Bio-based substrate face mask coating thickness control method based on online viscosity feedback

Abstract

The application provides a kind of dynamic control method of bio-based substrate mask coating thickness based on online viscosity feedback, it is related to bio-based substrate mask production technical field, comprising the following steps: in response to the speed change instruction signal from production line, call corresponding transient compensation curve, transient compensation curve is determined in advance based on the thixotropy of gel, it defines the sequence of compensation amount of metering pump reference speed with time changes in transient process;Within the duration of transient process, the compensation amount output in real time by transient compensation curve is superimposed on the current speed setting value of metering pump, to form composite pump speed control signal;According to composite pump speed control signal, drive metering pump to run, to suppress the coating thickness deviation caused by gel viscosity transient change.The scheme directly suppresses the discharge fluctuation caused by viscosity transient change, fundamentally effectively reduces the defective film length generated by each start-stop and speed change, greatly improves the production yield and material utilization.

Description

Technical Field

[0001] This application relates to the field of bio-based mask production technology, specifically to a dynamic control method for the coating thickness of bio-based masks based on online viscosity feedback. Background Technology

[0002] In the modern coating production of bio-based masks, ensuring uniform coating thickness is a core quality indicator. Current mainstream production processes typically employ a static control method with pre-set parameters. This involves setting fixed metering pump speeds, doctor blade gaps, and conveyor belt speeds on the production line based on the gel formulation and process parameters determined in the laboratory.

[0003] However, gel-like substances (such as carbomer and sodium alginate) exhibit significant thixotropic (shear-thinning) and non-Newtonian fluid properties, making their apparent viscosity extremely sensitive to temperature, shear history, and settling time. This leads to drastic and rapid dynamic changes in gel viscosity during actual production processes, especially during transient processes such as production line start-up, shutdown, and speed increases / decreases. Fixed pump speed parameters cannot respond to these instantaneous viscosity fluctuations, resulting in quality issues such as transient thickness control failure. During acceleration, the viscosity of the gel drops sharply due to high shear. If the pump speed is not adjusted accordingly, it will result in excessive discharge, forming an ultra-thick section. During deceleration or shutdown, the shear force disappears, causing the viscosity to recover rapidly. This increases pumping resistance, resulting in insufficient discharge and forming ultra-thin or broken coating sections. Each speed change produces a section of unusable defective product, severely restricting the improvement of the production yield. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, this application aims to provide a method for dynamic control of the coating thickness of a bio-based film based on online viscosity feedback, comprising the following steps: In response to a speed change command signal from the production line, the corresponding transient compensation curve is invoked. The transient compensation curve is pre-determined based on the thixotropic properties of the gel and is defined as a sequence of the amount of compensation for the reference speed of the metering pump changing over time during the transient process. During the duration of the transient process, the compensation amount output in real time by the transient compensation curve is superimposed on the current speed setting value of the metering pump to form a composite pump speed control signal. The metering pump is driven to operate according to the composite pump speed control signal in order to suppress coating thickness deviation caused by transient changes in gel viscosity.

[0005] According to the technical solution provided in this application, before invoking the transient compensation curve, the following steps are also included: The real-time online viscosity value of the gel before it enters the transient process is obtained and used as the first viscosity value. The process of invoking the transient compensation curve includes the following steps: Retrieve and traverse the pre-stored curve database to obtain a family of compensation curves corresponding to the speed change command signal. The family of compensation curves includes multiple compensation curves and the viscosity value corresponding to each compensation curve. The compensation curve in the family of compensation curves that corresponds to the first viscosity value is taken as the transient compensation curve.

[0006] According to the technical solution provided in this application, the pre-stored curve database is established through the following steps: For bio-based mask formulations, obtain the corresponding viscosity fluctuation range; N gel samples with different experimental viscosity values ​​were selected within the viscosity fluctuation range, and experiments were conducted on M common speed change modes on the production line: For each gel sample with the experimental viscosity value, the speed change corresponding to each speed change mode was performed, and the thickness deviation data on the time series was recorded synchronously at a high sampling frequency during the transient process; Based on the thickness deviation data recorded in each experiment, the ideal pump speed compensation sequence that varies with time to eliminate the thickness deviation is calculated by inversion. Each ideal pump speed compensation sequence is defined as a transient compensation curve, and it is associated with the corresponding speed change mode and the initial viscosity value of the gel sample used in the experiment to obtain the pre-stored curve database.

[0007] According to the technical solution provided in this application, the following steps are also included: During the steady-state operation of the production line, a preset step speed change of the metering pump is performed, and the system transmission delay time from the issuance of the pump speed command to the start of the flow response of the coating system is calculated. The step of adding the compensation amount output in real time from the transient compensation curve to the current speed setting value of the metering pump includes the following steps: Based on the system transmission delay duration, a timing alignment function F(t) = C(t - Td) is defined, where C(t) is the original compensation output function of the transient compensation curve with speed command triggering as the zero point, and Td is the system transmission delay duration; During the duration of the transient process, the compensation amount calculated in real time according to the timing alignment function is superimposed on the current speed setting value of the metering pump.

[0008] According to the technical solution provided in this application, the following steps are also included: After each compensation for a transient velocity change process is completed based on the transient compensation curve, the actual thickness deviation data for that transient process is obtained. The actual thickness deviation data is compared with the thickness deviation threshold, and the comprehensive performance index of the compensation is calculated. The comprehensive performance index is used to drive the self-optimization process of the transient compensation curve.

[0009] According to the technical solution provided in this application, after calculating the comprehensive effectiveness index of the compensation, the following steps are also included: If the comprehensive performance index is lower than or equal to the optimization trigger threshold, the current speed change command signal, the corresponding real-time online viscosity value, and the actual thickness deviation data will be stored as a set of optimization samples in a temporary database. When the number of optimized samples in the temporary database corresponding to the same velocity change pattern and with initial viscosity values ​​within the same viscosity range reaches a preset sample number threshold, the following steps are performed: The temporary database is used to call multiple sets of optimized samples that meet the conditions, and a new transient compensation curve is generated based on the same inversion calculation principle as the establishment of the pre-stored curve database. Replace the original transient compensation curve in the pre-stored curve database with the updated transient compensation curve, and clear the optimized samples already used in the temporary database.

[0010] According to the technical solution provided in this application, the method further includes the following steps: During the installation and commissioning phase of the production line, the first-order inertial time constant of the drive system from the issuance of the speed command to the actual stabilization of the linear velocity of the substrate on the production line was identified through step response testing. Based on the first-order inertial time constant of the drive system, a dynamic linear velocity prediction model for the substrate of the production line is established. This model is used to describe the actual dynamic response process of the substrate linear velocity under the speed change command.

[0011] According to the technical solution provided in this application, the step of adding the compensation amount calculated in real time based on the timing alignment function to the current speed setting value of the metering pump includes the following steps: Based on the dynamic linear velocity prediction model, the corresponding velocity scaling factor is calculated. The compensation amount output in real time by the transient compensation curve is multiplied by the speed proportional factor and then added to the current speed setting value of the metering pump.

[0012] According to the technical solution provided in this application, before driving the metering pump to operate based on the composite pump speed control signal, the following steps are further included: The transverse online viscosity values ​​at at least three key locations distributed along the width direction of the coating head are acquired in real time, and the real-time statistical variance of the transverse online viscosity values ​​is calculated and used as the system health index. The step of driving the metering pump to operate based on the composite pump speed control signal includes the following steps: If the system health index is less than or equal to a preset health threshold, the metering pump is driven to operate according to the composite pump speed control signal.

[0013] According to the technical solution provided in this application, after calculating the real-time statistical variance of the transverse online viscosity value and using it as a system health index, the method further includes the following steps: If the system health index is greater than the preset health threshold, a maintenance warning signal pointing to the coating head will be generated and output.

[0014] Compared with the prior art, the beneficial effects of this application are as follows: I. Achieving a leap from static to dynamic: It changed the static control mode that relied on fixed parameters, and realized real-time, adaptive dynamic control of coating thickness by responding to speed change commands and calling a dedicated transient compensation curve, enabling the control system to actively adapt to changes in production status.

[0015] Second, it solves the core pain point of transient thickness loss control: By using a compensation curve pre-determined for the thixotropic properties of the gel, the metering pump speed is precisely compensated at the moment of speed change in the production line, which directly suppresses the fluctuation of output caused by transient viscosity changes. This fundamentally and effectively reduces the length of defective film generated by each start-up, shutdown, and speed change, and significantly improves the production yield and material utilization rate.

[0016] Third, it has the ability to proactively control: It does not rely on lagging thickness feedback signals, but acts in advance based on impending events (speed changes) and the immediate state of the material (viscosity characteristics), thus avoiding thickness deviations at the source rather than correcting them after the fact, making the control effect more timely and accurate. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the steps of the dynamic control method for coating thickness of bio-based substrate films based on online viscosity feedback provided in this application. Detailed Implementation

[0018] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Example 1 As mentioned in the background section, to address the problems in the prior art, this application proposes a dynamic control method for the coating thickness of bio-based film based on online viscosity feedback, such as... Figure 1 As shown, it includes the following steps: S1. In response to the speed change command signal from the production line, the corresponding transient compensation curve is invoked. The transient compensation curve is pre-determined based on the thixotropic properties of the gel and is defined as the sequence of the compensation amount for the metering pump reference speed changing over time during the transient process. Specifically, the speed change command signal refers to the instruction issued by the production line control system to change the operating speed of the production line, usually transmitted in the form of digital or analog signals. This signal may contain information such as the target speed value and the rate of change. The transient compensation curve is a sequence of compensation data pre-determined experimentally and stored in the control system. This curve is based on the thixotropic properties of the gel, which refer to the characteristic of a gel-like fluid where its viscosity decreases under shear stress and gradually recovers after shearing stops. In actual implementation, when the control system detects a speed change command, it immediately retrieves the compensation curve matching the speed change pattern from the stored compensation curve library. The process of establishing the transient compensation curve includes: first, simulating the speed change process of the production line in an experimental environment, recording the change in gel viscosity over time using high-precision sensors; then analyzing the correspondence between viscosity change and coating thickness deviation; and finally, calculating the pump speed compensation amount sequence required to eliminate the thickness deviation through inversion. This sequence uses time as the independent variable and the compensation amount as the dependent variable to form a complete compensation curve. In implementation, the compensation curve is usually stored in a PLC or industrial computer in the form of a lookup table or mathematical function.

[0021] S2. During the duration of the transient process, the compensation amount output in real time by the transient compensation curve is superimposed on the current speed setting value of the metering pump to form a composite pump speed control signal. Specifically, the duration of a transient process refers to the time interval from the start of the speed change to the system reaching a steady state again. During this process, the control system reads the corresponding compensation value from the transient compensation curve according to a predetermined sampling period (e.g., 10ms). This compensation value is superimposed on the metering pump's reference speed setpoint by an adder to generate a composite pump speed control signal. For example, when the production line accelerates, the compensation curve may provide a positive compensation value, causing the metering pump speed to temporarily exceed the steady-state setpoint to compensate for the instantaneous reduction in output caused by the shear thinning effect.

[0022] S3. Drive the metering pump according to the composite pump speed control signal to suppress coating thickness deviation caused by transient changes in gel viscosity.

[0023] Specifically, after D / A conversion and signal conditioning, the composite pump speed control signal drives the servo motor or stepper motor of the metering pump to perform corresponding speed changes. The metering pump uses a high-precision gear pump or peristaltic pump, which can quickly respond to speed commands. Through this feedforward compensation control, fluctuations in the output caused by gel thixotropy can be effectively offset, thereby ensuring that the coating thickness remains stable during changes in production line speed.

[0024] This implementation scheme achieves highly stable control of coating thickness during production line speed variations. Traditional control methods often exhibit significant thickness deviations when speed changes, while this method, through a feedforward compensation mechanism, can control thickness fluctuations within ±2%, significantly improving product quality consistency. Simultaneously, this method reduces reliance on subsequent thickness detection feedback, achieving a more timely control response. Its scientific principle is based on the rheological properties of non-Newtonian fluids, particularly the thixotropic behavior of gel materials. When the production line speed changes, the shear rate experienced by the gel alters, causing its apparent viscosity to change over time. This viscosity change directly affects the output, thus influencing coating thickness deviations.

[0025] In a preferred embodiment, the step of invoking the transient compensation curve is further included: The real-time online viscosity value of the gel before it enters the transient process is obtained and used as the first viscosity value. The process of invoking the transient compensation curve includes the following steps: Retrieve and traverse the pre-stored curve database to obtain a family of compensation curves corresponding to the speed change command signal. The family of compensation curves includes multiple compensation curves and the viscosity value corresponding to each compensation curve. The compensation curve in the family of compensation curves that corresponds to the first viscosity value is taken as the transient compensation curve.

[0026] Specifically, the real-time online viscosity value is obtained through an online viscometer installed in the feed line. Commonly used online viscometers include rotary viscometers, vibrating plate viscometers, or capillary viscometers based on pressure difference calculations. The viscometer continuously measures the gel viscosity at a sampling frequency of 1-10 Hz and transmits the measured value to the control system. Before entering the transient process, typically referring to the steady-state operation phase before the speed change command is issued, the system takes the viscosity measurement value at this moment as the first viscosity value, which represents the basic rheological properties of the current batch of gel. The pre-stored curve database is a structured database stored in an industrial control computer or PLC, containing a large number of compensation curves pre-determined for different operating conditions. These curves are categorized and stored according to speed change patterns (such as acceleration, deceleration, specific change amplitude) and initial viscosity values. A compensation curve family refers to a set of compensation curves measured for the same speed change pattern but under different initial viscosity conditions. For example, for a speed change mode where the production line accelerates from 20 m / min to 30 m / min, the database stores compensation curves corresponding to multiple viscosity levels such as 500 cP, 1000 cP, and 1500 cP. These curves together constitute a family of compensation curves for this speed change mode. After acquiring the current initial viscosity value, the control system selects the best-matching compensation curve from the corresponding family of compensation curves using an interpolation algorithm. For example, if the measured viscosity is 1200 cP, and the database contains two curves at 1000 cP and 1500 cP, the system will calculate the compensation curve corresponding to 1200 cP through linear interpolation. This dynamic curve selection mechanism based on real-time viscosity can effectively adapt to viscosity differences between different batches of gel and viscosity changes within the same batch of gel during the production process.

[0027] The core effect of this scheme is the achievement of adaptive capability in the control system. By dynamically selecting the compensation curve based on real-time viscosity measurements, the system can automatically adapt to changes in material properties, significantly improving control accuracy and robustness. Especially when handling gel materials from different batches, with different formulations, or affected by ambient temperature, this adaptive mechanism ensures that the most suitable compensation parameters are always used, improving thickness control accuracy by more than 15%. Its technical principle is based on similarity theory and constitutive relations of non-Newtonian fluids in fluid mechanics. The thixotropic response characteristics of gels are closely related to their initial viscosity state; under different initial viscosities, the response law of gels to changes in shear rate exhibits systematic differences. This method, by establishing a family of compensation curves with initial viscosity as a parameter, essentially constructs a two-dimensional feedforward controller, with one dimension being time and the other being the initial viscosity state, thereby achieving a more accurate description and compensation of the process's dynamic characteristics.

[0028] In a preferred embodiment, the pre-stored curve database is established through the following steps: For bio-based mask formulations, obtain the corresponding viscosity fluctuation range; N gel samples with different experimental viscosity values ​​were selected within the viscosity fluctuation range, and experiments were conducted on M common speed change modes on the production line: For each gel sample with the experimental viscosity value, the speed change corresponding to each speed change mode was performed, and the thickness deviation data on the time series was recorded synchronously at a high sampling frequency during the transient process; Based on the thickness deviation data recorded in each experiment, the ideal pump speed compensation sequence that varies with time to eliminate the thickness deviation is calculated by inversion. Each ideal pump speed compensation sequence is defined as a transient compensation curve, and it is associated with the corresponding speed change mode and the initial viscosity value of the gel sample used in the experiment to obtain the pre-stored curve database.

[0029] Specifically, the viscosity fluctuation range was determined through extensive sample testing and statistical analysis of a specific formulation of gel. First, all viscosity samples that might occur during production were collected, including gels produced from different batches of raw materials and under different process parameters. The viscosity values ​​of these samples were measured using a laboratory viscometer, and then statistical analysis was conducted to determine their viscosity distribution range, for example, from 800 cP to 2000 cP. This range ensured that subsequent experiments could cover all viscosity conditions that might be encountered in actual production. Within the determined viscosity range, N representative viscosity points (e.g., 6 points) were selected at equal intervals or based on the actual distribution density. Simultaneously, historical operating data from the production line was analyzed to summarize M of the most common speed variation patterns (e.g., 4: low to medium speed, medium to high speed, high to medium speed, medium to low speed). For each viscosity sample, each speed variation pattern was sequentially executed on the experimental coating equipment to ensure that the experimental conditions were consistent with actual production. During the experiment, high-precision thickness measuring instruments, such as laser thickness gauges or beta-ray thickness gauges, were used to measure the coating thickness in real time at a sampling frequency of 100-1000 Hz. Simultaneously, the starting time of the velocity change is precisely recorded to ensure accurate correspondence between thickness data and the time series. Thickness deviation data refers to the difference between the measured thickness and the target thickness. This data is recorded according to the time series, forming a thickness deviation curve for each experimental condition. The inversion calculation is based on the principle of flow balance. First, a relationship model between thickness deviation and the required flow change is established. Then, based on the flow-speed characteristic curve of the metering pump, the flow change is converted into a pump speed compensation amount. The specific calculation process is as follows: for the thickness deviation at each sampling time, the theoretically required increase in output to eliminate the deviation is calculated, and then the corresponding pump speed compensation value is calculated based on the relationship between output and pump speed. Connecting the compensation values ​​at all time points forms an ideal pump speed compensation amount sequence. All the calculated compensation curves are organized according to two dimensions: velocity change mode and initial viscosity value, to establish a structured database. In the database, each compensation curve contains complete metadata information: the corresponding velocity change mode (such as starting velocity, target velocity, rate of change), gel initial viscosity value, compensation amount time series data, etc. This data is stored in the form of a matrix or data table for easy subsequent quick retrieval and retrieval.

[0030] In a preferred embodiment, the following steps are also included: During the steady-state operation of the production line, a preset step speed change of the metering pump is performed, and the system transmission delay time from the issuance of the pump speed command to the start of the flow response of the coating system is calculated. The step of adding the compensation amount output in real time from the transient compensation curve to the current speed setting value of the metering pump includes the following steps: Based on the system transmission delay duration, a timing alignment function F(t) = C(t - Td) is defined, where C(t) is the original compensation output function of the transient compensation curve with speed command triggering as the zero point, and Td is the system transmission delay duration; During the duration of the transient process, the compensation amount calculated in real time according to the timing alignment function is superimposed on the current speed setting value of the metering pump.

[0031] Specifically, steady-state operation refers to the stage where the production line runs stably at a constant speed. In this state, the control system sends a step speed change command with a known amplitude to the metering pump (e.g., a sudden increase in speed from 100 rpm to 120 rpm). Simultaneously, a high-response flow sensor (e.g., a Coriolis mass flow meter) monitors the actual flow rate change of the coating system in real time. The system transmission delay Td is obtained through precise timing: the time difference Δt = t2 - t1 between the time point t1 when the speed command is issued and the time point t2 when the flow sensor detects a significant change in flow rate is the system transmission delay. To ensure accuracy, this test can be repeated multiple times and the average value taken. The timing alignment function is a mathematical compensation function used to correct the time reference of the compensation curve. Here, C(t) is the original compensation output function of the transient compensation curve, with the speed command trigger as the zero point, and Td is the previously measured system transmission delay. The significance of this function is that it shifts the original compensation curve forward by Td time on the time axis, causing the compensation action to occur earlier, precisely offsetting the inherent delay of the system. In engineering implementation, this function can be achieved by modifying the triggering timing of the compensation curve in the PLC or control computer. When the speed change command is issued, the control system immediately starts the timing alignment function F(t). In each control cycle (e.g., 10ms), the system calculates the value of F(t) based on the current time t, i.e., obtains the pump speed compensation amount after delay compensation. This compensation amount is fused with the reference speed setpoint of the metering pump through a real-time superposition algorithm. For example, at t=Td, the original compensation curve C(t) has just begun to output the compensation amount, while the timing alignment function F(t) calculates the value of C(0) at t=Td. In fact, the compensation action has already started at t=0, which exactly cancels out the system delay of time Td.

[0032] This implementation achieves precise timing synchronization between viscosity compensation and production line dynamics. By compensating for the inherent delay of the compensation system, the timing accuracy of the compensation action is improved to the millisecond level, solving the problems of compensation being too late or too early due to delay. This timing alignment can further reduce the thickness fluctuation amplitude in transient processes, significantly improving control accuracy.

[0033] In a preferred embodiment, the following steps are also included: After each compensation for a transient velocity change process is completed based on the transient compensation curve, the actual thickness deviation data for that transient process is obtained. The actual thickness deviation data is compared with the thickness deviation threshold, and the comprehensive performance index of the compensation is calculated. The comprehensive performance index is used to drive the self-optimization process of the transient compensation curve.

[0034] Specifically, the transient process refers to the complete transition phase from the start of a velocity change to the system reaching stability again. After this process, the control system retrieves the actual thickness deviation data recorded during this time period from the thickness detection system (such as an online thickness gauge). This data includes complete information such as the time series of thickness deviations, the maximum deviation value, and the duration of the deviation. The data acquisition frequency is typically 50-100Hz to ensure that the complete dynamic response process can be captured. The thickness deviation threshold is a pre-set allowable deviation range based on product quality standards, such as ±3μm. The comparison process includes: checking whether the actual thickness deviation exceeds the threshold range at any moment, statistically analyzing the proportion of over-limit times, and calculating the sum of squares of the deviations. The comprehensive performance index is a numerical value that quantitatively evaluates the compensation effect. It is usually calculated using a weighted average of multiple indicators, including: the absolute value of the maximum deviation, the integral value of the deviation area (IAE), the overshoot, and the settling time. The smaller the index value, the better the compensation effect. The comprehensive performance index serves as the input signal for the compensation curve optimization algorithm. When the index value continuously deviates from the ideal range, a self-optimization process is triggered. The optimization process is based on machine learning or optimization algorithms, which automatically adjust the parameters or shape of the compensation curve by analyzing historical compensation effect data. For example, if systematic overcompensation is found after compensation, the amplitude of the compensation curve is reduced proportionally; if the response speed is found to be insufficient, the slope of the leading edge of the compensation curve is adjusted.

[0035] This implementation method achieves continuous self-optimization and quality self-improvement of the control system. By establishing a quantitative performance evaluation system, the system can automatically identify the quality of compensation effects and continuously improve control parameters accordingly. This closed-loop optimization mechanism enables the control system to adapt to long-term changing factors such as equipment aging and slow changes in material properties, always maintaining optimal control performance and keeping the product defect rate at a low level.

[0036] In a preferred embodiment, after calculating the comprehensive effectiveness index of the compensation, the following steps are further included: If the comprehensive performance index is lower than or equal to the optimization trigger threshold, the current speed change command signal, the corresponding real-time online viscosity value, and the actual thickness deviation data will be stored as a set of optimization samples in a temporary database. When the number of optimized samples in the temporary database corresponding to the same velocity change pattern and with initial viscosity values ​​within the same viscosity range reaches a preset sample number threshold, the following steps are performed: The temporary database is used to call multiple sets of optimized samples that meet the conditions, and a new transient compensation curve is generated based on the same inversion calculation principle as the establishment of the pre-stored curve database. Replace the original transient compensation curve in the pre-stored curve database with the updated transient compensation curve, and clear the optimized samples already used in the temporary database.

[0037] Specifically, the optimization trigger threshold is a pre-set performance threshold. When the overall performance index falls below this threshold, it indicates that the compensation effect is not ideal, and data needs to be recorded for future optimization. For example, if the threshold is set to 0.8 (normalized index), data recording is triggered when the measured value is 0.75. The optimization sample is a structured data set containing three key elements: detailed information on the velocity change command (change amplitude, rate, etc.), the real-time online viscosity value at the start of the transient process, and the actual thickness deviation time series data throughout the transient process. These samples are stored in a specially established temporary database. The system automatically classifies the samples in the temporary database according to the velocity change pattern and viscosity range. The velocity change pattern is identified and classified by comparing the initial value, target value, and rate of change of the velocity change. The viscosity range divides continuous viscosity values ​​into several discrete intervals, such as 0-1000 cP, 1000-2000 cP, etc. The preset sample size threshold is usually set to 5-10 groups to ensure sufficient statistical significance. When the number of samples in a certain category reaches the threshold, the optimization execution condition is triggered. The optimization process first extracts thickness deviation data from multiple sets of optimization samples that meet the criteria. Then, it employs the same inversion calculation principle as described above, but instead of relying solely on the statistical characteristics of the data, it calculates the values ​​based on these multiple sets. Specifically, this includes calculating the average or weighted average of the thickness deviation data to eliminate the influence of random interference; then, it inverts and calculates a new compensation sequence based on the average deviation curve. This new compensation sequence is the updated transient compensation curve, which incorporates experience from multiple actual runs and is more consistent with actual operating conditions than the original curve based on laboratory data. While ensuring the quality of the new curve, the system performs a curve update operation. The update process uses a version management mechanism, retaining the old curve as a backup while setting the new curve as the default. The sample data used to generate the new curve in the temporary database is cleared to make room for collecting new optimization samples. The entire update process is completed automatically without manual intervention, but the system generates an optimization log for technical personnel to review.

[0038] In a preferred embodiment, the method further includes the following steps: During the installation and commissioning phase of the production line, the first-order inertial time constant of the drive system from the issuance of the speed command to the actual stabilization of the linear velocity of the substrate on the production line was identified through step response testing. Based on the first-order inertial time constant of the drive system, a dynamic linear velocity prediction model for the substrate of the production line is established. This model is used to describe the actual dynamic response process of the substrate linear velocity under the speed change command.

[0039] Specifically, the step response test is conducted under no-load conditions on the production line. During the test, the control system sends a step speed command with a known amplitude to the production line drive motor (e.g., a sudden increase from 20 m / min to 25 m / min), while simultaneously monitoring the actual change in the substrate conveying speed in real time using a high-precision encoder or linear velocity sensor. The first-order inertial time constant of the drive system is obtained through curve fitting: the measured speed response curve is modeled using a first-order inertial element: v(t) = V0 + (V1 - V0)(1 - e^(-V ... (-t / τ) A least-squares fit is performed, where V0 is the initial velocity, V1 is the target velocity, and τ is the time constant to be identified. To ensure accuracy, the test is repeated 3-5 times in different velocity ranges, and the average value is taken. The dynamic linear velocity prediction model is a differential equation based on a first-order inertial element: dv(t) / dt = [v_cmd(t) - v(t)] / τ, where v_cmd(t) is the velocity command value, and v(t) is the predicted actual linear velocity of the substrate. In the discrete-time implementation, it is transformed into a difference equation using the Euler method: v(k+1) = v(k) + [v_cmd(k) - v(k)]×Δt / τ, where Δt is the control period. This model is implemented in a PLC or control computer with a built-in algorithm, and can predict the dynamic change process of the substrate linear velocity in real time based on a known velocity command sequence.

[0040] This implementation achieves accurate prediction of the dynamic changes in the linear velocity of the substrate. By establishing an accurate system dynamic model, the control system can predict the detailed process of velocity changes, providing an accurate time reference for feedforward compensation, reducing the phase error of velocity tracking, and significantly improving the synchronization of compensation control.

[0041] In a preferred embodiment, the step of adding the compensation amount calculated in real time according to the timing alignment function to the current speed setting value of the metering pump includes the following steps: Based on the dynamic linear velocity prediction model, the corresponding velocity scaling factor is calculated. The compensation amount output in real time by the transient compensation curve is multiplied by the speed proportional factor and then added to the current speed setting value of the metering pump.

[0042] Specifically, the speed scaling factor K_v is a dimensionless coefficient that changes in real time. The formula is: K_v(t) = v(t) / v_ref, where v(t) is the real-time predicted speed output by the dynamic linear velocity prediction model, and v_ref is the reference speed (usually the target speed or initial speed during the speed change process). This factor reflects the proportional relationship between the current instantaneous speed and the reference speed, gradually approaching the target scaling value from an initial value of 1 during speed changes. A multiplication step is added to the existing timing alignment compensation. In each control cycle, the system simultaneously reads the base compensation amount Δn_base(t) output by the transient compensation curve and the current speed scaling factor K_v(t), calculating the final compensation amount after speed scaling adjustment: Δn_final(t) = Δn_base(t) × K_v(t). This adjusted compensation amount is then superimposed on the reference speed setting of the metering pump. For example, during production line acceleration, as the actual linear speed increases, the speed scaling factor K_v gradually increases, correspondingly amplifying the compensation amount to ensure that the coating amount per unit area remains constant.

[0043] This implementation achieves precise coating amount control across the entire speed range. Through dynamic adjustment of the speed scaling factor, the compensation intensity automatically follows changes in linear speed, ensuring coating consistency across various operating conditions from low to high speeds. The adaptive compensation mechanism reduces the coefficient of variation in coating amount at different speeds.

[0044] In a preferred embodiment, before driving the metering pump to operate according to the composite pump speed control signal, the following steps are further included: The transverse online viscosity values ​​at at least three key locations distributed along the width direction of the coating head are acquired in real time, and the real-time statistical variance of the transverse online viscosity values ​​is calculated and used as the system health index. The step of driving the metering pump to operate based on the composite pump speed control signal includes the following steps: If the system health index is less than or equal to a preset health threshold, the metering pump is driven to operate according to the composite pump speed control signal.

[0045] Specifically, at least three representative monitoring locations are selected along the width of the coating head: the left end, the middle, and the right end. An independent online viscosity sensor is installed at each location, forming a lateral monitoring array. The sensors preferably employ a non-contact or bypass design to avoid disturbing the main fluid. The data acquisition system synchronously reads the viscosity values ​​at each location at a frequency of 1-5 Hz to ensure that dynamic changes in the lateral distribution can be captured.

[0046] The formula for calculating real-time statistical variance is: σ 2 = Σ(η_i - η_mean) 2 / (n-1), where η_i is the viscosity value at each measurement point, η_mean is the average value, and n is the number of measurement points (n≥3). The system health index HI is defined as the normalized variance value: HI=σ / η_mean, making it a dimensionless index, which facilitates the setting of a unified threshold. This index reflects the degree of deterioration of the lateral uniformity of the coating system; the larger the HI value, the worse the system health. The preset health threshold is set according to historical data and process requirements, usually between 0.05 and 0.15. The control system compares HI with the threshold in real time. When HI≤threshold, the system is considered to be in a healthy state, allowing normal compensation control to be executed. The composite pump speed control signal is directly transmitted to the metering pump driver to maintain normal production.

[0047] This implementation achieves status monitoring and intelligent protection of the coating system. By monitoring the lateral viscosity distribution, potential faults such as localized clogging of the coating head, distributor wear, and seal aging can be detected early, providing timely warnings before problems escalate. This preventative maintenance mechanism can reduce unplanned downtime by more than 30%, avoiding large-scale quality incidents.

[0048] Furthermore, after calculating the real-time statistical variance of the transverse online viscosity value and using it as a system health index, the process further includes the following steps: If the system health index is greater than the preset health threshold, a maintenance warning signal pointing to the coating head will be generated and output.

[0049] Specifically, when the system health index HI continuously exceeds a preset health threshold (e.g., 0.1) for a predetermined time (e.g., 30 seconds), the control system triggers an early warning mechanism. The maintenance early warning signal is a structured alarm message containing the following key information: the specific location of the anomaly (determined by analyzing the relative viscosity values ​​at each measurement point), the health index value, the duration of exceeding the threshold, and recommended maintenance measures (e.g., "The viscosity on the left side is abnormally high; it is recommended to check the left-end filter"). This signal is displayed via a human-machine interface (HMI) and simultaneously uploaded to the factory management system (MES). A three-level early warning mechanism is established: when HI is between 0.1 and 0.15, it is a Level 1 warning (observation level), the system continues to run but records the anomaly; between 0.15 and 0.25, it is a Level 2 warning (warning level), the system reduces its operating speed and prompts for planned maintenance; when it exceeds 0.25, it is a Level 3 warning (action level), the system recommends stopping the machine for inspection. Simultaneously, the system automatically records process data for a period of time prior to the anomaly, providing a complete historical record for fault diagnosis.

[0050] This implementation achieves a complete closed loop from anomaly detection to maintenance guidance. Through intelligent early warning grading and targeted maintenance suggestions, it significantly shortens the response time for fault diagnosis and handling, improving the efficiency and quality of maintenance work. Intelligent early warning can reduce the mean time to repair (MTBT).

[0051] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are only preferred embodiments of this application. It should be noted that due to the limitations of written expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this invention, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A method for dynamic control of coating thickness of bio-based film based on online viscosity feedback, characterized in that, Includes the following steps: In response to a speed change command signal from the production line, the corresponding transient compensation curve is invoked. The transient compensation curve is pre-determined based on the thixotropic properties of the gel and is defined as a sequence of the amount of compensation for the reference speed of the metering pump changing over time during the transient process. During the duration of the transient process, the compensation amount output in real time by the transient compensation curve is superimposed on the current speed setting value of the metering pump to form a composite pump speed control signal. Based on the composite pump speed control signal, the metering pump is driven to operate in order to suppress coating thickness deviation caused by transient changes in gel viscosity; It also includes the following steps: During the steady-state operation of the production line, a preset step speed change of the metering pump is performed, and the system transmission delay time from the issuance of the pump speed command to the start of the flow response of the coating system is calculated. The step of adding the compensation amount output in real time from the transient compensation curve to the current speed setting value of the metering pump includes the following steps: Based on the system transmission delay duration, a timing alignment function F(t) = C(t - Td) is defined, where C(t) is the original compensation output function of the transient compensation curve with speed command triggering as the zero point, and Td is the system transmission delay duration; During the duration of the transient process, the compensation amount calculated in real time according to the timing alignment function is superimposed on the current speed setting value of the metering pump. The method also includes the following steps: During the installation and commissioning phase of the production line, the first-order inertial time constant of the drive system from the issuance of the speed command to the actual stabilization of the linear velocity of the substrate on the production line was identified through step response testing. Based on the first-order inertial time constant of the drive system, a dynamic linear velocity prediction model for the substrate of the production line is established. This model is used to describe the actual dynamic response process of the substrate linear velocity under the speed change command. The step of adding the compensation amount, calculated in real time based on the timing alignment function, to the current speed setting of the metering pump includes the following steps: Based on the dynamic linear velocity prediction model, the corresponding velocity scaling factor is calculated. The compensation amount output in real time by the transient compensation curve is multiplied by the speed proportional factor and then added to the current speed setting value of the metering pump.

2. The method for dynamic control of coating thickness of bio-based film based on online viscosity feedback according to claim 1, characterized in that: Before invoking the transient compensation curve, the following steps are also included: The real-time online viscosity value of the gel before it enters the transient process is obtained and used as the first viscosity value. The process of invoking the transient compensation curve includes the following steps: Retrieve and traverse the pre-stored curve database to obtain a family of compensation curves corresponding to the speed change command signal. The family of compensation curves includes multiple compensation curves and the viscosity value corresponding to each compensation curve. The compensation curve in the family of compensation curves that corresponds to the first viscosity value is taken as the transient compensation curve.

3. The method for dynamic control of coating thickness of bio-based film based on online viscosity feedback according to claim 2, characterized in that: The pre-stored curve database is established through the following steps: For bio-based mask formulations, obtain the corresponding viscosity fluctuation range; N gel samples with different experimental viscosity values ​​were selected within the viscosity fluctuation range, and experiments were conducted on M common speed change modes on the production line: For each gel sample with the experimental viscosity value, the speed change corresponding to each speed change mode was performed, and the thickness deviation data on the time series was recorded synchronously at a high sampling frequency during the transient process; Based on the thickness deviation data recorded in each experiment, the ideal pump speed compensation sequence that varies with time to eliminate the thickness deviation is calculated by inversion. Each ideal pump speed compensation sequence is defined as a transient compensation curve, and it is associated with the corresponding speed change mode and the initial viscosity value of the gel sample used in the experiment to obtain the pre-stored curve database.

4. The method for dynamic control of coating thickness of bio-based film based on online viscosity feedback according to claim 1, characterized in that: It also includes the following steps: After each compensation for a transient velocity change process is completed based on the transient compensation curve, the actual thickness deviation data for that transient process is obtained. The actual thickness deviation data is compared with the thickness deviation threshold, and the comprehensive performance index of the compensation is calculated. The comprehensive performance index is used to drive the self-optimization process of the transient compensation curve.

5. The method for dynamic control of coating thickness of bio-based film based on online viscosity feedback according to claim 4, characterized in that: After calculating the comprehensive effectiveness index of this compensation, the following steps are also included: If the comprehensive performance index is lower than or equal to the optimization trigger threshold, the current speed change command signal, the corresponding real-time online viscosity value, and the actual thickness deviation data will be stored as a set of optimization samples in a temporary database. When the number of optimized samples in the temporary database corresponding to the same velocity change pattern and with initial viscosity values ​​within the same viscosity range reaches a preset sample number threshold, the following steps are performed: The temporary database is used to call multiple sets of optimized samples that meet the conditions, and a new transient compensation curve is generated based on the same inversion calculation principle as the establishment of the pre-stored curve database. Replace the original transient compensation curve in the pre-stored curve database with the updated transient compensation curve, and clear the optimized samples already used in the temporary database.

6. The method for dynamic control of coating thickness of bio-based film based on online viscosity feedback according to claim 1, characterized in that: Before driving the metering pump to operate based on the composite pump speed control signal, the following steps are also included: The transverse online viscosity values ​​at at least three key locations distributed along the width direction of the coating head are acquired in real time, and the real-time statistical variance of the transverse online viscosity values ​​is calculated and used as the system health index. The step of driving the metering pump to operate based on the composite pump speed control signal includes the following steps: If the system health index is less than or equal to a preset health threshold, the metering pump is driven to operate according to the composite pump speed control signal.

7. The method for dynamic control of coating thickness of bio-based film based on online viscosity feedback according to claim 6, characterized in that: After calculating the real-time statistical variance of the transverse online viscosity value and using it as a system health index, the process further includes the following steps: If the system health index is greater than the preset health threshold, a maintenance warning signal pointing to the coating head will be generated and output.

Images