A high-orientation PET release film production system with segmented stretching precision control

By dynamically optimizing multi-source sensing and collaborative control modules, the problem of strong coupling interference of multiple physical fields in PET release film production was solved, enabling high-precision and stable production of high-orientation PET release films and improving product yield.

CN123113709APending Publication Date: 2026-11-17YANGZHOU MINGTAI FILM CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611104761.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-24
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing PET release film production systems suffer from nonlinear amplification effects due to strong coupling interference from multiple physical fields under complex stretching environments, leading to uneven film thickness, scratches, and even breakage. Traditional control systems lack global quantitative assessment of the coupling characteristics of multi-source disturbances, making it difficult to achieve high-frequency adaptive vibration suppression compensation.

Method used

Multi-source sensing modules are used to collect production parameters in real time. A collaborative control module is used to build models of thermal stability, mechanical disturbance and air film interference. Combined with servo drivers, dynamic optimization control is performed to adjust the speed of the stretching roller in real time to suppress multi-physics interference.

Benefits of technology

It effectively solves the nonlinear amplification effect caused by strong coupling interference of multi-physics fields, improves the yield and stretching control accuracy of highly oriented PET release film, and ensures the high dynamic stability and molding quality of the film.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113709A_ABST
    Figure CN123113709A_ABST
Patent Text Reader

Abstract

This invention belongs to the field of PET release film production technology, specifically a high-orientation PET release film production system with segmented stretching precision control. It includes: a multi-source sensing module for real-time acquisition of production parameters, comprising a temperature detection unit, a displacement sensor, a differential pressure transmitter, a machine vision unit, an encoder, and a speedometer; a stretching execution module, comprising a preheating roller, a stretching roller, and a servo driver; and a collaborative control module, communicatively connected to the multi-source sensing module and the stretching execution module. The collaborative control module, based on the standard deviation of the axial temperature and the rate of temperature change on the surface of the preheating roller obtained by the temperature detection unit, solves the problem of nonlinear amplification effect caused by strong coupling interference of multiple physical fields under complex stretching conditions. This effectively avoids uneven film thickness, scratches, and even breakage caused by large-scale movement of the stretching point, significantly improving the yield of high-orientation PET release films.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of PET release film production technology, specifically a high-orientation PET release film production system with segmented stretching precision control. Background Technology

[0002] Due to its excellent flatness and thermal dimensional stability, the optical and mechanical properties of high-orientation PET release films are directly determined by the control precision of the stretching stage in their molding process. However, the actual stretching environment is extremely complex. During continuous operation, the film is often subjected to strong coupling interference from multiple physical fields. For example, the axial temperature difference and transient temperature fluctuations on the surface of the preheating roller can disrupt the transverse consistency of the rheological properties of the polymer material. At the same time, the unavoidable small radial runout of the stretching roller, coupled with the micro-pressure difference fluctuations brought about by the ambient airflow, can easily induce normal vibration of the film in the stretching gap and excessive edge retraction.

[0003] When the aforementioned thermodynamic, mechanical, and aerodynamic disturbances are superimposed, they often produce a strong nonlinear amplification effect, leading to abnormal relative slippage between the film and the stretching roller. This results in significant movement of the stretching point (neckline), ultimately causing uneven film thickness, surface scratches, or even breakage. Furthermore, traditional film production control systems mostly monitor and process single variables in isolation, lacking global quantitative assessment methods for the coupling characteristics of multi-source disturbances. This non-cooperative static control logic cannot dynamically perceive the overall steady-state tolerance of the stretching system, making it difficult for the servo drive's speed control to achieve targeted high-frequency adaptive vibration suppression compensation. This has become a core bottleneck restricting the improvement of the yield of ultra-thin, highly oriented PET release films.

[0004] Therefore, the present invention provides a high-orientation PET release film production system with segmented stretching precision control. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: a high-orientation PET release film production system with segmented stretching precision control, comprising: A multi-source sensing module is used to collect production parameters in real time. The multi-source sensing module includes a temperature detection unit, a displacement sensor, a differential pressure transmitter, a machine vision unit, an encoder, and a speed measuring instrument. The stretching execution module includes a preheating roller, a stretching roller, and a servo driver; A collaborative control module, communicatively connected to the multi-source sensing module and the stretching execution module, is used for: constructing a thermal stability model to obtain a thermal stability coefficient based on the standard deviation of the axial temperature of the preheating roller surface and the rate of temperature change obtained by the temperature detection unit; constructing a mechanical disturbance model to obtain a mechanical disturbance coefficient based on the radial runout of the stretching roller and the lateral vibration amplitude of the diaphragm obtained by the displacement sensor; constructing an air film interference model to obtain an air film interference coefficient based on the airflow micro-differential pressure fluctuation obtained by the micro-differential pressure transmitter and the edge retraction rate obtained by the machine vision unit; constructing a multi-field coupled steady-state synergy model to obtain a synergy degree based on the real-time slip rate obtained by the encoder and tachometer, the standard deviation of the necking line position of the stretching point obtained by the machine vision unit, and the thermal stability coefficient and mechanical disturbance coefficient; and constructing an optimization model based on the synergy degree, the air film interference coefficient, and the current speed fluctuation, calculating the high-frequency fluctuation standard deviation of the target stretching roller speed, and sending it to the servo driver for execution.

[0007] The beneficial effects of this invention are as follows: 1. The high-orientation PET release film production system of the present invention, which controls the segmented stretching precision, collects and integrates parameters such as thermal stability coefficient, mechanical disturbance coefficient, real-time slip rate and stretching point movement variance in real time. It performs global quantitative evaluation of multi-source disturbance factors such as thermodynamics, mechanical dynamics and aerodynamics, solves the problem of nonlinear amplification effect caused by strong coupling interference of multi-physics fields under complex stretching environment, effectively avoids uneven film thickness, scratches or even breakage caused by large movement of stretching points, and significantly improves the yield of high-orientation PET release film.

[0008] 2. The high-orientation PET release film production system with segmented stretching precision control described in this invention uses the comprehensively calculated synergy degree and air film interference coefficient as adjustment factors, and combines the minimum speed fluctuation of the servo system to dynamically calculate the high-frequency fluctuation standard deviation of the target stretching roller speed. When the system is in good condition, the control target is moderately relaxed to allow a certain fluctuation range. When the system's anti-disturbance condition deteriorates, this mechanism makes the control strategy no longer a static and rigid preset, but can intelligently approach the optimal fluctuation range according to the real-time working conditions, greatly enhancing the stretching control precision and system robustness under multi-variable interference environments.

[0009] 3. The high-orientation PET release film production system with segmented stretching precision control described in this invention receives the optimized target fluctuation constraint benchmark value in real time and continuously monitors the actual high-frequency speed fluctuation of the stretching roller. Once the actual fluctuation exceeds the dynamic target set value, the driver immediately and actively triggers the internal motor control algorithm to perform precise stiffness compensation and harmonic suppression. This can quickly offset the inherent vibration and unexpected harmonic interference in mechanical operation, ensuring that theoretical control commands can be transformed into strong vibration suppression at the physical level, providing a solid guarantee for maintaining the high dynamic stability of the thin film. Attached Figure Description

[0010] The invention will now be further described with reference to the accompanying drawings.

[0011] Figure 1 This is the main flowchart of the system collaborative control in this invention; Figure 2 This is a flowchart illustrating the construction process of the thermal stability model, mechanical disturbance model, and air film interference model in this invention. Figure 3 This is a flowchart of the calculation process for the multi-field coupled steady-state synergy model and optimization model in this invention. Detailed Implementation

[0012] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0013] This invention discloses a high-orientation PET release film production system with segmented stretching precision control, comprising a multi-source sensing module for real-time acquisition of production parameters. The multi-source sensing module includes a temperature detection unit, a displacement sensor, a differential pressure transmitter, a machine vision unit, an encoder, and a speed measuring instrument. The stretching execution module includes a preheating roller, a stretching roller, and a servo driver; The collaborative control module communicates with the multi-source sensing module and the stretching execution module. The collaborative control module is used for: constructing a thermal stability model to obtain the thermal stability coefficient based on the standard deviation of the axial temperature of the preheating roller surface and the rate of temperature change obtained by the temperature detection unit; constructing a mechanical disturbance model to obtain the mechanical disturbance coefficient based on the radial runout of the stretching roller and the lateral vibration amplitude of the diaphragm obtained by the displacement sensor; constructing an air film interference model to obtain the air film interference coefficient based on the airflow micro-pressure difference fluctuation obtained by the micro-differential pressure transmitter and the edge retraction rate obtained by the machine vision unit; constructing a multi-field coupled steady-state synergy model to obtain the synergy degree based on the real-time slip rate obtained by the encoder and tachometer, the standard deviation of the necking line position of the stretching point obtained by the machine vision unit, as well as the thermal stability coefficient and the mechanical disturbance coefficient; and constructing an optimization model based on the synergy degree, the air film interference coefficient, and the current speed fluctuation, calculating the high-frequency fluctuation standard deviation of the target stretching roller speed, and sending it to the servo driver for execution.

[0014] The multi-source sensing module can contain only temperature and displacement sensors to monitor the temperature of the preheating roll and the radial runout of the stretching roll. Alternatively, the module can be manually inspected, with operators periodically recording production parameters instead of real-time data acquisition by automated sensors. In another implementation, the multi-source sensing module can be configured with only encoders and tachometers to acquire the roll surface linear velocity and the actual film linear velocity to calculate the slip ratio. The configuration of these sensors is intended to provide the system with basic production status information.

[0015] The preheating roller and stretching roller can be driven by a conventional AC motor or DC motor, and their speed can be adjusted by a simple frequency converter. In this case, the adjustment of the roller speed may depend on preset fixed parameters or manual intervention by the operator. Alternatively, the stretching execution module can adopt a mechanical tension control system to maintain the tension of the film through mechanical friction or pneumatic devices, rather than through precise speed synchronization control by a servo driver.

[0016] The collaborative control module is configured to construct a thermal stability model based on the standard deviation of the axial temperature on the surface of the preheating roller and the rate of temperature change obtained by the temperature detection unit to obtain the thermal stability coefficient. For example, the thermal stability model can use a simple linear function to weight and sum the axial temperature difference and the rate of temperature change to obtain a preliminary thermal stability assessment value. Alternatively, the model can set a fixed temperature threshold, and when the temperature difference or rate of change exceeds the threshold, it can directly output a binary stability state, indicating whether the thermal stability is within an acceptable range.

[0017] The collaborative control module constructs a mechanical disturbance model based on the radial runout of the stretching roller and the lateral vibration amplitude of the diaphragm obtained by the displacement sensor to obtain the mechanical disturbance coefficient. For example, the mechanical disturbance model can be evaluated based solely on the radial runout data of the stretching roller, ignoring the influence of the lateral vibration of the diaphragm. Alternatively, the model can use a simple averaging method to average the radial runout and lateral vibration amplitudes to obtain a rough mechanical disturbance index.

[0018] Furthermore, the collaborative control module constructs an air film interference model to obtain air film interference coefficients based on the airflow micro-pressure difference fluctuations obtained from the micro-differential pressure transmitter and the edge retraction rate obtained from the machine vision unit. For example, the air film interference model can use a simple proportional relationship to directly map the airflow micro-pressure difference fluctuations and the edge retraction rate to a coefficient. Alternatively, the model can set a fixed air film interference compensation amount based on empirical values, without performing real-time model calculations and coefficient acquisition.

[0019] Based on this, the collaborative control module constructs a multi-field coupled steady-state synergy model to obtain the synergy degree by using the real-time slip rate obtained by the encoder and tachometer, the standard deviation of the necking line position obtained by the machine vision unit, as well as the thermal stability coefficient and mechanical disturbance coefficient. For example, the multi-field coupled steady-state synergy degree model can use a simple linear weighted summation method to combine the coefficients and parameters to obtain a comprehensive synergy degree. Alternatively, the model can only consider the thermal stability coefficient and mechanical disturbance coefficient, without including the real-time slip rate and the standard deviation of the necking line position in the synergy degree evaluation.

[0020] Finally, the collaborative control module constructs an optimization model based on the degree of collaboration, the air film interference coefficient, and the current speed fluctuation. It calculates the high-frequency fluctuation standard deviation of the target stretching roller speed and sends it to the servo driver for execution. For example, the optimization model can use a preset lookup table or fixed rules to directly output a fixed speed fluctuation adjustment amount based on the degree of collaboration and the air film interference coefficient. Alternatively, the high-frequency fluctuation standard deviation of the target stretching roller speed can be manually adjusted by the operator based on experience, rather than being automatically calculated by the model.

[0021] The temperature detection unit continuously monitors the temperature distribution and temperature change rate along the width direction of the preheating roller surface; the displacement sensor detects the radial runout of the stretching roller and the lateral vibration amplitude of the diaphragm in the stretching gap in real time; the differential pressure transmitter monitors the micro-pressure fluctuation of the ambient airflow in the stretching area; the machine vision unit continuously acquires diaphragm images to track the edge shrinkage rate and the position of the stretching point (neckline); the encoder and tachometer acquire the roller surface linear velocity and the actual film linear velocity, respectively, to calculate the real-time slip rate.

[0022] The real-time acquired data is transmitted to the collaborative control module. First, based on the data from the temperature detection unit, the collaborative control module constructs a thermal stability model and calculates the current thermal stability coefficient. For example, when a large axial temperature difference or temperature change rate occurs on the surface of the preheating roller, the thermal stability coefficient will decrease accordingly, indicating fluctuations in thermorheological properties. Next, based on the data from the displacement sensor, the collaborative control module constructs a mechanical disturbance model and calculates the mechanical disturbance coefficient. For example, when the radial runout of the stretching roller or the lateral vibration amplitude of the diaphragm increases, the mechanical disturbance coefficient will decrease, indicating a decline in mechanical operational stability. Simultaneously, based on the data from the differential pressure transmitter and the machine vision unit, the collaborative control module constructs an air-film interference model and calculates the air-film interference coefficient to assess the impact of airflow disturbances on the diaphragm edge.

[0023] Subsequently, the collaborative control module inputs the calculated thermal stability coefficient and mechanical disturbance coefficient, along with the real-time slip ratio obtained by the encoder and tachometer, and the standard deviation of the necking position of the stretching point obtained by the machine vision unit, into the multi-field coupled steady-state synergy model. This model comprehensively evaluates the disturbance rejection state of these multi-dimensional physical fields and outputs a synergy degree that reflects the overall steady-state operating capability of the system. For example, when thermal stability, mechanical stability, slip ratio, and stretching point position are all in an ideal state, the synergy degree will be high; conversely, when any one or more factors are abnormal, the synergy degree will decrease.

[0024] Finally, the collaborative control module constructs an optimization model based on the calculated degree of collaboration, air film interference coefficient, and current speed fluctuation of the stretching roller. This optimization model dynamically calculates a high-frequency fluctuation standard deviation of the target stretching roller speed based on these input parameters. This target standard deviation is sent to the servo driver in the stretching execution module. After receiving the target standard deviation, the servo driver adjusts the operation of the stretching roller in real time according to this benchmark value, performing targeted high-frequency adaptive vibration suppression compensation. Thus, even in complex environments with strong coupling disturbances in multiple physical fields, the speed fluctuation of the stretching roller can be precisely controlled, thereby effectively stabilizing the stretching point position, reducing the risk of uneven film thickness and breakage, and improving the forming quality of the film.

[0025] The production system proposed in this application, through the aforementioned collaborative working mechanism, effectively addresses multi-physics field coupling disturbances during the production of high-orientation PET release film. Compared to traditional production systems that monitor and process single variables in isolation, this system can perform a global quantitative assessment of the coupling characteristics of multi-source disturbances such as thermodynamics, mechanical dynamics, and aerodynamics. This shift from single-variable monitoring to global collaborative control enables the system to dynamically perceive the overall steady-state tolerance of the stretching system.

[0026] This system constructs a multi-field coupled steady-state synergy model to integrate and evaluate these seemingly independent disturbance factors, thereby obtaining a more comprehensive understanding of the system state. As a result, the speed control of the servo drive is no longer static and preset, but can perform targeted high-frequency adaptive vibration suppression compensation based on real-time operating conditions. This effectively solves the core bottleneck of the traditional solution where the servo drive is difficult to achieve high-frequency adaptive vibration suppression compensation. This synergistic control strategy significantly improves the accuracy and stability of the film stretching process and has a positive effect on improving the yield of ultra-thin, highly oriented PET release films.

[0027] The thermal stability model built by the collaborative control module is configured to perform exponential decay calculations. The specific calculation process includes: The ratio of the actual standard deviation of the axial temperature on the preheating roll surface to the maximum allowable standard deviation threshold of the axial temperature is made dimensionless, and the ratio of the absolute value of the actual rate of change of the preheating roll surface temperature over time to the maximum allowable rate of change threshold of the temperature is also made dimensionless. The two sets of dimensionless ratios are weighted and summed to obtain the comprehensive temperature disturbance term. The negative value of the comprehensive temperature disturbance term is used as the exponent to perform a monotonically decreasing exponential function operation, and the thermal stability coefficient with a value between zero and one is output.

[0028] The thermal stability model constructed by the collaborative control module adopts the natural exponential decay function, and its formula is as follows: in, Indicates the thermal stability coefficient; The standard deviation of the temperature along the width direction of the preheating roller surface (unit: ); This indicates the maximum permissible standard deviation threshold for axial temperature (unit: ); This represents the rate of change of the surface temperature of the preheating roller over time (unit: ); This represents the maximum permissible rate of temperature change threshold (unit: ); and These represent the spatial temperature difference weight and the time rate of change weight, respectively.

[0029] The collaborative control module performs data processing, state assessment, and decision-making based on preset control logic and models. Its main function is to coordinate and optimize the complex production process globally, ensuring that the system maintains high-precision operation even under multi-physics coupled disturbances. The collaborative control module is responsible for constructing and applying a thermal stability model to accurately quantify the thermal environment state of the preheating roll, providing crucial input for subsequent stretching accuracy control. This thermal stability model aims to quantify the stability of the thermal environment on the preheating roll surface, and its output thermal stability coefficient... This model directly reflects the uniformity of the temperature distribution and the stability of dynamic changes of the preheating roller. By comprehensively considering the spatial temperature difference and time temperature fluctuation on the surface of the preheating roller, the model evaluates the influence of the thermal environment on the rheological properties of PET release film, thus providing a basis for the precise control of the stretching process.

[0030] The natural exponential decay function is a mathematical function characterized by its exponentially decreasing value as the input variable increases. In thermal stability models, it can be used to transform the temperature non-uniformity (spatial temperature difference) and temperature fluctuation (time rate of change) of the preheating roller surface into a thermal stability coefficient between 0 and 1. This functional form can sensitively reflect the nonlinear effect of temperature disturbances on system stability. Even small temperature deviations or fluctuations can lead to a significant decrease in the thermal stability coefficient, thus allowing for a more accurate assessment of the degree of thermal environment deterioration.

[0031] thermal stability coefficient It is the output of the thermal stability model, used to quantify the stability of the thermal environment on the surface of the preheating roller. Its value range is usually between 0 and 1, where 1 represents an ideal and completely thermally stable state, while a value close to 0 indicates that there is a serious disturbance in the thermal environment. As an important input parameter of the multi-field coupled steady-state synergy model, this coefficient directly affects the system's assessment of the overall steady-state level.

[0032] Standard deviation of temperature along the width direction of the preheating roller surface To measure the uniformity of temperature distribution along the width of the film on the surface of the preheating roller, the method is to measure in real time by uniformly arranging multiple temperature sensors along the axial direction on the surface of the preheating roller, and then calculate the statistical standard deviation of these measurements.

[0033] Larger The value indicates that uneven temperature distribution on the surface of the preheating roller may lead to differences in the rheological properties of the film in the transverse direction, thus affecting the stretching quality. This represents the maximum permissible standard deviation threshold for axial temperature. This is the upper limit of the preset, acceptable standard deviation of temperature along the width of the preheating roller surface. It represents the maximum degree of temperature non-uniformity the system can tolerate while ensuring film production quality. This threshold can be empirically set based on the characteristics of the PET material, product quality requirements, and historical production data, or it can be determined through experimental optimization. When the temperature approaches or exceeds this threshold, it indicates that the uniformity of the thermal environment has reached a critical state.

[0034] Rate of change of surface temperature of preheating roller over time It is used to measure the degree of fluctuation of the surface temperature of the preheating roller over time. It can be calculated by performing time differentiation or difference operations on continuous measurement data from one or more temperature sensors over a period of time.

[0035] Larger The value indicates that the surface temperature of the preheating roller fluctuates rapidly, which may adversely affect the transient rheological behavior of the film. This represents the maximum permissible temperature change rate threshold. This is a preset, acceptable upper limit for the rate of change of the preheating roller surface temperature over time. It represents the maximum temperature fluctuation rate the system can tolerate while ensuring film production stability. This threshold can also be set based on material properties, process requirements, and empirical data. When the temperature approaches or exceeds this threshold, it indicates that the dynamic stability of the thermal environment has been challenged.

[0036] Spatial temperature difference weighting Weighted by the rate of change over time Used to adjust the temperature difference in space and rate of change over time The relative importance of these two factors in the thermal stability model can be adjusted by modifying their weights to flexibly configure the model's response to these two types of thermal disturbances, based on specific production processes, material properties, or the varying sensitivities of products to temperature uniformity and temperature stability. For example, if a certain PET release film has extremely high requirements for lateral thickness uniformity, the weights can be appropriately increased. If higher requirements are needed for transient tensile stability, the [capacity] can be increased. These two weights are typically determined through expert experience, historical data analysis, or optimization algorithms, and usually satisfy the following conditions: The normalization conditions.

[0037] The collaborative control module first obtains the temperature data along the width direction of the preheating roller surface from the temperature detection unit and calculates its standard deviation. This reflects the uniformity of temperature distribution. Simultaneously, the module also acquires the rate of change of the preheating roller surface temperature over time. To capture transient temperature fluctuations, and to quantify these multidimensional temperature characteristics in a unified manner, the collaborative control module will... Compared with the preset maximum allowable axial temperature standard deviation threshold Perform normalization processing, and similarly... With the maximum allowable rate of temperature change threshold Normalization is performed.

[0038] Subsequently, through spatial temperature difference weighting Weight of time change rate The two normalized values ​​are weighted and summed to form a comprehensive temperature perturbation index. Finally, this comprehensive index is input into the natural exponential decay function to calculate the thermal stability coefficient. This exponential decay characteristic means that even small temperature disturbances can lead to... The significant decrease in temperature sensitivity reflects the deterioration trend of the thermal environment, overcoming the limitations of traditional single temperature monitoring. It organically combines the two key factors of spatial temperature difference and time fluctuation on the surface of the preheating roller and maps them into a unified thermal stability coefficient in a nonlinear manner. This provides a comprehensive and accurate thermal environment assessment parameter for the subsequent multi-field coupled steady-state synergy model, effectively supporting the decision-making of the entire segmented stretching precision control system and improving the ability to perceive and respond to thermal disturbances during the production of high-orientation PET release film.

[0039] By setting the weights for spatial temperature difference and time rate of change, the system can flexibly adjust the weighting of each factor on the thermal stability coefficient according to different process requirements, making the thermal stability assessment more targeted. Ultimately, the obtained thermal stability coefficient... It can intuitively and sensitively characterize the steady-state level of the current thermal environment. As an important input to the multi-field coupled steady-state synergy model, it significantly improves the perception and response speed of the entire segmented stretching precision control system to thermal disturbances.

[0040] The mechanical disturbance model constructed by the collaborative control module is configured to perform rational fractional damping calculations. The specific calculation process includes: The ratio of the actual radial peak-to-peak runout of the stretching roller to the maximum allowable radial runout threshold is made dimensionless and a square term is calculated. Similarly, the ratio of the actual transverse vibration amplitude of the diaphragm to the critical vibration amplitude that triggers rupture is made dimensionless and a square term is calculated. Penalty coefficients are assigned to the above two sets of square terms and the weighted sum is obtained to obtain the comprehensive mechanical penalty term. A rational fraction is constructed using the sum of the fundamental constant and the comprehensive mechanical penalty term as the denominator, and the mechanical disturbance coefficient that converges nonlinearly with mechanical disturbance is output.

[0041] The mechanical disturbance model constructed by the collaborative control module adopts a rational fractional damping function, the formula of which is: in, Indicates the mechanical disturbance coefficient; Indicates the radial peak-to-peak runout of the stretching roller (unit: ); Indicates the maximum permissible radial runout threshold (unit: ); Indicates the lateral vibration amplitude of the diaphragm in the tension gap (unit: ); This represents the critical vibration amplitude that triggers membrane rupture (unit: ); and These represent the penalty coefficients for mechanical runout and diaphragm vibration, respectively.

[0042] This collaborative control module is responsible for receiving various production parameters from the multi-source sensing module and performing complex model building, state assessment, and control command issuance based on these parameters. It integrates multi-source information to achieve global optimization and precise control of the production process. The mechanical disturbance model aims to quantify the degree of interference to the tensile stability of the film caused by mechanical factors during the production process. It transforms the complex mechanical motion state and its impact on the film into a calculable value, providing a basis for subsequent control decisions.

[0043] Rational fractional damping functions are a mathematical function form often used to describe systems with nonlinear damping or attenuation characteristics. They are used to construct mechanical disturbance models and can effectively nonlinearly map and quantify the negative impact of mechanical disturbances (such as bouncing and vibration) on system stability, making the model more sensitive to disturbances, especially when the disturbance is close to the critical value.

[0044] The mechanical disturbance coefficient is a quantitative indicator output by the mechanical disturbance model. Its value reflects the degree of influence of mechanical disturbance on system stability under the current production state. This coefficient can be a value between 0 and 1. For example, the closer the value is to 1, the smaller the mechanical disturbance and the more stable the system; the closer the value is to 0, the larger the mechanical disturbance and the less stable the system.

[0045] The radial peak-to-peak runout of the stretching roll refers to the difference between the maximum and minimum deviations of the radial position of the stretching roll surface during rotation. This can be measured in real time by a non-contact displacement sensor (e.g., a laser displacement sensor or an eddy current sensor).

[0046] This indicates the maximum permissible radial runout threshold, which is an upper limit set according to the production process requirements and equipment precision standards of PET release film. When the radial runout of the stretching roller exceeds this threshold, it may have an unacceptable impact on the film's thickness uniformity, surface quality, etc. This threshold can be determined based on experimental data, industry standards, or expert experience.

[0047] The transverse vibration amplitude of the diaphragm within the stretching gap refers to the maximum displacement of the film's oscillation or vibration perpendicular to the stretching direction within the stretching region. It can be measured in real-time by a machine vision unit (e.g., a high-speed camera with image processing algorithms) capturing changes in the diaphragm's edge position, or through non-contact detection using an optical sensor array. This represents the critical vibration amplitude that triggers membrane rupture. It is the maximum transverse vibration amplitude that the membrane can withstand during stretching. Once the transverse vibration amplitude of the membrane exceeds this critical value, the membrane may tear, break, or be severely deformed, leading to production interruption or product scrap. This critical value is usually determined through material mechanics testing, process experiments, or simulation analysis.

[0048] and These represent the penalty coefficients for mechanical runout and diaphragm vibration, respectively. These coefficients are used to adjust the impact of radial runout of the stretching roller and lateral vibration of the diaphragm on the mechanical disturbance factor. The relative influence weight can be adjusted, for example, if a certain film is particularly sensitive to stretching roller runout, the weight can be appropriately increased. The value can be increased if the diaphragm is more sensitive to vibration. The values ​​of these coefficients can be optimized through historical production data analysis, machine learning algorithm training, or expert experience.

[0049] The collaborative control module first obtains the radial peak-to-peak runout of the stretching roller. The transverse vibration amplitude of the diaphragm in the tension gap These real-time measurements were then compared with a preset maximum permissible radial runout threshold. and the critical vibration amplitude that triggers membrane rupture By performing ratio processing, the motion deviation of the mechanical structure and the dynamic performance of the thin film can be effectively normalized, making the disturbance effects under different working conditions comparable.

[0050] Next, these normalized ratios are substituted into the rational fractional damping function, and the deviation is amplified by the squared term, thereby significantly enhancing the model's sensitivity to anomalous disturbances. Simultaneously, a mechanical runout penalty coefficient is introduced. and diaphragm vibration penalty coefficient The collaborative control module can flexibly adjust the coefficient of mechanical disturbance of the stretching roller runout and film vibration to the overall mechanical disturbance based on the characteristics of different production environments or the sensitivity of film materials. The weighted contributions of each factor ultimately determine the mechanical disturbance coefficient output by the model. The impact of current mechanical disturbances on system stability is precisely quantified. This modeling method based on the ratio of physical quantities and the penalty mechanism provides a reliable basis for the quantification of mechanical disturbances for global optimization by the collaborative control module, effectively solving the problem that mechanical dynamic disturbances are difficult to perceive and quantify accurately.

[0051] By constructing a mechanical disturbance model based on a rational fractional damping function, the system can accurately quantify the impact of radial runout of the stretching roller and lateral vibration of the film on the film forming quality. This avoids the limitations of relying solely on experience or a single parameter for judgment. Especially when the runout of the stretching roller or the vibration of the film approaches the critical value, the nonlinear characteristics of the rational fractional damping function can rapidly amplify the impact of the disturbance, enabling the system to identify potential risks in a timely manner. This significantly improves the flatness, thickness uniformity, and overall mechanical quality of high-orientation PET release films, effectively reduces the risk of film defects and breakage caused by mechanical disturbances, and thus improves product yield.

[0052] The air-film interferometry model constructed by the collaborative control module is configured to perform inverse trigonometric function scaling operations. The specific calculation process includes: Calculate the ratio of the actual micro-pressure difference fluctuation of the ambient airflow in the stretching region to the maximum critical pressure difference threshold, and the ratio of the actual edge retraction rate to the maximum allowable edge retraction rate threshold; assign response coefficients to the above two sets of ratios and sum them by weight to obtain the comprehensive interference term; substitute the comprehensive interference term into the arctangent function to obtain the smooth scaling value; subtract the smooth scaling value from the ideal state reference value to output the air film interference coefficient with the limit state saturation characteristics.

[0053] The air-film interference model constructed by the collaborative control module adopts an inverse trigonometric function scaling model, and its formula is: in, Indicates the air film interference coefficient; This represents the micro-pressure difference fluctuation of the ambient airflow in the stretching region (unit: ); This represents the maximum critical differential pressure threshold (unit: ); This indicates the edge shrinkage rate of the diaphragm before it enters the cross-stretching machine (unit: ); This represents the maximum permissible edge shrinkage rate threshold (unit: ); and These represent the pressure difference and the retraction response coefficient, respectively.

[0054] This collaborative control module can be implemented using an industrial-grade programmable logic controller (PLC), an industrial personal computer (IPC), or an embedded system based on a digital signal processor (DSP) to ensure real-time performance and stability. The air film interference model is a software or hardware module within the collaborative control module. Its function is to quantify the impact of aerodynamic disturbances on the film stretching process mathematically. This model transforms physical phenomena into calculable coefficients, providing a basis for subsequent collaborative control. The inverse trigonometric scaling model is the specific mathematical structure adopted by the air film interference model. It utilizes the nonlinear characteristics of the arctangent function to map the input micro-pressure difference fluctuations and edge retraction rates into an air film interference coefficient between 0 and 1. This nonlinear mapping enables the model to exhibit higher sensitivity when the input variables approach critical values, thereby more accurately reflecting the degree of disturbance.

[0055] in, The air film interference coefficient is a dimensionless value used to characterize the degree of air film interference. Its magnitude directly reflects the impact of aerodynamic disturbances on the stability of the thin film; a smaller value generally indicates more severe air film interference. This represents the micro-pressure difference fluctuation of the airflow in the stretching region. It is a physical quantity collected in real time by a micro-differential pressure transmitter, reflecting the instantaneous change in airflow pressure within the stretching region. Such fluctuations may cause uneven stress on the diaphragm during the stretching process. It is the maximum critical differential pressure threshold, which is a preset, allowable maximum micro-differential pressure fluctuation value.

[0056] When the actual measurement When the airflow disturbance approaches or exceeds this threshold, it indicates that the airflow disturbance has reached a level that may affect production quality. This threshold is usually determined through experimental calibration or process experience.

[0057] This indicates the edge shrinkage rate of the diaphragm before it enters the stretching machine. This is the degree of edge shrinkage of the diaphragm detected by the machine vision unit. Edge shrinkage is a common defect in the film stretching process and is affected by a variety of factors, including airflow disturbances.

[0058] It is the maximum allowed edge shrinkage rate threshold, and Similarly, it is a preset, maximum permissible edge retraction value used to define the acceptable state of the membrane edge. and These represent the differential pressure and the retraction response coefficient, respectively. They are weighting parameters in the model, used to adjust the relative contributions of micro-pressure fluctuations and edge retraction rates to the air film interference coefficient. These coefficients can be adjusted through system identification, optimization algorithms, or expert experience to better adapt the model to the characteristics of the actual production line.

[0059] The collaborative control module receives the micro-differential pressure fluctuations of the ambient airflow in the stretching region from the micro-differential pressure transmitter in the multi-source sensing module. And the edge retraction rate of the membrane before it enters the cross-stretching machine, obtained by the machine vision unit. These real-time data are input into the air-film interference model inside the collaborative control module. This model uses an inverse trigonometric function scaling structure, combined with a preset maximum critical differential pressure threshold. Maximum allowable edge shrinkage rate threshold and pressure difference and retraction response coefficient and Calculate the current air film interference coefficient. .

[0060] The inverse trigonometric scaling model can map the nonlinear effects of these input variables into a unified interference coefficient. Especially when the input variables approach critical values, the model provides a more sensitive response, thus accurately reflecting the actual impact of airflow disturbances on film stability. The calculated air film interference coefficients... Subsequently, it serves as a key input parameter in the multi-field coupled steady-state synergy model, working in conjunction with other factors such as the thermal stability coefficient and mechanical disturbance coefficient to comprehensively evaluate the steady-state synergy of the entire production system. This integrated approach ensures that aerodynamic disturbances are incorporated into the global control strategy, enabling the system to more comprehensively perceive and respond to multi-physics coupled disturbances during the production process. This provides a more accurate basis for calculating the high-frequency fluctuation standard deviation of the target stretching roll speed in subsequent optimization models.

[0061] By introducing an air-film interference model, the micro-pressure difference fluctuations of the ambient airflow in the stretching region and the edge retraction phenomenon of the diaphragm before entering the stretching machine can be accurately quantified. The model adopts an inverse trigonometric function scaling structure, which can effectively capture the nonlinear effects of these aerodynamic factors on the stability of the film, thereby overcoming the problem that traditional control systems lack a targeted compensation mechanism when facing airflow disturbances.

[0062] The obtained air film interference coefficients Integrating it into the multi-field coupled steady-state synergy model enables the system to more comprehensively evaluate the multi-physics coupling state in the production process. This not only improves the consideration of thermodynamic and mechanical dynamic disturbances, but also incorporates aerodynamic disturbances into the global control strategy, significantly enhancing the system's adaptability to complex production environments and its ability to resist disturbances.

[0063] Ultimately, through precise sensing and quantification of air film interference, the collaborative control module can generate a more accurate high-frequency fluctuation standard deviation of the target stretching roller speed, guiding the servo driver to perform more refined vibration suppression compensation. This effectively suppresses problems such as film normal vibration, excessive edge retraction, and stretching point drift caused by unstable airflow, thereby significantly improving the flatness, surface quality, and overall yield of high-orientation PET release films.

[0064] The multi-field coupled steady-state synergy model is configured to perform a series exponential function operation on multidimensional state variables. The specific operation process includes: Positive benefit elasticity exponents are assigned to the thermal stability coefficient and the mechanical disturbance coefficient, respectively; the ratio of the real-time slip rate to the maximum tolerable slip rate is introduced into the bias constant to construct a slip penalty term, and a negative penalty elasticity exponent is assigned to it; the ratio of the standard deviation of the actual necking line position to the boundary value of the maximum allowable free variance is introduced into the bias constant to construct a necking penalty term, and a negative penalty elasticity exponent is assigned to it; the system coefficients with positive benefit elasticity exponents and the process penalty term with negative penalty elasticity exponents are multiplied together to output the multi-field coupling steady-state synergy degree of the dynamic equilibrium of the system.

[0065] The multi-field coupled steady-state synergy model uses a power function structure to comprehensively evaluate the disturbance resistance state of multidimensional physical fields. Its formula is: in, Indicates the steady-state coordination degree of multi-field coupling; Real-time slip ratio (unit: ); Maximum tolerable slip ratio (unit: ); The standard deviation of the necking line position at the stretching point (unit: ); This represents the maximum allowable wandering variance boundary value of the necking line (unit: ); and This represents the positive return elasticity coefficient; and This represents the negative penalty elasticity coefficient.

[0066] This multi-field coupled steady-state synergy model aims to quantify the system's ability to resist external disturbances and maintain stable operation under the interaction of multiple physical fields (such as thermal, mechanical, and airflow fields) during the production process. It adopts a power function structure, which means that the influence factors of different physical fields are combined in the form of a product and weighted exponentially to reflect the contribution or weakening effect of each factor on the overall stability in a nonlinear way. This structure can capture the complex nonlinear coupling relationship between the physical fields, rather than a simple linear superposition, so as to more accurately evaluate the comprehensive steady-state performance of the system under multi-source disturbances. It can be implemented by calculation through a pre-established mathematical model.

[0067] Multi-field coupled steady-state synergy It is a comprehensive dimensionless index used to characterize the overall stability and disturbance resistance of the stretching process in a high-orientation PET release film production system under the combined action of multiple physical fields. The value of this coefficient is usually between 0 and 1. The higher the value, the more stable the system operation and the stronger the ability to resist disturbances. Conversely, the lower the value, the worse the system stability. The calculation result of this coefficient can be used as the basis for subsequent control strategy adjustment, such as for optimizing the high-frequency fluctuation standard deviation of the stretching roller speed.

[0068] Real-time slip ratio The slip ratio refers to the relative difference between the linear velocity of the film itself and the linear velocity of the stretching roller surface during the stretching process. It reflects whether there is relative motion between the film and the stretching roller and is a key indicator for measuring the uniformity of force and transmission efficiency of the film in the stretching area. A high slip ratio may lead to scratches on the film surface, uneven thickness, or wandering of the stretching point. The real-time slip ratio can be obtained in various ways, such as by calculating the difference ratio between the linear velocity of the roller surface fed back by the encoder and the actual linear velocity of the film measured by a laser Doppler velocimeter.

[0069] Maximum tolerable slip ratio This refers to the highest real-time slip rate threshold that the system can accept while ensuring the quality of thin film products and the stability of the production process. Exceeding this threshold means that the relative motion between the film and the stretching roller has reached an unacceptable level, which may lead to product defects or production interruptions. This threshold is usually determined through experimental testing, process experience, or simulation analysis, and is adjusted according to different product specifications and production conditions.

[0070] Standard deviation of necking point position The standard deviation is an indicator of the degree of lateral positional fluctuation of the necking-off line (the region where significant plastic deformation begins to occur in the film) during the stretching process. The stability of the necking-off line position is crucial for ensuring film thickness uniformity and consistent mechanical properties. A large standard deviation indicates significant lateral movement of the necking-off line, which may lead to local overstretching or understretching of the film. This standard deviation can be obtained by continuously acquiring image sequences using a machine vision unit and calculating them in real time using an edge detection algorithm.

[0071] Maximum allowable wandering variance boundary value of the necking line This refers to the maximum standard deviation of the necking line position fluctuation that the system can tolerate while ensuring product quality and process stability. Exceeding this boundary value may result in the film's thickness uniformity, optical properties, or mechanical properties failing to meet requirements. This boundary value is typically set based on factors such as product standards, process windows, and equipment precision.

[0072] Positive return elasticity coefficient and It is used to adjust the thermal stability coefficient. and mechanical disturbance coefficient Steady-state cooperability of multi-field coupling Weighting factors for contribution levels; these coefficients are typically positive, and their magnitude reflects the degree of positive impact of the corresponding physical field stability on the overall synergy. For example, a larger... This indicates that thermal stability makes a more significant positive contribution to overall synergy. These coefficients can be calibrated through expert experience, historical data analysis, or optimization algorithms to accurately reflect the relative importance of different physical fields in actual production.

[0073] Negative penalty elasticity coefficient and It is used to adjust the real-time slip ratio. Standard deviation of the necking line position at the stretching point Steady-state coherence of multi-field coupling The weighting factors for the inhibitory effect are typically positive, and their magnitude reflects the degree of negative impact of the corresponding perturbation factor on the overall synergy. For example, a larger... This indicates that an increase in the real-time slip ratio has a more significant punitive effect on the decline in overall coordination. These coefficients can also be calibrated through expert experience, historical data analysis, or optimization algorithms to accurately reflect the sensitivity of different disturbance factors in actual production.

[0074] The thermal stability coefficient is determined through the model. and mechanical disturbance coefficient As positive benefit factors, these factors are calculated using thermal stability and mechanical disturbance models constructed from the standard deviation of axial temperature on the preheating roller surface and the rate of temperature change, and the radial runout of the stretching roller and the transverse vibration amplitude of the diaphragm, respectively. Higher values ​​of these two coefficients indicate better thermal and mechanical stability of the system and a greater contribution to overall synergy. The degree of their influence is determined by the positive benefit elasticity coefficient. and Adjustments were made.

[0075] The model incorporates real-time slip ratio. Standard deviation of the necking line position at the stretching point As a negative penalty factor, real-time slip rate This reflects the relative motion between the film and the stretching roller, while the standard deviation of the necking line position at the stretching point... These two indicators characterize the uniformity of film deformation in the stretching region. The larger the values ​​of these two indicators, the more unstable the film's operating state, and the more significant the negative impact on overall synergy. The penalty intensity decreases from a negative penalty to a positive penalty for the elastic coefficient. and Adjustments were made.

[0076] By combining these positive reward factors and negative penalty factors in the form of a power function, the model can dynamically output a comprehensive multi-field coupled steady-state synergy. This synergy index not only reflects the system's operating level under ideal conditions, but more importantly, it can sense and quantify the dynamic deviations caused by slippage and necking wire movement during the thin film operation in real time. This provides a comprehensive and accurate assessment of the system's disturbance rejection status, overcoming the limitations of traditional methods that isolate and monitor a single variable. It provides a more accurate quantitative benchmark for subsequent stretching accuracy control, enabling the servo drive to perform targeted high-frequency adaptive vibration suppression compensation.

[0077] By constructing a multi-field coupled steady-state synergy model, a global quantitative assessment of the multi-physics field disturbance resistance state during the production process is achieved. The model adopts a power function structure, with thermal stability coefficient and mechanical disturbance coefficient as positive benefit factors. Through elastic coefficient weighting, it reflects the contribution of thermodynamic and mechanical dynamic stability to the overall synergy, and can effectively characterize the system's operating level under ideal conditions.

[0078] Meanwhile, the real-time slip ratio and the standard deviation of the necking line position at the stretching point are introduced as negative penalty factors. Through the normalized power function form, the dynamic deviation during the operation of the thin film is transformed into a suppression effect on the degree of cooperation. This not only captures the damage to the system stability caused by the slip ratio and necking line movement, but also achieves differentiated responses to different disturbance sources by adjusting the elastic coefficient.

[0079] By organically combining positive benefits with negative penalties, the model can dynamically output a comprehensive synergy index, thereby solving the problem of inaccurate system state assessment under multi-physics field coupling interference. This enables the system to more accurately perceive the overall steady-state tolerance of the stretching process, providing a more targeted basis for high-frequency adaptive vibration suppression compensation for servo drives, and significantly improving the production yield and product quality of ultra-thin, highly oriented PET release films.

[0080] The real-time slip ratio is calculated by the difference ratio between the roller surface linear velocity fed back by the encoder and the actual linear velocity of the film measured by the tachometer; the standard deviation of the actual necking line position is obtained by continuously acquiring image sequences by the machine vision unit and using edge detection algorithm to track and calculate in real time.

[0081] Real-time slip ratio The standard deviation of the necking line position at the stretching point is calculated by the ratio of the difference between the roller surface linear velocity fed back by the encoder and the actual film linear velocity measured by the laser Doppler velocimeter. Image sequences are continuously acquired by a machine vision unit and edge detection algorithms are used for real-time tracking and calculation.

[0082] Real-time slip ratio The standard deviation of the necking line position at the stretching point is calculated by the ratio of the difference between the roller surface linear velocity fed back by the encoder and the actual film linear velocity measured by the laser Doppler velocimeter. Image sequences are continuously acquired by a machine vision unit and edge detection algorithms are used for real-time tracking and calculation.

[0083] Among them, real-time slip rate The slip ratio refers to the relative difference between the actual running speed of the film and the linear velocity of the stretching roller surface during the stretching process. This parameter directly reflects the stability of the traction relationship between the film and the stretching roller and is a key indicator for evaluating the dynamic stability of the stretching process. It can be calculated by obtaining the roller surface linear velocity through a high-precision rotary encoder installed on the stretching roller and directly measuring the actual running linear velocity of the film using a non-contact laser Doppler velocimeter. Then, the difference between the two velocities is calculated as a ratio to one of the velocities (e.g., the roller surface linear velocity). Another approach is to use an image processing-based method, capturing images of the contact area between the film and the roller surface using a high-speed camera and calculating the relative motion between the film and the roller surface using a feature point tracking algorithm to obtain the slip ratio.

[0084] Standard deviation of necking point position The standard deviation refers to the degree of fluctuation in the lateral position of the area where the film undergoes plastic deformation and thins rapidly during the film stretching process (i.e., the necking line). The stability of the necking line is directly related to the thickness uniformity and mechanical properties of the film. The calculation of this standard deviation can be achieved by a machine vision unit. For example, a machine vision unit can consist of one or more high-resolution industrial cameras, combined with an appropriate lighting system (such as a backlight or ring light), to continuously acquire a sequence of film images of the stretched area.

[0085] Subsequently, edge detection algorithms in image processing, such as Canny, Sobel, or Prewitt operators, are used to identify and extract the edge contours of the film, thereby determining the position of the necking line. By statistically analyzing the position of the necking line in consecutive frames of images, its standard deviation over a period of time can be calculated. Alternatively, a multi-point laser rangefinder array can be used to measure the width change of the film in the stretched region in real time. Combined with a pre-defined geometric model, the position and fluctuation of the necking line can be indirectly deduced.

[0086] By introducing high-precision physical quantification methods, key input parameters are provided for the multi-field coupled steady-state synergy model, thereby realizing closed-loop perception of the dynamic characteristics of the stretching process. By calculating the differential ratio between the roller surface linear velocity fed back by the encoder and the actual film linear velocity measured by the laser Doppler velocimeter, the system can capture the tiny relative slippage between the film and the stretching roller caused by nonlinear disturbances in real time. This overcomes the hysteresis caused by relying solely on roller speed control, enabling the system to directly quantify the traction stability of the film during the stretching process and providing core dynamic basis for evaluating the steady-state synergy under multi-field coupling.

[0087] Meanwhile, by continuously acquiring image sequences through a machine vision unit and using an edge detection algorithm to track and calculate the standard deviation of the necking line position in real time, the abstract movement of the stretching point can be transformed into a quantifiable statistical indicator. The necking line is a key area determining the thickness uniformity in the PET release film stretching process. By tracking its position fluctuations in real time, the system can accurately perceive the deformation stability of the film within the stretching gap. This edge detection technology based on visual feedback effectively solves the problem that traditional control methods cannot intuitively monitor the positional offset of the stretching point, ensuring that the coordination model can be evaluated based on the actual physical deformation state, thereby providing a high-confidence constraint benchmark for the subsequent adaptive vibration damping compensation of the servo drive.

[0088] The above technical solutions provide more accurate and real-time data on the film's operating status and the dynamic position of the stretching point for the multi-field coupled steady-state synergy model. This significantly improves the accuracy of the synergy model in assessing the disturbance rejection status of the production system. As a result, the subsequent optimization model can calculate a more accurate standard deviation of the high-frequency fluctuation of the target stretching roller speed based on more reliable input. This guides the servo driver to perform more targeted high-frequency adaptive vibration suppression compensation, effectively suppressing problems such as uneven thickness, surface scratches, and breakage caused by strong coupling interference of multiple physical fields during the film stretching process. This significantly improves the production yield and product quality of high-orientation PET release film.

[0089] The optimized model configuration uses a nonlinear relaxation-tightening algorithm to adaptively approximate the current fluctuations. The specific calculation process includes: Calculate the weighted average of the steady-state coordination degree of multi-field coupling and the air film interference coefficient, and perform a power operation greater than one on the weighted average to obtain the dynamic control factor; calculate the difference between the current actual high-frequency fluctuation standard deviation and the standard deviation of the minimum speed fluctuation of the servo system; multiply the difference with the dynamic control factor and attenuate it, then add it to the standard deviation of the minimum speed fluctuation, and output the high-frequency fluctuation standard deviation of the target stretching roller speed as the desired setpoint of the control closed loop.

[0090] The optimization model uses a nonlinear polynomial relaxation-tightening algorithm to adaptively approximate the current fluctuations, and its formula is as follows: in, The standard deviation of high-frequency fluctuations in the target stretching roll speed (unit: ); This represents the standard deviation of the high-frequency fluctuation of the current actual stretching roll speed (unit: ); This represents the standard deviation of the minimum achievable speed fluctuation of the servo system (unit: ); and These represent the weighting coefficients for coordination and airflow stability, respectively. Indicates the intensity index of regulation, and satisfies .

[0091] The optimization model here refers to a mathematical or computational model whose purpose is to achieve an optimal goal by adjusting system parameters. The nonlinear polynomial relaxation-tightening algorithm is an iterative optimization method that dynamically adjusts control parameters through nonlinear mathematical functions to gradually bring the system performance closer to an ideal state. This algorithm can adaptively adjust its approximation strategy based on real-time feedback of the system state, thereby achieving precise control of the speed fluctuation of the stretching roller in a complex and ever-changing production environment. For example, this algorithm can adopt a nonlinear optimization method based on gradient descent.

[0092] High-frequency fluctuation standard deviation of target stretching roll speed It is the upper limit of the expected speed fluctuation of the stretching roller under the current operating conditions. It is not a fixed value, but a value that is dynamically calculated based on the real-time status of the production system. It serves as the benchmark for high-frequency vibration suppression control of the servo drive. This target value aims to balance production efficiency and product quality, and ensure the best film stretching effect within the allowable fluctuation range.

[0093] Current standard deviation of high-frequency fluctuations in actual stretching roll speed This refers to the actual fluctuation degree obtained through statistical analysis of real-time monitoring data of the stretching roller's rotational speed. This value reflects the stability of the stretching roller's speed during the current production process and is one of the input parameters for the optimization model to make adaptive adjustments. For example, it can be obtained by high-pass filtering and statistical calculation of the rotational speed signal fed back from the servo motor encoder, and is the minimum standard deviation of the rotational speed fluctuation. Under ideal conditions, the minimum speed fluctuation level that the servo drive system's hardware and control algorithm can achieve is a physical limit determined by the inherent characteristics of the servo system. It represents the highest stability that the system can achieve without any external disturbances or under optimal control. This value is usually obtained through technical parameters provided by the servo system supplier or through precise testing in a laboratory environment.

[0094] Weighting ratio coefficient and Used to balance synergy and air film interference coefficient The relative importance of these coefficients in the optimization model can be determined based on actual production experience, thin film product characteristics, or through experimental optimization. For example, when a certain product is particularly sensitive to airflow disturbances, the coefficients can be appropriately increased. The weight of the multiphysics coupling effect can be increased; conversely, if the multiphysics coupling effect is more significant, the weight of the multiphysics coupling effect can be increased. The weight.

[0095] Regulation intensity index It is a nonlinear adjustment parameter used to control the strength of the optimization model's relaxation or tightening of the current fluctuation. When the system is in good condition, the index makes the optimization model more sensitive to the deterioration of the system state (i.e., a decrease in coordination or airflow stability), thus tightening the target fluctuation standard deviation more actively and prompting the system to implement stricter control. Conversely, when the system state is good, the adjustment of the target fluctuation standard deviation is relatively gentle. For example, the index can be set to 2, 3 or higher based on production experience to achieve different degrees of nonlinear response.

[0096] By introducing an optimization model and employing a nonlinear polynomial relaxation-tightening algorithm, the aim is to improve the steady-state synergy of multi-field coupling in the production system. and air film interference coefficient This translates into specific control commands for the high-frequency fluctuations in the stretching roll speed. The optimization model receives the standard deviation of the current actual high-frequency fluctuations in the stretching roll speed. The minimum standard deviation of speed fluctuation that the servo system can achieve. and preset weighting coefficients , Using the regulation intensity index m as input, during operation, the collaborative control module first calculates the multi-field coupling steady-state coordination degree based on the production parameters collected by the multi-source sensing module. and air film interference coefficient .

[0097] These coefficients comprehensively reflect the production system's resilience across multiple dimensions, including thermal stability, mechanical disturbance, slip ratio, standard deviation of necking point position, and airflow interference. Subsequently, the optimization model utilizes these coefficients and its embedded nonlinear polynomial relaxation-tightening algorithm to optimize the standard deviation of high-frequency fluctuations in the actual stretching roll speed. This algorithm performs adaptive approximation calculations using a nonlinear function. and The weighted average value is used as an adjustment factor, which is applied to the difference between the actual volatility and the minimum volatility. .

[0098] When the system is in good condition (i.e.) and When the target volatility standard deviation is high, the adjustment factor is large, resulting in a higher standard deviation of the target volatility. Towards Directional relaxation allows for a certain range of fluctuations; however, when the system state deteriorates (i.e., ... and When the level is lower, the adjustment factor decreases, making Towards Tightening the direction requires the servo system to perform more stringent fluctuation suppression. (Control intensity index) (and This further enhances the nonlinear response, ensuring that the control target can be tightened quickly and forcefully when the system is in poor condition.

[0099] Finally, the high-frequency fluctuation standard deviation of the target stretching roll speed was calculated. The data is sent to the servo driver in the stretching execution module, and the servo driver will... As the control reference constraint value of its high-frequency vibration suppression closed-loop control module, it can achieve precise and dynamic control of the stretching roller speed. This mechanism makes the control system no longer statically pursue a fixed fluctuation target, but can adaptively adjust the control strategy according to the real-time changes in the production environment, thereby effectively suppressing the stretching roller speed fluctuation caused by strong coupling interference of multiple physical fields, and significantly improving the control accuracy and product quality of the film stretching process.

[0100] By introducing an optimization model based on a nonlinear polynomial relaxation-tightening algorithm, the steady-state synergy of multi-field coupling is improved. and air film interference coefficient The comprehensive evaluation indicators are converted into the standard deviation of the high-frequency fluctuation of the target stretching roller speed that the servo driver can execute. This allows the control system to dynamically set a control benchmark that ensures production stability and meets the servo system's response capabilities based on the real-time disturbance rejection status of the production environment.

[0101] This achieves closed-loop optimization from system state assessment to precise control commands, avoiding the shortcomings of traditional control logic that lack dynamic perception and targeted compensation, through weighting coefficients. , With the flexible configuration of the intensity index m, the system can comprehensively consider the effects of multi-source disturbances such as thermodynamics, mechanical dynamics, and aerodynamics, and provide differentiated responses to different disturbance levels. When the system state is unfavorable, the target fluctuation standard deviation is [not specified]. The system will be tightened, prompting the servo drive to more actively suppress fluctuations; when the system is in good condition, it is allowed to relax appropriately to maintain the stability of production. This adaptive and precise control mechanism significantly improves the control accuracy and system robustness of the stretching process, effectively suppresses small fluctuations in the stretching process, and thus ensures the excellent quality and high yield of high-orientation PET release film.

[0102] The servo driver in the stretching execution module has an embedded high-frequency vibration suppression closed-loop control module, which is configured to receive the high-frequency fluctuation standard deviation of the target stretching roller speed output by the collaborative control module in real time as the control reference constraint value. When the actual measured high-frequency fluctuation standard deviation is higher than the control reference constraint value, the servo driver triggers the motor control algorithm to perform stiffness compensation and harmonic suppression.

[0103] The servo driver in the stretching execution module has an embedded high-frequency vibration damping closed-loop control module, configured to receive the output of the collaborative control module in real time. As a control baseline constraint value; when the actual measured high-frequency fluctuation standard deviation Higher than the target value At this time, the servo driver triggers the motor control algorithm to perform stiffness compensation and harmonic suppression.

[0104] The servo driver in the stretching execution module has an embedded high-frequency vibration damping closed-loop control module, configured to receive the output of the collaborative control module in real time. As a control baseline constraint value; when the actual measured high-frequency fluctuation standard deviation Higher than the target value At this time, the servo driver triggers the motor control algorithm to perform stiffness compensation and harmonic suppression.

[0105] The servo drive incorporates a high-frequency vibration suppression closed-loop control module. This module is an integrated functional unit within the servo drive, specifically designed to detect and suppress high-frequency vibrations generated by the motor or its driven load under high-speed or high-dynamic response conditions. The closed-loop control feature allows it to adjust in real time based on the deviation between the actual output and the desired output. This module can be implemented based on a high-performance digital signal processor (DSP) or a field-programmable gate array (FPGA). It uses high-speed sampling of motor current, speed, or position signals and employs advanced control algorithms (such as adaptive notch filtering, resonance suppression algorithms, and feedforward control) to counteract vibrations.

[0106] In addition, this module can also be implemented by integrating a dedicated vibration suppression algorithm library into the firmware of the servo driver. These algorithms can analyze the system's resonant frequency and generate a reverse compensation signal to counteract the inherent vibration of the mechanical system.

[0107] The servo drive is configured to receive the output of the cooperative control module in real time. As a control baseline constraint, this configuration aims to incorporate the macroscopic optimization results of the collaborative control module (i.e., the high-frequency fluctuation standard deviation of the target stretching roll speed) into the control baseline constraint. This information is used to transform the data into the basis for the micro-control of the servo drive. The servo drive can exchange data with the collaborative control module via industrial Ethernet (e.g., EtherCAT, Profinet) or high-speed serial communication interfaces (e.g., RS-485, CANopen), receiving data in real time. .

[0108] When the actual measured standard deviation of high-frequency fluctuations Higher than the target value At this time, the servo drive triggers the motor control algorithm to perform stiffness compensation and harmonic suppression. This is the core triggering mechanism of closed-loop control, ensuring that the servo drive can actively intervene to suppress vibrations when they exceed the allowable range. The controller inside the servo drive continuously monitors the actual high-frequency fluctuations of the motor speed or load and calculates its standard deviation. .

[0109] once Exceeding the preset The controller will activate preset stiffness compensation algorithms (e.g., increasing system stiffness by adjusting the current loop gain or speed loop gain) and harmonic suppression algorithms (e.g., canceling harmonic vibrations of the motor or mechanical system by injecting reverse harmonic current).

[0110] In addition, the servo drive can utilize its built-in vibration analysis function to analyze the motor's operating spectrum in real time and identify vibrations exceeding the limits. Once the high-frequency components of the threshold are identified, the driver will dynamically adjust its internal parameters, such as increasing the P gain of the proportional-integral-derivative (PID) controller to improve the response speed, or enabling a specific notch filter to attenuate vibrations at specific frequencies, thereby achieving stiffness compensation and harmonic suppression.

[0111] By integrating a high-frequency vibration suppression closed-loop control module into the servo driver, a closed-loop feedback link from the collaborative control model to the actuator is constructed, realizing the underlying hardware implementation of the control strategy. Based on production parameters collected in real time by multi-source sensing modules, the collaborative control module constructs thermal stability models, mechanical disturbance models, air film interference models, and multi-field coupling steady-state coordination models. Combined with the optimization model, it calculates the high-frequency fluctuation standard deviation of the target stretching roll speed. .

[0112] Should The value is sent in real time to the servo driver of the stretching execution module, and the servo driver receives this. Subsequently, this value was used as a control baseline constraint, and the actual high-frequency fluctuation standard deviation of the stretching roll was continuously monitored. When actual fluctuations Exceeding the target value At that time, the servo driver immediately triggers its embedded motor control algorithm to perform stiffness compensation and harmonic suppression. The core of this mechanism is to transform the macroscopic collaborative control logic into the microscopic motor drive response. By dynamically adjusting the output characteristics of the motor, it actively cancels the unexpected high-frequency vibration generated by the stretching roller during operation, thereby ensuring the stability of the stretching roller speed. This enables the system to transform theoretically optimized control into actual physical execution, effectively suppressing the stretching point movement caused by multi-physical field coupling interference, and providing a reliable execution guarantee for the production of high-precision, high-orientation PET release film.

[0113] By integrating a high-frequency vibration suppression closed-loop control module into the servo driver and receiving optimized target values ​​from the collaborative control module in real time, precise and adaptive suppression of high-frequency fluctuations in the stretching roller speed is achieved. This allows stiffness compensation and harmonic suppression during the stretching process to be dynamically adjusted according to actual working conditions, significantly improving the operational stability of the stretching roller. Consequently, it effectively suppresses the normal vibration of the film in the stretching gap and the drift of the stretching point, ultimately improving the production accuracy and product yield of high-orientation PET release film.

[0114] The terms "front," "back," "left," "right," "top," and "bottom" all refer to the figures in the accompanying drawings. Figure 1 Based on the perspective of the observer, the side of the device facing the observer is defined as the front, the left side of the observer is defined as the left, and so on.

[0115] In the description of this invention, it should be understood that the terms "center", "longitudinal", "lateral", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this invention.

[0116] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A high orientation PET release film production system with segmental draw precision control, characterized by, include: A multi-source sensing module is used to collect production parameters in real time. The multi-source sensing module includes a temperature detection unit, a displacement sensor, a differential pressure transmitter, a machine vision unit, an encoder, and a speed measuring instrument. The stretching execution module includes a preheating roller, a stretching roller, and a servo driver; A collaborative control module, communicatively connected to the multi-source sensing module and the stretching execution module, is used for: constructing a thermal stability model to obtain a thermal stability coefficient based on the standard deviation of the axial temperature of the preheating roller surface and the rate of temperature change obtained by the temperature detection unit; constructing a mechanical disturbance model to obtain a mechanical disturbance coefficient based on the radial runout of the stretching roller and the lateral vibration amplitude of the diaphragm obtained by the displacement sensor; constructing an air film interference model to obtain an air film interference coefficient based on the airflow micro-differential pressure fluctuation obtained by the micro-differential pressure transmitter and the edge retraction rate obtained by the machine vision unit; constructing a multi-field coupled steady-state synergy model to obtain a synergy degree based on the real-time slip rate obtained by the encoder and tachometer, the standard deviation of the necking line position of the stretching point obtained by the machine vision unit, and the thermal stability coefficient and mechanical disturbance coefficient; and constructing an optimization model based on the synergy degree, the air film interference coefficient, and the current speed fluctuation, calculating the high-frequency fluctuation standard deviation of the target stretching roller speed, and sending it to the servo driver for execution.

2. The precision controlled, high orientation PET release film production system of claim 1, wherein: The thermal stability model constructed by the collaborative control module is configured to perform exponential decay calculations, the specific calculation process of which includes: The ratio of the actual standard deviation of the axial temperature on the preheating roll surface to the maximum allowable standard deviation threshold of the axial temperature is made dimensionless, and the ratio of the absolute value of the rate of change of the actual preheating roll surface temperature over time to the maximum allowable rate of change threshold of the temperature is also made dimensionless. The two sets of dimensionless ratios are weighted and summed to obtain the comprehensive temperature disturbance term. The negative value of the comprehensive temperature disturbance term is used as the exponent to perform a monotonically decreasing exponential function operation, and the thermal stability coefficient with a value between zero and one is output.

3. A high orientation PET release film production system with segmental precision control of stretching according to claim 1, characterized in that: The mechanical disturbance model constructed by the collaborative control module is configured to perform rational fractional damping calculations, the specific calculation process of which includes: The ratio of the actual radial peak-to-peak runout of the stretching roller to the maximum allowable radial runout threshold is made dimensionless and a square term is calculated. Similarly, the ratio of the actual transverse vibration amplitude of the diaphragm to the critical vibration amplitude that triggers rupture is made dimensionless and a square term is calculated. Penalty coefficients are assigned to the two sets of square terms and the weighted sum is obtained to obtain a comprehensive mechanical penalty term. A rational fraction is constructed using the sum of the fundamental constant and the comprehensive mechanical penalty term as the denominator, and the mechanical disturbance coefficient that converges nonlinearly with mechanical disturbance is output.

4. The precision controlled, high orientation PET release film production system of claim 1, wherein: The air-film interference model constructed by the collaborative control module is configured to perform inverse trigonometric function scaling operations, the specific calculation process of which includes: Calculate the ratio of the actual micro-pressure difference fluctuation of the ambient airflow in the stretching region to the maximum critical pressure difference threshold, and the ratio of the actual edge retraction rate to the maximum allowable edge retraction rate threshold; assign response coefficients to the above two sets of ratios and sum them by weight to obtain the comprehensive interference term; substitute the comprehensive interference term into the arctangent function to obtain the smooth scaling value; subtract the smooth scaling value from the ideal state reference value to output the air film interference coefficient with the limit state saturation characteristics.

5. The precision controlled, high orientation PET release film production system of claim 1, wherein: The multi-field coupled steady-state synergy model is configured to perform a series of exponential function operations on multidimensional state variables. The specific operation process includes: Positive benefit elasticity exponents are assigned to the thermal stability coefficient and the mechanical disturbance coefficient, respectively; the ratio of real-time slip rate to maximum tolerable slip rate is introduced into the bias constant to construct a slip penalty term, and a negative penalty elasticity exponent is assigned to it; the ratio of the standard deviation of the actual necking line position to the boundary value of the maximum allowable free variance is introduced into the bias constant to construct a necking penalty term, and a negative penalty elasticity exponent is assigned to it; the system coefficient with the positive benefit elasticity exponent and the process penalty term with the negative penalty elasticity exponent are multiplied together to output the multi-field coupling steady-state synergy degree for evaluating the dynamic equilibrium of the system.

6. The precision controlled, high orientation PET release film production system of claim 5, wherein: The real-time slip ratio is calculated by the difference ratio between the roller surface linear velocity fed back by the encoder and the actual film linear velocity measured by the tachometer; the standard deviation of the actual necking line position is obtained by continuously acquiring image sequences by the machine vision unit and using an edge detection algorithm for real-time tracking and calculation. The real-time slip rate The standard deviation of the necking position at the stretching point is calculated by the ratio of the difference between the roller surface linear velocity fed back by the encoder and the actual linear velocity of the film measured by the laser Doppler velocimeter. Image sequences are continuously acquired by a machine vision unit and edge detection algorithms are used for real-time tracking and calculation.

7. The precision controlled, high orientation PET release film production system of claim 5, wherein: The optimization model is configured to perform adaptive approximation calculations on the current fluctuations based on a nonlinear relaxation-tightening algorithm. The specific calculation process includes: Calculate the weighted average of the multi-field coupling steady-state synergy degree and the air film interference coefficient, and perform a power operation greater than one on the weighted average to obtain the dynamic control factor; calculate the difference between the current actual high-frequency fluctuation standard deviation and the servo system's minimum speed fluctuation standard deviation; multiply the difference by the dynamic control factor and attenuate it, then add it to the minimum speed fluctuation standard deviation, and output the high-frequency fluctuation standard deviation of the target stretching roller speed as the desired setpoint of the control closed loop.

8. The precision controlled, high orientation PET release film production system of claim 7, wherein: The servo driver in the stretching execution module has a built-in high-frequency vibration suppression closed-loop control module, which is configured to receive the high-frequency fluctuation standard deviation of the target stretching roller speed output by the collaborative control module in real time as the control reference constraint value; when the actual measured high-frequency fluctuation standard deviation is higher than the control reference constraint value, the servo driver triggers the motor control algorithm to perform stiffness compensation and harmonic suppression. The servo driver in the stretching execution module has an embedded high-frequency vibration suppression closed-loop control module, configured to receive the output of the collaborative control module in real time. As a control baseline constraint value; when the actual measured high-frequency fluctuation standard deviation Higher than the target value At that time, the servo driver triggers the motor control algorithm to perform stiffness compensation and harmonic suppression.