization to generate a dynamic tension target value with feedforward offsetting properties.

[0085] S407, the tension decoupling unit quantifies the impact of wrap angle changes on the actual tension of the paper web using a linear elastic deformation model, and calculates the baseline pre-calendering wrap angle relative to the current moment. The difference between the geometric wrap angle target value and the calculated difference is multiplied by the tension coupling elastic coefficient to obtain the tension compensation increment. This tension compensation increment is then added to the base set tension to generate the dynamic tension target value. The specific tension compensation calculation is based on the following algebraic equation: ; in, Indicates the current time The target value of dynamic tension; This indicates the basic tension setting used to maintain normal transmission. Indicates the elastic coefficient of tension coupling; Indicates the reference pre-rolling wrap angle; Indicates the current time The target value of the geometric envelope angle.

[0086] S408, Tension Coupling Elasticity Coefficient in Algebraic Equations This coefficient characterizes the increase in internal tension of paper caused by a unit change in wrap angle. The value of this coefficient is determined by a combination of the Young's modulus of a specific batch of food packaging paper and the suspended mechanical span between the pre-calendering tension guide roller and the extrusion cut point. For fiber-based materials with standard basis weights, the calibration range for the tension coupling elastic coefficient is set between 50 N / rad and 150 N / rad.

[0087] The multivariable control module calculates the target dynamic tension value. The signal is encapsulated as a tension speed command and sent to the upstream servo traction roller. For the closed-loop control process in which the underlying frequency converter receives the tension signal and performs torque limiting and speed fine-tuning, those skilled in the art can use the standard proportional-integral-derivative tension control algorithm. The cascaded configuration of its underlying current loop and speed loop is a well-known technology in the field and will not be described in detail here.

[0088] In S409, within the same control cycle addressing the tension coupling problem, the rheological decoupling unit of the multivariable control module tackles the risk of melt edge overflow caused by high-pressure compensation. When the target mechanical pressure value at the current moment increases compared to the basic mechanical linear pressure, the polymer melt in a viscous flow state experiences stronger normal compression between the composite pressure roller and the rubber back roller.

[0089] Due to the incompressibility and lateral spreading effect of non-Newtonian fluids, excess extrusion pressure forces the polymer melt to accelerate towards the edges of the calendering cut. This uncontrolled lateral flow causes the coating edge to exceed the physical width of the food packaging paper, and the overflowing high-temperature resin will directly adhere to the exposed rubber backing roller surface, causing equipment contamination and downtime.

[0090] S410, the rheological decoupling unit extracts the target mechanical pressure value at the current moment and calculates the physical overflow amount of the melt's lateral widening. This physical overflow amount is obtained by multiplying the calculated difference with the lateral spreading overflow coefficient, and the physical spacing of the servo mechanical baffles inside the die head is adjusted accordingly. The specific rheological opening target value is calculated using the following equation: ; in, Indicates the current time The target value of rheological aperture; Indicates the reference rheostat opening degree; Indicates the lateral spread overflow coefficient; Indicates the current time The target value of mechanical pressure; This indicates the basic mechanical line pressure.

[0091] S411, the lateral spread-over coefficient in the equation The increase in mechanical pressure is mapped to the reduction in the die extrusion width. This coefficient depends on the melt index of the current polymer and the geometric flow resistance characteristics of the die orifice, and is typically set between 0.1 mm / kPa and 0.5 mm / kPa. When the system applies high pressure to resist the phase change water vapor resistance, the die baffle simultaneously contracts towards the center, reducing the supply of melt at the edge from the source.

[0092] The multivariable control module packages the above calculation results into a rheological opening command and sends it to the servo mechanical baffles at both ends of the die. Upon receiving the rheological opening command, the servo mechanical baffles perform micro-displacement actions, changing the edge melt extrusion amount. By clarifying the quantitative algebraic relationships between the wrap angle and tension, and pressure and rheological width, this invention not only achieves direct physical countermeasures against water vapor resistance but also completely constructs a multivariable decoupled closed loop to eliminate the side effects of countermeasures, ensuring the continuous and stable operation of the high-speed coating system under multidimensional physical boundaries.

[0093] After completing the calculation of the multi-dimensional decoupling instructions, the multivariable control module enters the adaptation process of the underlying physical execution stage. To ensure the absolute safety of equipment operation and the physical synchronization of multi-dimensional actions, this process is executed through the following sub-steps.

[0094] S501, the decoupling commands output by the theoretical calculation model will generate step signals when dealing with extreme transient water vapor burst conditions. These step signals represent the ideal compensation target expected by the system, but if they are directly applied to the underlying hydraulic and servo physical equipment, they may easily exceed the physical tolerance limit of the mechanical structure or cause paper damage.

[0095] To prevent hardware-level damage, the multivariable control module incorporates a boundary safety limiting mechanism. The multivariable control module applies bidirectional physical threshold clamping to the input mechanical pressure target value, geometric wrap angle target value, dynamic tension target value, and rheological opening target value, rejecting dangerous commands that exceed the equipment's safe operating range.

[0096] S502, taking the target value of mechanical pressure, a core parameter for counteracting physical resistance, as an example, uses a multivariable control module to perform hard limiting calculations on it using a saturated nonlinear function. The specific limiting logic is implemented based on the following algebraic equations: ; in, Indicates the current time The safe mechanical pressure; Indicates the current time The target value of mechanical pressure; Indicates the upper limit of safe pressure; This indicates the lower limit of safe pressure.

[0097] S503, the upper limit of safe pressure in the equation. The upper limit of the safe pressure is determined by the critical stress at which the fibers of the food packaging paper in a specific batch break. To prevent the high-pressure mechanical force from directly crushing the paper structure and causing physical breakage, the upper limit of the safe pressure is usually set between 8 MPa and 12 MPa. The lower limit of the safe pressure... This is equivalent to the basic mechanical line pressure set during the initialization phase, ensuring that the polymer melt possesses the most basic calendering adhesion. After the amplitude limiting process, each command then enters the dynamic response matching phase.

[0098] S504. Different physical actuators in industrial settings have inherent differences in mechanical inertia and response bandwidth. For example, the response speed of a variable frequency servo motor that adjusts tension and geometric wrap angle is in the millisecond range, while a hydraulic servo proportional valve that controls pressure compensation typically has a physical delay of tens of milliseconds due to the compressibility of the oil and the flow resistance of the pipeline. Without intervention, the synchronization commands issued by the control system will manifest as disjointed actions in the physical space.

[0099] To ensure that the physical timing of multi-dimensional intervention actions coincides at the calendering cut point, the multivariable control module must perform time constant matching on the faster-responding actuators. The multivariable control module identifies the actuator with the slowest physical response in the system (i.e., the hydraulic servo proportional valve controlling pressure compensation) as the benchmark reference, and uses a digital filtering algorithm to artificially lag-shape other faster control commands.

[0100] The S505 multivariable control module employs a first-order inertial lag model to dynamically shape fast-response commands. Due to safety mechanical pressure... The corresponding physical execution response is the slowest, and the system uses it as the absolute benchmark for time alignment and sends it directly; here, the geometric envelope target value with a faster response is used. Taking the time-domain shaping process as an example, the specific smoothing filtering equation is as follows: ; in, Indicates the current time The final execution geometry envelope angle; This indicates the final execution geometry envelope of the previous control cycle; Indicates the dynamic matching filter coefficients; Indicates the current time The target value of the geometric containment angle; This indicates a fixed control period. Similarly, the target values ​​for tension and rheological opening are also subject to the same hysteresis smoothing process.

[0101] S506, Dynamic Matched Filter Coefficients The dynamic gradient slope used to adjust the faster command signal is calculated by combining the physical time constant of the slowest actuator (hydraulic servo proportional valve) with the fixed control cycle. To maintain the mathematical convergence and stability of the system, the dynamic matched filter coefficients... The value range is limited to 0 to 1. Through first-order inertial hysteresis processing, the digital instructions that originally had abrupt changes are smoothly transformed into a continuous and gradual trajectory that conforms to the mechanical response characteristics of the equipment, thus eliminating rigid impacts and oscillations in the mechanical transmission process.

[0102] After dynamic matching, each instruction is converted from a digital quantity into a fieldbus control message. For the pulse-width modulation output conversion of the multi-actuator low-level control instructions and the low-level packet encapsulation mechanism of the industrial Ethernet bus protocol stack, those skilled in the art can implement it using general programmable automation controller hardware configurations and standard distributed input / output slave stations. The underlying electrical isolation design and message communication handshake logic are well-known technologies in the field and will not be elaborated upon here. Smooth instruction output ensures that feedforward control actions can be applied to the production object in a safe and physically synchronized state.

[0103] See attached document Figure 6 After the multidimensional physical intervention actions are smoothly output and applied to the production object, the system needs to handle the handover of control after the disappearance of local abnormal states. In actual lamination production, the uneven local moisture absorption of food packaging paper has obvious transient characteristics. The multivariable control module continuously monitors the alignment vapor barrier data released by the timing alignment module, and executes the smooth exit of feedforward instructions and the disturbance-free takeover of steady-state closed loop based on the dynamic change trajectory of this data. This handover mechanism is specifically operated through the following sub-steps.

[0104] S507, the multivariable control module performs a check on the current time in each control cycle. Alignment of vapor barrier data A threshold comparison is performed. When this value falls below the preset steady-state baseline threshold, it indicates that the high moisture content region causing physical resistance has completely crossed the extrusion calendering tangent. At this point, the system must promptly withdraw the high-intensity feedforward physical intervention to avoid overcompensation in the normal drying paper web range, which could lead to mechanical fatigue or cause the coated material to break.

[0105] The aforementioned steady-state reference threshold is used to define the physical boundary between anomalous phase change water vapor bursts and conventional background noise. This value is typically determined based on the measurement fluctuation variance of the infrared moisture meter under standard drying conditions and the ambient humidity drift, and its range is generally set between 5 kPa and 15 kPa.

[0106] After confirming the end of the local transient abnormal condition, the multivariable control module immediately activates the smooth attenuation mechanism of the feedforward compensation component. If the target values ​​of mechanical pressure, geometric wrap angle, and dynamic tension are switched back to the initial basic settings at this time, the underlying servo actuator will produce a step-back action. This mechanical rigid impact can easily cause the paper web to break while it is being transported at high speed.

[0107] To achieve a flexible transition of control commands, the multivariable control module introduces an exponential decay model to dynamically weaken the feedforward compensation increment generated in the early stage. The specific decay coefficient calculation relies on the following algebraic equation: ; in, Indicates the current time The feedforward attenuation coefficient; This represents the cumulative elapsed time when the aligned vapor pressure barrier data falls below the steady-state baseline threshold; This represents the decay time constant.

[0108] S509, the decay time constant in the equation This determines the physical lag rate at which the feedforward intervention action exits. The value of this constant must match the rotational inertia of the pre-compression tension guide roller mechanical structure and the unloading and oil return characteristics of the hydraulic proportional valve to ensure that the mechanical actuators can smoothly release the elastic potential energy accumulated in the early stages. For conventional extrusion compounding equipment, the calibration range of the decay time constant is set between 100ms and 500ms.

[0109] The multivariable control module will calculate the feedforward attenuation coefficient. As a global multiplicative factor, it acts synchronously on the incremental mechanical pressure compensation, geometric wrap angle adjustment, and tension compensation of the previous output. As the accumulated elapsed time increases, the feedforward attenuation coefficient decays exponentially, forcing each feedforward compensation action to smoothly and gradually approach zero in the time domain.

[0110] In S510, on the same timeline as the feedforward command dynamically decays, the multivariable control module gradually relinquishes the system's underlying dominant control to the conventional proportional-integral-derivative steady-state closed-loop algorithm. To achieve a smooth, disturbance-free transition during control mode switching, the multivariable control module performs a dynamic pre-loading operation on its internal integral accumulator at the initial moment when the steady-state closed-loop algorithm re-engages.

[0111] The preloaded initial value of the integrator is mapped to the difference between the actual physical state value being executed at the current moment and the basic process setting value. This state preset mechanism can effectively avoid integral storms or output oscillations caused by sudden deviation changes in the steady-state closed-loop algorithm at the moment of takeover, ensuring that the underlying servo system maintains continuous torque output and position lock before and after the switch.

[0112] In S511, as the feedforward attenuation coefficient eventually reaches zero, the steady-state closed-loop algorithm takes over equipment control. The system reverts to relying on the feedback speed of the speed encoder and the real-time measurement values ​​of the tension sensor to maintain stable operation of the basic mechanical line pressure and the reference pre-coating wrap angle. For the discretized difference equation operation logic of the conventional proportional-integral-derivative algorithm, and its internal anti-integral saturation limiting mechanism, those skilled in the art can configure and call it using the standard general-purpose closed-loop control function library within the industrial automation controller. Its error tracking and gain adjustment logic are well-known technologies in the field and will not be elaborated upon here. This smooth cyclical handover between transient feedforward countermeasures and steady-state closed-loop maintenance ensures the high-speed coating production line's efficient correction capability and life-cycle operational stability in the face of complex physical disturbances.

[0113] Specific application examples:

[0114] This embodiment uses a food-grade kraft paper (food packaging paper basis weight) as the standard. Taking the actual production conditions of 80g / m² extruded composite LDPE (low-density polyethylene) as an example, the system has a fixed control cycle. Set to 5ms (0.005s).

[0115] During the steady-state initialization phase, the main controller reads the following baseline process parameters: basic mechanical linear pressure. The reference pre-calendering wrap angle is 2.0 MPa. The linear velocity is π / 6 rad (approximately 0.5236 rad). The effective transverse calendering width of the extrusion die is 5.0 m / s. It is 1.0m.

[0116] Continuous monitoring by multiple sensors. At that moment, the infrared moisture meter captured a localized water spot with high moisture content, the absolute moisture content was... The percentage increased from a steady-state 4% to a peak of 25%. The polymer melt extrusion temperature was measured simultaneously using thermocouples. The boiling point constant of water is 573.15 K (300℃). The constant is 373.15K.

[0117] Based on the algebraic calculation model: Take the paper base porosity release coefficient (Including unit conversion factors). When the absolute moisture content reaches 25%, the peak value of the instantaneously generated normal reaction force in the water spot region is: Similarly, when the system is under normal, undisturbed operating conditions (i.e., steady-state absolute moisture content of 4%), the typical background vapor pressure noise it generates is: .

[0118] System material feeding path length The absolute lag time is 2.5m. Delayed cycle beat: .

[0119] The timing alignment module pushes the calculated 40.0 kPa pressure barrier data into the shift register, and after a precise delay of 100 control cycles (0.5 seconds), releases it synchronously with the arrival of the paper web as alignment vapor pressure barrier data. .

[0120] Based on the equation: Take the pressure compensation gain coefficient The peak pressure target value is then increased to: .

[0121] At the same time, the geometric wrap angle opens to release air: When the water-spotted area moves away from the tangent point, When the pressure drops below the set steady-state reference threshold (10.0 kPa), the system triggers a decay handover. This is based on the decay time constant. implement The multidimensional feedforward increments exhibit a smooth exponential decline, and the steady-state PID controller takes over without disturbance, successfully avoiding paper web breakage caused by mechanical step pullback.

[0122] To verify the effectiveness of the present invention, a continuous winding test of 10,000 meters of paper web was conducted under the above process parameters, and 20 high moisture content patches (with a relative increase of more than 15% compared to the baseline) were set in the paper web. The relevant state variables and output control commands during the test were recorded by the main controller, and the results are as follows: Figure 7 and Figure 8 As shown.

[0123] Reference Appendix Figure 7The horizontal axis of the graph represents the system's operating time (s), and the vertical axis represents the calculated vapor pressure barrier data (kPa). The graph contains two data curves: the dashed line represents the transient pressure barrier at the acquisition end, and the solid line represents the pressure barrier after time alignment.

[0124] Based on the data analysis of the embodiments, in At this point, the dashed line shows a peak value, reaching 40.0 kPa. This change corresponds to an abnormal region detected by the upstream infrared moisture meter, where the local absolute moisture content reached 25%. The thermodynamic solution module then calculates the value based on the equation... The absolute moisture content was calculated. The increase in pressure caused the transient phase change vapor pressure barrier to rise to 40.0 kPa.

[0125] The solid line has the same waveform characteristics as the dashed line, but it is shifted to the right on the time axis, and its peak appears at... The reason for this translation is that the timing alignment module adjusts the material path length according to the material alignment module. With running linear velocity Using equations The calculated absolute lag time is 0.5s. The system stores the abnormal data calculated at the acquisition end into the shift register queue and releases it after a 0.5s delay. The translation trajectory of the solid line indicates that the system compensates for the physical distance between the sensor detection point and the rolling cut point in time, aligning the release time of the feedforward data with the arrival time of the moisture anomaly area at the physical rolling cut point. (Note: The value of the solid line before 0.5s is 0, corresponding to the data emptying and filling process of the shift register in the initial operation phase; after 0.5s, the value of the non-abnormal area of ​​the solid line recovers to the normal background noise of 6.4kPa corresponding to a steady-state 4% moisture content).

[0126] See attached document Figure 8 The horizontal axis of the graph represents the system's operating time (s), and the vertical axis represents the target mechanical pressure (MPa) issued by the multivariable control module to the servo proportional valve. The graph includes two comparison curves: the dashed line represents the traditional constant baseline pressure, and the solid line represents the multivariable feedforward compensation trajectory.

[0127] The dashed line remains horizontal at 2.0 MPa, indicating that traditional single closed-loop control only operates based on the preset baseline mechanical linear pressure. The system failed to implement feedforward regulation of the line pressure based on the water vapor phase change. The solid line initially maintained at 2.0 MPa, and then... It then began to rise, and in The pressure reached a peak of approximately 2.048 MPa. This upward trend is due to the multivariable control module receiving the timing-aligned release pressure barrier data and applying the equation... According to the pressure compensation gain coefficient The base setting is compensated (2.0 + 1.2 × 0.04 = 2.048 MPa). The increased mechanical pressure is used to overcome the reaction force of phase change water vapor generated in the abnormal region.

[0128] To the right of the peak, when Within the 0.85s to 0.9s range, the aligned vapor pressure barrier data drops below the preset 10kPa steady-state baseline threshold. At this point, the solid line does not exhibit a step-like vertical drop, but rather shows an exponentially decaying downward trend. The system gradually aligns with the 2.0 MPa baseline on both sides. This trajectory corresponds to the system's feedforward exit and control handover mechanism, with the multivariable control module invoking the exponential decay model. (decay time constant) The attenuation calculation is performed on the feedforward compensation increment superimposed in the previous stage (set to 200ms). This smooth transition trajectory avoids command steps in the servo actuator and achieves a smooth control handover to the steady-state closed-loop algorithm.

[0129] Experimental data shows that under traditional constant pressure control (dashed line corresponding mode) without feedforward compensation, the peel strength of this batch of coatings decreased from an average of 3.2 N / 15 mm to 0.8 N / 15 mm at the moisture anomaly point, and 17 micro-delamination defects were detected. After adopting the control method of this invention (solid line corresponding mode), through the handover of pressure compensation based on time alignment and control power based on exponential decay, the coating peel strength remained above 3.0 N / 15 mm in the anomaly area, and the number of micro-delamination defects was reduced. Simultaneously, the smooth decay handover process avoided mechanical strip breakage caused by sudden tension changes, improving the stability of production operations and product yield.

Claims

1. A food packaging paper lamination precision compensation control system, characterized in that, It includes a physical hardware support environment and a main controller. The physical hardware support environment includes a multi-source sensor array and an execution module. The main controller includes: The data acquisition module is used to acquire reference process parameters and read the measurement signals of the multi-source sensor array to generate a real-time state variable sequence including the running linear velocity. Thermodynamic calculation module, which is used to calculate based on the real-time state variable sequence to obtain transient phase change vapor pressure barrier data; The timing alignment module is used to calculate the delay period based on the running linear velocity, cache the transient phase change vapor pressure barrier data, and release it after the delay period is reached to generate aligned vapor pressure barrier data. The multivariable control module is used to receive the aligned vapor pressure barrier data, synchronously generate mechanical compensation instructions, geometric wrap angle instructions, tension speed instructions and rheological opening instructions, and send the mechanical compensation instructions, the geometric wrap angle instructions, the tension speed instructions and the rheological opening instructions to the execution module respectively.

2. The food packaging paper lamination accuracy compensation control system according to claim 1, characterized in that, The execution module includes an extrusion and calendering module, a rheology adjustment module, and a geometry and tension execution module; The extrusion and calendering module includes a servo actuator, the rheology adjustment module includes a servo mechanical flow deflector, and the geometry and tension execution module includes a servo electric cylinder and a servo traction roller. The multivariable control module sends the mechanical compensation command to the servo actuator, the geometric wrap angle command to the servo electric cylinder, the tension speed command to the servo traction roller, and the rheological opening command to the servo mechanical baffle.

3. The food packaging paper lamination accuracy compensation control system according to claim 2, characterized in that, The multi-source sensor array includes an infrared moisture meter, a speed encoder, and a thermocouple. The real-time state variable sequence includes absolute moisture content, running linear velocity, and polymer melt extrusion temperature; The data acquisition module reads the measurement signal from the infrared moisture meter to generate the absolute moisture content, reads the measurement signal from the speed encoder to generate the running linear velocity, and reads the measurement signal from the thermocouple to generate the polymer melt extrusion temperature. The reference process parameters include basic mechanical linear pressure, reference pre-calendering wrap angle, reference rheological opening, and basic set tension.

4. The food packaging paper lamination accuracy compensation control system according to claim 3, characterized in that, The thermodynamic calculation module contains an algebraic calculation model and calls upon the internally stored effective transverse calendering width and the basis weight of food packaging paper. The real-time state variable sequence is substituted into the algebraic calculation model to perform the following calculations: The difference between the polymer melt extrusion temperature and the boiling point constant of water is calculated to obtain the thermal enthalpy difference; Calculate the product of the basis weight of the food packaging paper, the absolute moisture content, and the running line speed, and then quotient the calculated product with the effective transverse calendering width to obtain the absolute moisture mass flux distribution density. The transient phase change vapor pressure barrier data are generated by multiplying the enthalpy difference, the absolute moisture mass flux distribution density, and the paper-based pore release coefficient.

5. The food packaging paper lamination accuracy compensation control system according to claim 3, characterized in that, The timing alignment module has an internal shift register queue and performs the following operations: Extract the material path length from the infrared moisture meter to the calendering cut point of the extrusion calendering module; The absolute lag time is obtained by calculating the quotient of the material path length and the running linear speed. Calculate the quotient of the absolute lag time and the fixed control period, and perform a rounding mapping operation on the obtained quotient to generate the delay period; The transient phase change vapor barrier data is pushed into the first address of the shift register queue, and the delay period is used as the offset index to perform reverse addressing in the shift register queue, extracting the historical value in the corresponding target storage address as the aligned vapor barrier data.

6. The food packaging paper lamination accuracy compensation control system according to claim 3, characterized in that, The multivariable control module contains a control matrix, and uses the control matrix to perform the following synchronous calculations: The product of the aligned vapor barrier data and the pressure compensation gain coefficient is calculated to obtain the pressure compensation increment. The pressure compensation increment is then superimposed on the basic mechanical linear pressure to generate a mechanical pressure target value. The mechanical pressure target value is then encapsulated as the mechanical compensation command. The ratio of the aligned vapor pressure barrier data to the reference normalized pressure is calculated. The obtained ratio is added to the numerical value and then the natural logarithm is calculated. The logarithm is multiplied by the wrap angle guidance sensitivity coefficient to obtain the wrap angle adjustment increment. The wrap angle adjustment increment is superimposed on the reference pre-calculated wrap angle to generate the geometric wrap angle target value. The geometric wrap angle target value is then encapsulated as the geometric wrap angle command.

7. The food packaging paper lamination accuracy compensation control system according to claim 6, characterized in that, The multivariable control module is also used to perform the following compensation calculations: Calculate the difference between the reference pre-rolling wrap angle and the target value of the geometric wrap angle, and multiply the obtained difference by the tension coupling elastic coefficient to obtain the tension compensation increment; The basic set tension is added to the tension compensation increment to generate a dynamic tension target value, and the dynamic tension target value is encapsulated as the tension speed command; The difference between the target mechanical pressure value and the base mechanical linear pressure is calculated. The difference is multiplied by the lateral spreading overflow coefficient to obtain the physical overflow amount. The physical overflow amount is subtracted from the reference rheological opening to generate the rheological opening target value. The rheological opening target value is then encapsulated as the rheological opening instruction.

8. The food packaging paper lamination accuracy compensation control system according to claim 6, characterized in that, The multivariable control module internally incorporates a boundary safety limiting mechanism and a first-order inertial hysteresis model, and performs the following processing: The target mechanical pressure value is subjected to bidirectional physical threshold clamping. When the target mechanical pressure value is greater than the upper limit safety pressure, the upper limit safety pressure is output. When the target mechanical pressure value is less than the lower limit safety pressure, the lower limit safety pressure is output. When the target mechanical pressure value is between the upper limit safety pressure and the lower limit safety pressure, the current target mechanical pressure value is output, thus generating a safe mechanical pressure. The final execution geometric envelope angle of the previous control cycle is extracted using the first-order inertial hysteresis model and multiplied by the dynamic matching filter coefficient to obtain the first product. The difference between the value and the dynamic matching filter coefficient is calculated, the current geometric envelope angle target value is extracted, and multiplied by the difference to obtain the second product. The first product and the second product are added together to generate the final execution geometric envelope angle at the current moment.

9. The food packaging paper lamination accuracy compensation control system according to claim 7, characterized in that, The multivariate control module compares the aligned vapor pressure barrier data with a preset steady-state reference threshold in each control cycle, and when the aligned vapor pressure barrier data is less than the steady-state reference threshold, it calls the exponential decay model to perform the following feedforward decay operation: The quotient of the cumulative elapsed time of the aligned vapor pressure barrier data falling below the steady-state reference threshold and the decay time constant is calculated. The negative value of the quotient is used as the exponent of the natural constant to calculate the feedforward decay coefficient. The feedforward attenuation coefficient is used as a global multiplication factor and is applied synchronously to the internally generated pressure compensation increment, wrap angle adjustment increment, and tension compensation increment.

10. The food packaging paper lamination accuracy compensation control system according to claim 9, characterized in that, The multivariable control module incorporates a steady-state closed-loop algorithm. During the dynamic decay of the feedforward attenuation coefficient, the underlying dominant control is gradually returned to the steady-state closed-loop algorithm. At the initial moment when the steady-state closed-loop algorithm re-intervenes, a dynamic pre-loading operation is performed on the integral accumulator within the steady-state closed-loop algorithm. The difference between the actual physical state value being executed at the current moment and the corresponding reference process parameter is mapped to the preloaded initial value of the integral accumulator.

Images