Intelligent control method and device for photopolymerization process of flexible resin plate

Abstract

The application relates to the technical field of process control, and discloses a kind of intelligent control method and device for the photopolymerization reaction process of flexible resin plate.The method comprises the following steps: collecting light intensity, temperature and conversion rate feedback along the exposure partition, calculating the deviation based on the nominal model and obtaining the field model parameters through closed-loop setting, feeding forward the predicted deviation of downstream conversion rate and temperature according to the predicted deviation, calculating the adjustment amount of the transmission roller and the light source actuator, and driving the actuator to move after constraining the speed difference of adjacent partitions.The application solves the series of interrelated problems in the existing flexible resin plate exposure control scheme, such as the deviation of feedback object from the actual progress of the reaction, the lag of execution adjustment response behind the reaction heat release change, and the inconsistency of tension in adjacent areas caused by independent execution of partitions.

Description

Technical Field

[0001] This application relates to the field of process control technology, and in particular to an intelligent control method and apparatus for a flexible resin plate photopolymerization reaction process. Background Technology

[0002] Flexible resin plate exposure equipment typically allows for pre-setting of exposure time, light source power, and operating mode via a touchscreen. It also supports integration with other workstations on the production line via foot pedal control, I / O signal interfaces, or 0-10V analog input. Some UV curing light sources are equipped with a dual closed-loop feedback structure, using a temperature control module and a light power feedback algorithm to monitor and compensate for the lamp's junction temperature and output irradiance, maintaining long-term stability of the light source's output intensity. The determination of exposure process parameters relies on empirical settings based on the plate type and light source strength. Too short a main exposure time will result in overly straight image slopes, curved lines, and the washing away of fine details; too long a time will lead to plate adhesion and blurred text. This empirical setting logic has long constituted the primary method for exposure process control in this field.

[0003] However, existing closed-loop feedback structures on the light source side monitor and compensate for the stability of the light source's output. The feedback quantity remains focused on input parameters such as light intensity and lamp temperature, without addressing the actual cross-linking conversion degree of the resin layer—the reaction result itself. When there are differences in resin batches, ambient humidity, or photoinitiator activity, the same combination of light intensity and exposure time settings may correspond to different actual curing degrees, resulting in a disconnect between the controlled object and the result that the process truly cares about. In existing control schemes, conveyor belt speed and light source power are mostly set independently in an open-loop combination, lacking a mechanism for dynamic coordination based on the actual reaction progress. Once the light intensity fluctuates or the condition of the incoming plate changes, relying solely on a pre-set fixed speed cannot provide corresponding compensation.

[0004] Because ultra-thin flexible resin plates have low heat capacity, heat dissipation conditions vary at different locations along the width of the conveyor belt. If the overall, single-point parameter setting method is continued, it will be impossible to perceive the actual local heating rate differences in different zones of the plate during exposure, making it difficult to adjust the exposure intensity and conveyor speed of that local area. Even if closed-loop feedback is introduced to reduce control deviation, the inherent response lag of the feedback means that by the time abnormal changes in temperature or conversion rate are detected, the local reaction has already entered the self-accelerating exothermic stage, and post-correction alone cannot avoid uneven curing in local areas. Furthermore, if the detection and execution links are set independently by zone to improve the specificity of the response, the resulting speed difference between adjacent zones will create uneven traction on the plate in the conveyor direction. This raises a new technical problem: how to adjust the zones independently while avoiding local tension imbalance on the ultra-thin plate due to execution differences between adjacent zones. Summary of the Invention

[0005] This application provides an intelligent control method and device for the photopolymerization reaction process of flexible resin plates, which solves a series of interrelated problems in the existing exposure control schemes for flexible resin plates, such as the feedback object deviating from the actual reaction process, the execution adjustment response lagging behind the reaction exothermic changes, and the independent execution of zones causing inconsistent tension in adjacent areas. This enables the crosslinking conversion rate and surface temperature of the ultrathin flexible resin plate along the exposure zone to be maintained within the preset target range.

[0006] In a first aspect, this application provides an intelligent control method for the photopolymerization process of flexible resin plates, the intelligent control method for the photopolymerization process of flexible resin plates comprising: Step S1: Obtain the feedback amount of light source irradiance intensity, plate surface temperature and resin crosslinking conversion rate corresponding to the exposure zone to obtain the feedback amount group; Step S2: Based on the nominal model of the controlled object preset in the controller, perform state calculation on the feedback quantity group to obtain the theoretical conversion rate increment. Compare the difference between the theoretical conversion rate increment and the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group to obtain the model deviation. Based on the model deviation, perform closed-loop tuning on the model parameters in the nominal model of the controlled object to obtain the field model parameters corresponding to the exposure zone. Substitute the field model parameters into the nominal model of the controlled object and perform feedforward calculation on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone as it moves to the downstream detection station. Based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature, obtain the predicted control deviation. Step S3: Calculate the speed adjustment amount and light intensity adjustment amount for the conveyor roller actuator and the light source actuator respectively based on the predicted control deviation; Step S4: Constrain the rotation speed adjustment amount according to the difference between the rotation speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount, and control the conveyor roller actuator and the light source actuator to operate according to the execution adjustment amount.

[0007] Secondly, this application provides an intelligent control device for the photopolymerization process of flexible resin plates, the intelligent control device for the photopolymerization process of flexible resin plates comprising: The acquisition module is used to acquire the feedback amount of light source irradiation intensity, plate surface temperature and resin crosslinking conversion rate corresponding to the exposure zone, and obtain the feedback amount group; The calculation module is used to perform state calculation on the feedback quantity group based on the nominal model of the controlled object preset in the controller to obtain the theoretical conversion rate increment. The difference between the theoretical conversion rate increment and the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group is compared to obtain the model deviation. The model parameters in the nominal model of the controlled object are closed-loop tuned according to the model deviation to obtain the field model parameters corresponding to the exposure zone. The field model parameters are substituted into the nominal model of the controlled object to perform feedforward calculation on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone as it moves to the downstream detection station. The predicted control deviation is obtained based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature. The calculation module is used to calculate the speed adjustment amount and light intensity adjustment amount respectively for the conveyor roller actuator and the light source actuator based on the predicted control deviation. The control module is used to constrain the rotation speed adjustment amount based on the difference between the rotation speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount, and control the conveyor roller actuator and the light source actuator to operate according to the execution adjustment amount.

[0008] Thirdly, a smart control device for the photopolymerization process of flexible resin plates is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the smart control device for the photopolymerization process of flexible resin plates to execute the aforementioned smart control method for the photopolymerization process of flexible resin plates.

[0009] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the above-described intelligent control method for the photopolymerization reaction process of flexible resin plates.

[0010] The technical solution provided in this application obtains and integrates the feedback quantities of light source irradiance, plate surface temperature, and resin crosslinking conversion rate corresponding to the exposure zone into a feedback quantity group. This transforms the feedback object relied upon by the controller from the output parameters of the light source itself to the actual crosslinking process of the resin layer. Compared with existing solutions that only provide closed-loop feedback on light intensity or lamp temperature, this avoids the disconnect between the same set of light intensity and exposure time settings corresponding to different actual curing degrees under different resin batches and environmental conditions. By performing state estimation on the feedback quantity group and comparing the difference with the measured increment to obtain the model deviation, the model parameters in the nominal model of the controlled object are then closed-loop tuned to obtain the field model parameters. The controller no longer uses the factory-specified reaction rate, but gradually adjusts it to the actual reactivity level of the resin based on the measured data generated by the exposure zone. This tuning process enables the controller to adapt to factors such as resin batch differences and environmental humidity fluctuations. By substituting the field model parameters into the nominal model of the controlled object and performing feedforward calculations on the feedback quantity group, the predicted conversion rate and predicted temperature of the exposure zone when it reaches the downstream detection station are obtained. This allows the controller to know the deviation trend that will appear in the zone before the reaction reaches the self-accelerating exothermic stage. Compared with the conventional feedback method that only makes post-event adjustments based on the measured deviation at the current moment, this provides a lead time for subsequent adjustment actions.

[0011] Based on the predictive control deviation, the speed adjustment amount and light intensity adjustment amount are calculated separately for the conveyor roller actuator and the light source actuator. This transforms the two process variables, conveyor speed and light source intensity, which were originally set independently, into a dual-channel output that is calculated collaboratively based on the same set of predictive deviations. This overcomes the shortcomings of the existing scheme, where speed and light intensity are set independently and lack dynamic coordination based on actual process. The speed adjustment amount is constrained based on the difference between the speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount. This ensures that the adjustment amount calculated independently for each zone is checked for consistency between adjacent areas before execution. This avoids the new problem of inconsistent local tension on ultra-thin plates caused by the improved local response of zoned control. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram of an embodiment of the intelligent control method for the photopolymerization process of flexible resin plates in this application. Figure 2This is a schematic diagram illustrating the change in model deviation along the exposure zone number after closed-loop tuning of the field model parameters in the embodiments of this application. Figure 3 This is a schematic diagram comparing the conversion rate prediction deviation and temperature prediction deviation for each exposure zone in the embodiments of this application. Detailed Implementation

[0014] This application provides an intelligent control method and apparatus for the photopolymerization process of flexible resin plates. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0015] This application presents an intelligent control method for the photopolymerization process of flexible resin plates, applied to an exposure device with a conveyor belt laid along the conveying direction. The exposure device is divided into multiple exposure zones along the width of the conveyor belt. Each exposure zone is equipped with an independent light source actuator and a conveyor roller actuator. The light source actuator adjusts the light source irradiation intensity of the corresponding area of ​​the exposure zone, and the conveyor roller actuator adjusts the conveyor belt speed of the corresponding area of ​​the exposure zone. The light source actuator and conveyor roller actuator of each exposure zone are independently controlled by a controller, and there is no mechanical coupling between them. Along the conveying direction, each exposure zone is equipped with an upstream detection station and a downstream detection station. The upstream detection station is equipped with an irradiation intensity sensor, a surface temperature sensor, and a cure degree sensor to obtain the real-time light source irradiation intensity value, real-time plate surface temperature value, and real-time resin crosslinking conversion rate value corresponding to the exposure zone. The downstream detection station is set at a preset fixed distance from the upstream detection station along the conveying direction to re-inspect the plate after exposure in the exposure zone, verifying whether the adjustment amount corresponding to the exposure zone has enabled the plate to reach the range limited by the preset target conversion rate and preset maximum temperature. There are no physical barriers between two adjacent exposure zones. The flexible resin plate is laid continuously across multiple exposure zones along the width of the conveyor belt. The same plate may be in different reaction processes in different areas of different exposure zones due to differences in heat dissipation conditions or incoming material conditions.

[0016] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the intelligent control method for the photopolymerization process of flexible resin plates in this application includes: Step S1: Obtain the feedback amount of light source irradiance intensity, plate surface temperature and resin crosslinking conversion rate corresponding to the exposure zone to obtain the feedback amount group; Specifically, the acquired feedback quantities of light source irradiance intensity, plate surface temperature, and resin crosslinking conversion rate are not isolated instantaneous readings. The resin crosslinking conversion rate feedback quantity refers to the value converted from the attenuation ratio of the spectral absorption intensity of unreacted double bonds on the surface of the flexible resin plate at a specific wavelength. The value ranges from 0 to 1, and the closer the value is to 1, the more likely the resin in that zone has become fully crosslinked. The feedback quantity group is a data set that organizes the three types of feedback quantities measured at several consecutive sampling times in the same exposure zone according to two dimensions: zone number and sampling time. This organization method is a prerequisite for subsequent steps to compare, differ, and substitute data from the same zone. If there is no correspondence between zone number and time among the feedback quantities, subsequent steps will not be able to determine which data points belong to the continuous observation of the same plate zone.

[0017] Step S2: Based on the nominal model of the controlled object preset in the controller, perform state calculation on the feedback quantity group to obtain the theoretical conversion rate increment. Compare the difference between the theoretical conversion rate increment and the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group to obtain the model deviation. Based on the model deviation, perform closed-loop tuning on the model parameters in the nominal model of the controlled object to obtain the field model parameters corresponding to the exposure zone. Substitute the field model parameters into the nominal model of the controlled object and perform feedforward calculation on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone moving to the downstream detection station. Based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature, obtain the predicted control deviation. Specifically, the nominal model of the controlled object refers to a first-order differential equation model based on the relationship between the photopolymerization reaction rate and the resin surface temperature, the light source irradiation intensity, and the crosslinking conversion rate. This model includes factory-calibrated parameters such as frequency factor and activation energy, which are measured by the resin supplier under given environmental conditions and belong to the initial settings of the model. The state extrapolation is the process of substituting the light source irradiation intensity feedback and the plate surface temperature feedback from the feedback quantity group into this differential equation, and progressively calculating the theoretical conversion rate increment forward according to the preset extrapolation time. The preset extrapolation time corresponds to the actual exposure interval length of the exposure zone. If the value is too short, the extrapolation result will only reflect the change trend in a very short time and will not be able to reflect the complete reaction process; if the value is too long, the extrapolation error will accumulate. Model deviation reflects the difference between nominal parameters and the actual reaction rate of the resin. Closed-loop tuning corrects model parameters such as the frequency factor according to this difference, based on a preset learning rate, and limits it within a preset parameter threshold range. This avoids excessive single corrections that could cause the model parameters to deviate from the physically reasonable range. The on-site model parameters are the set of parameters used to describe the actual reaction characteristics of the resin in that exposure zone after correction. The preset target conversion rate in the predictive control deviation is 0.92, which is the curing endpoint determined by considering both the printing plate's durability and the strength requirements of the image structure. The preset maximum temperature is 45 degrees Celsius, which is the upper limit below which ultra-thin plates will not warp or deform.

[0018] Step S3: Based on the predicted control deviation, calculate the speed adjustment amount and light intensity adjustment amount for the conveyor roller actuator and the light source actuator respectively. Specifically, the calculation of the speed adjustment and light intensity adjustment is based on the conversion rate prediction deviation and temperature prediction deviation in the predictive control deviation. The response speed of the conveyor roller actuator is faster than that of the light source actuator. The light source itself has thermal inertia. Therefore, the temperature prediction deviation is mainly compensated by the speed adjustment, and the conversion rate prediction deviation is mainly compensated by the light intensity adjustment. However, it should be noted that this division of labor based on response speed solves the problem that single variable adjustment cannot take into account the two different time scale disturbances of reaction exothermic lag and conversion rate cumulative lag. The effect of the conveyor roller actuator on the plate surface temperature is achieved by changing the residence time of the exposure zone in the irradiation zone. This effect takes effect synchronously with the speed command. However, due to the heat storage of the lamp source, the effect of the light source actuator on the plate surface temperature lags behind the light intensity command. The difference in response speed between the two in the temperature direction constitutes the direct basis for the above division of labor.

[0019] Step S4: Constrain the rotation speed adjustment amount according to the difference between the rotation speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount, and control the conveyor roller actuator and the light source actuator to operate according to the execution adjustment amount.

[0020] Specifically, the difference constraint processing between the speed adjustment amounts of adjacent exposure zones is aimed at the problem that local stretching is easily generated during the conveying process when the heat capacity of ultra-thin resin plates is small and the adjustment amounts of adjacent zones are inconsistent due to differences in reaction exothermic or heat dissipation conditions. The execution adjustment amount is the final value actually issued to the conveying roller actuator and the light source actuator after constraint processing. This value directly drives the actuator to complete the update of the process conditions of the current exposure zone, so that when the plate leaves the exposure area, the crosslinking conversion rate and temperature of each zone are within the range limited by the preset target conversion rate and the preset maximum temperature.

[0021] In one specific embodiment, step S1 includes: Based on the irradiance sensor, surface temperature sensor and curing degree sensor set in the upstream detection station corresponding to the exposure zone, the exposure zone is synchronously detected and processed to obtain the real-time light source irradiance value, real-time plate surface temperature value and real-time resin crosslinking conversion rate value corresponding to multiple consecutive sampling times. Based on the real-time light source irradiance value, real-time plate surface temperature value, and real-time resin crosslinking conversion rate value corresponding to two adjacent sampling times, differential calculation processing is performed on the exposure zone to obtain the light source irradiance change rate, plate surface temperature change rate, and resin crosslinking conversion rate change rate. Based on the real-time light source irradiance value, real-time plate surface temperature value, real-time resin crosslinking conversion rate value, light source irradiance change rate, plate surface temperature change rate and resin crosslinking conversion rate change rate, the exposure zones are classified and integrated to obtain the light source irradiance feedback quantity, plate surface temperature feedback quantity and resin crosslinking conversion rate feedback quantity corresponding to the exposure zones. The feedback quantities of light source irradiance intensity, plate surface temperature, and resin crosslinking conversion rate corresponding to the exposure zone are input into the zone data association module for marking and encapsulation processing to obtain the feedback quantity group.

[0022] Specifically, the cure degree sensor employs a specific wavelength spectral absorption principle. A beam of detection light of a fixed wavelength is irradiated onto the surface of the exposed zone. The carbon-carbon double bonds in the resin that have not participated in the cross-linking reaction produce characteristic absorption in this wavelength band. The absorption intensity gradually decreases as the double bonds are consumed. The ratio difference between the absorption intensity measured at the current moment and the initial absorption intensity when the resin is completely uncured is the real-time resin cross-linking conversion rate value, ranging from 0 to 1. A larger value indicates a higher degree of curing in that zone. The fixed wavelength is the wavelength corresponding to the near-infrared absorption peak around 1620 nm. This wavelength is located in the wavelength band where the characteristic absorption peak of the carbon-carbon double bonds in the resin is located. The spectral absorption principle selects this wavelength band instead of other wavelength bands because the absorption peak of the carbon-carbon double bonds in this band has the lowest degree of overlap with the background absorption of the resin matrix, substrate, and other excipients, thus avoiding interference from the absorption signals of other substances in the detection of the degree of double bond consumption. The initial absorption intensity of the resin when it is completely uncured is taken from the absorption intensity reading measured by the curing degree sensor before the plate corresponding to the exposure zone enters the upstream detection station and before entering the exposure area. Each plate undergoes an absorption intensity acquisition in the uncured state before exposure begins. This reading serves as the benchmark value for the ratio difference calculation during the production process of this plate. The initial absorption intensity varies between different batches of resin due to differences in formulation or photoinitiator activity. Therefore, each plate is measured individually when it is put into production instead of using a fixed factory experience value, thereby avoiding deviations in the conversion rate calculation benchmark due to batch differences.

[0023] The time interval between multiple consecutive sampling moments is 0.5 seconds. This interval is less than the time scale required for the resin surface temperature to change significantly due to the exothermic reaction. It can capture the early trend before the reaction rate changes abruptly. If the interval is too large, the resin will have already undergone a significant change in the reaction process between two adjacent sampling moments. The change rate calculated in this way cannot reflect the instantaneous reaction state. If the interval is too small, it will cause data redundancy and increase the computational burden without substantially improving the numerical accuracy of the change rate. The differential calculation process calculates the difference between the same physical quantity measured at two adjacent sampling moments and divides it by the sampling time interval to obtain the change amplitude of the physical quantity per unit time. This change rate, together with the instantaneous value, constitutes complete information describing the current process state of the exposure zone. Relying solely on the instantaneous value cannot determine whether the state of the zone is trending towards stability or is changing rapidly. Relying solely on the change rate cannot determine the current absolute level.

[0024] The classification and integration process categorizes and integrates six values—three instantaneous values ​​and three rates of change—corresponding to the same exposure zone and the same sampling time. These values ​​are then grouped according to three data lines: light source irradiance, plate surface temperature, and resin crosslinking conversion rate. Each data line contains both the instantaneous value and the rate of change of the physical quantity. The resulting three sets of values ​​are the light source irradiance feedback, plate surface temperature feedback, and resin crosslinking conversion rate feedback. The labeling and encapsulation process adds the corresponding exposure zone number and sampling time label to these three sets of feedback values ​​to prevent confusion between feedback values ​​generated in different zones and at different times during subsequent processing. The data set obtained after adding the labels is the feedback value group.

[0025] In one specific embodiment, in step S2, based on the nominal model of the controlled object preset in the controller, the state of the feedback quantity group is calculated to obtain the theoretical conversion rate increment. The theoretical conversion rate increment is then compared with the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group to obtain the model deviation, including: Based on the feedback quantities of light source irradiance intensity, plate surface temperature, and preset initial values ​​of field model parameters in the feedback quantity group, the nominal model of the controlled object is substituted and constructed to obtain the reaction rate differential equation. Based on the preset calculation time, the reaction rate differential equation is iteratively integrated to obtain the theoretical conversion rate increment corresponding to the exposure zone; Based on the real-time resin crosslinking conversion rate value at the start time of the preset estimated duration corresponding to the exposure zone, the difference calculation is performed on the real-time resin crosslinking conversion rate value at the end time of the preset estimated duration to obtain the measured increment. The deviation of the model is obtained by dividing the measured increment by the theoretical conversion rate increment.

[0026] Specifically, the nominal model of the controlled object is a first-order differential equation established based on the variation of the photopolymerization reaction rate with the resin surface temperature, light source irradiation intensity, and crosslinking conversion rate. The equation contains two parameters: frequency factor and activation energy. These two parameters are provided by the resin supplier based on the curing test of the resin model under standard environmental conditions. The substitution process involves filling the instantaneous values ​​of the light source irradiation intensity feedback and the plate surface temperature feedback corresponding to the current sampling moment in the feedback quantity group into the light intensity term and temperature term of the differential equation, respectively, and filling the initial values ​​of the field model parameters into the positions of the frequency factor and activation energy. This yields a reaction rate differential equation specific to the current exposure zone and the current moment. The initial values ​​of the field model parameters are directly taken from the nominal frequency factor and nominal activation energy provided by the resin supplier at the initial stage of the control system. As new deviation data are continuously generated in subsequent exposure zones, these initial values ​​will be gradually corrected. Here, they are only used as the starting point for calculation. In the first-order differential equation, the temperature term represents the law that the reaction rate increases with increasing temperature, which is represented by the exponential decay relationship between the activation energy and the surface temperature of the plate. The light intensity term represents the law that the reaction rate increases with increasing light intensity, which is represented by the product relationship between the frequency factor and the irradiance of the light source. The cross-linking conversion rate term represents the law that the reaction rate spontaneously decreases as the reactants are gradually consumed, which is represented by the proportion of resin that has not yet participated in the cross-linking reaction. The three terms are multiplied together to form the complete expression on the right side of the first-order differential equation, and the left side of the equation is the first derivative of the resin cross-linking conversion rate with respect to time.

[0027] The preset calculation duration corresponds to the actual length of the exposure interval where the exposure partition is located. It is used to limit the time span for the integral calculation to extrapolate forward. In this embodiment, the value is 2 seconds. This value corresponds to the typical exposure interval span obtained by dividing the physical length of the exposure partition along the conveyor direction by the normal operating speed of the conveyor belt. If the duration is set too short, the calculation result will only reflect the instantaneous changes in a very short time and will not be able to represent the cumulative reaction process of the partition in the entire exposure process. If the duration is set too long, the integral error will increase with time because the reaction rate itself changes continuously with temperature and conversion rate.

[0028] Iterative integration processing involves progressively extrapolating the already defined reaction rate differential equation using extremely short time steps, each step being 0.01 seconds. This step size is less than one-hundredth of the preset extrapolation time. A step size that is too large will increase the error generated when approximating the continuous change of the reaction rate over time, while a step size that is too small will significantly increase the computational load without noticeably improving the extrapolation accuracy. The conversion rate increment obtained from each extrapolation step serves as the starting conversion rate for the next extrapolation step. This process is repeated until the preset extrapolation time is reached. The increments obtained from each extrapolation step are then summed to obtain the theoretical conversion rate increment corresponding to that exposure zone. Compared to a one-time substitution solution, this progressive extrapolation method can adapt to the objective law that the reaction rate is not constant but gradually decreases with the accumulation of conversion rate. The measured increment is the difference between the real-time resin crosslinking conversion rate measured at the start time of the preset calculation duration and the real-time resin crosslinking conversion rate measured at the end time. This difference reflects the actual conversion rate change that occurs in the actual exposure process of the exposure zone. The model deviation is the ratio of the theoretical conversion rate increment to the measured increment. When the ratio is equal to 1, it means that the nominal model's description of the current reaction characteristics of the zone is consistent with the actual situation. When the ratio is greater than 1, it means that the reaction process calculated by the nominal model according to the current parameters is faster than the actual process of the resin, that is, the current actual reaction activity of the resin is lower than the nominal setting. When the ratio is less than 1, it means that the actual reaction activity of the resin is higher than the nominal setting.

[0029] In one specific embodiment, in step S2, the model parameters in the nominal model of the controlled object are closed-loop tuned according to the model deviation to obtain the field model parameters corresponding to the exposure zone, including: The parameter correction amount is obtained by weighting the preset tuning learning rate based on the model deviation. Candidate model parameters are obtained by superimposing the initial values ​​of the field model parameters based on the parameter correction amount; Based on a preset parameter threshold range, the candidate model parameters are subjected to boundary comparison processing to obtain the model parameters after boundary constraints. The model parameters in the nominal model of the controlled object are updated based on the model parameters after boundary constraints to obtain the field model parameters corresponding to the exposure zone.

[0030] Specifically, the preset learning rate is 0.3. The weighted processing multiplies the difference between the model deviation and 1 by this learning rate to obtain the parameter correction amount. When the model deviation is greater than 1, it indicates that the actual reactivity of the resin is lower than the nominal setting, and the difference is positive, and the parameter correction amount is also positive. When the model deviation is less than 1, the difference is negative, and the parameter correction amount is also negative. A learning rate of 0.3 means that each tuning only corrects 30% of the deviation rather than correcting the entire deviation at once. If the value is too high, the model parameters will fluctuate significantly and repeatedly across multiple exposure zones. If the value is too low, more data from multiple exposure zones will be needed for the model parameters to converge to a level that closely reflects the actual reactivity characteristics of the resin. The parameter correction amount applies to the frequency factor. The candidate model parameter is obtained by adding the parameter correction amount to the initial value of the current frequency factor. This addition operation reflects that the frequency factor is adjusted in the same direction as the actual reactivity of the resin deviates from the nominal setting in terms of both direction and magnitude.

[0031] The lower limit of the preset parameter threshold range is 0.7 times the nominal frequency factor value, and the upper limit is 1.3 times the nominal frequency factor value. The boundary comparison process compares the candidate model parameters with the upper and lower limits of the range one by one. When the candidate model parameter falls within the range, it is directly used as the model parameter after boundary constraint. When the candidate model parameter exceeds the upper limit, the upper limit value is used as the model parameter after boundary constraint. When it is below the lower limit, the lower limit value is used as the model parameter after boundary constraint. The basis for setting this range is that even if there are batch differences in the same type of resin, its actual frequency factor usually does not deviate from the nominal value by more than 30%. Exceeding this range often means that the test data itself is abnormal rather than the normal fluctuation of the resin characteristics. In this case, the original value of the candidate model parameter is not adopted, but the boundary value is used to avoid abnormal data pulling the model parameter into an unreasonable range. The update process involves replacing the boundary-constrained model parameters with the original frequency factor values ​​in the nominal model of the controlled object. The activation energy and other parameters in the model remain unchanged. The resulting field model parameters are identical to those of the original nominal model in terms of equation form, except that the frequency factor value has changed. The changed value reflects the actual reactivity level of the resin in the exposure zone under the current environmental and batch conditions.

[0032] Figure 2 This is a schematic diagram illustrating the change in model deviation along the exposure zone number after closed-loop tuning of the field model parameters in this embodiment of the application. Figure 2The horizontal axis represents the exposure zone number arranged in the order of exposure zones, and the vertical axis represents the model deviation. The solid line represents the model deviation value corresponding to the field model parameters obtained after closed-loop tuning, the dashed line represents the model deviation value corresponding to the nominal frequency factor without closed-loop tuning, and the dotted line represents the theoretical benchmark of model deviation equal to 1, which corresponds to the state where the nominal model and the current actual reaction characteristics of the resin are completely consistent. As can be seen from the figure, although the model deviation without closed-loop tuning gradually approaches the benchmark as the exposure zones progress, the convergence speed is slow, and it always maintains a value far from the benchmark in the first few exposure zones. However, the model deviation after closed-loop tuning converges to the vicinity of the benchmark within a smaller number of exposure zones and maintains a small fluctuation around it. This indicates that the process of updating the field model parameters by gradually superimposing parameter corrections and boundary constraints enables the nominal model of the controlled object to quickly approach the actual reactivity level of the batch of resin under the current field conditions.

[0033] In one specific embodiment, in step S2, the field model parameters are substituted into the nominal model of the controlled object, and feedforward calculation is performed on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone moving to the downstream detection station. Based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature, the predicted control deviation is obtained, including: Substitute the field model parameters into the nominal model of the controlled object for value assignment to obtain the field reaction rate differential equation. Based on the conveyor belt speed corresponding to the exposure zone, the travel time required for the exposure zone to travel from the upstream inspection station to the downstream inspection station is calculated to obtain the predicted travel time. Based on the predicted travel time and the feedback amount of plate surface temperature in the feedback group, the differential equation of on-site reaction rate is iteratively integrated according to the preset calculation time to obtain the predicted conversion rate and predicted temperature corresponding to the exposure zone traveling to the downstream detection station. The difference between the predicted conversion rate and the preset target conversion rate is calculated, and the difference between the predicted temperature and the preset maximum temperature is calculated to obtain the conversion rate prediction deviation and the temperature prediction deviation. The conversion rate prediction deviation and the temperature prediction deviation are then classified and integrated to obtain the prediction control deviation.

[0034] Specifically, the assignment process involves filling the tuned frequency factors from the field model parameters into the corresponding positions in the original nominal model. The resulting field reaction rate differential equation is actually a set of simultaneous equations. One term describes the rate of change of conversion rate with light intensity, temperature, and cross-linked conversion rate. The other term describes the change of resin surface temperature with the combined effects of reaction exothermics and external heat dissipation. This temperature change relationship is based on the physical law that the photopolymerization reaction itself releases heat, the temperature rise is determined by the resin density and specific heat capacity, and some heat is lost through heat exchange between the plate and the surrounding environment. The two equations share the same conversion rate variable, and the calculated conversion rate is directly used as the input of the exothermic term in the temperature change relationship. The two are calculated synchronously during the iteration process, which is why the field reaction rate differential equation obtained after the assignment process can simultaneously provide both the predicted conversion rate and the predicted temperature.

[0035] The predicted travel time is obtained by dividing the fixed distance between the upstream and downstream inspection stations by the conveyor belt speed corresponding to the exposure zone. This distance is determined to be 400 mm during equipment installation and is a structural parameter of the equipment rather than an adjustable parameter of the process. The conveyor belt speed is taken from the current actual operating speed recorded in the feedback quantity group. The predicted travel time obtained by dividing the two reflects the actual time required for the exposure zone to travel from the upstream station to the downstream station.

[0036] The preset calculation time is used as a fixed step size for each iteration of the integral in this step, with a value of 0.1 seconds. The iteration process is to repeatedly calculate according to this step size. The conversion rate increment and temperature change obtained in each step are added to the conversion rate and temperature values ​​of the previous step as the starting point of the next step. This process is repeated until the accumulated calculation time reaches the predicted travel time. The conversion rate and temperature obtained at this time are the predicted conversion rate and predicted temperature corresponding to the exposure zone traveling to the downstream detection station. This method of accumulating to the total time with a fixed step size can adapt to the actual situation that the conversion rate and temperature affect each other during the calculation process and the reaction rate itself is not constant. Compared with the method of substituting the total time at once, this method can adapt to the actual situation that the conversion rate and temperature affect each other during the calculation process and the reaction rate itself is not constant. The conversion rate prediction deviation is the value obtained by subtracting the predicted conversion rate from the preset target conversion rate of 0.92. The temperature prediction deviation is the value obtained by subtracting the predicted temperature from the preset maximum temperature of 45 degrees Celsius. The classification and integration process retains these two deviations separately, marks them according to their respective exposure zones and the corresponding downstream detection stations, and then combines them to form the prediction control deviation. Therefore, the prediction control deviation includes independent deviation values ​​in both the conversion rate and temperature directions, rather than merging the two into a single value. This method of retaining independent values ​​allows subsequent steps to call execution components with different response speeds to compensate for the conversion rate deviation and temperature deviation respectively.

[0037] Figure 3 This is a schematic diagram comparing the conversion rate prediction deviation and temperature prediction deviation for each exposure zone in the embodiments of this application. Figure 3 The horizontal axis represents several exposure zones arranged in sequence, the left vertical axis represents the conversion rate prediction deviation, the right vertical axis represents the temperature prediction deviation, the diagonally filled bars represent the conversion rate prediction deviation value corresponding to each exposure zone, and the cross-line filled bars represent the temperature prediction deviation value corresponding to each exposure zone. Both deviations are calculated from the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature. As can be seen from the figure, along the arrangement order of zones one to five, both prediction deviations show a gradually decreasing trend, and the relative magnitudes of the conversion rate prediction deviation and temperature prediction deviation within the same exposure zone are different. For example, both deviations in zone one are at a relatively high level in this set of data, while both deviations in zone five drop to a relatively low level; the relative magnitudes of the two deviations within the same zone are not completely consistent.

[0038] In one specific embodiment, step S3 includes: Based on the conversion rate prediction deviation and temperature prediction deviation corresponding to two adjacent sampling times, differential calculation processing is performed on the exposure zone to obtain the conversion rate prediction deviation change rate and temperature prediction deviation change rate. The weights of the rotation speed channel and the light intensity channel are obtained by normalizing the conversion rate prediction deviation and the temperature prediction deviation. Based on the temperature prediction deviation, the rate of change of the temperature prediction deviation, and the speed channel weight, the preset speed proportional-integral-derivative parameters are weighted and processed to obtain the speed adjustment amount. Based on the conversion rate prediction deviation, the conversion rate prediction deviation change rate, and the light intensity channel weight, the preset light intensity proportional integral differential parameters are weighted and processed to obtain the light intensity adjustment quantity.

[0039] Specifically, in the differential calculation process, two adjacent sampling times correspond to a set of conversion rate prediction deviations and temperature prediction deviations obtained by the exposure zone at different sampling times after continuous calculation through the preceding steps. The time interval between the two adjacent sampling times is the same as the time interval between multiple consecutive sampling times in step S1, which is 0.5 seconds. The conversion rate prediction deviation at the current sampling time is subtracted from the conversion rate prediction deviation at the previous sampling time, and then divided by the time interval between the two sampling times to obtain the conversion rate prediction deviation change rate. The temperature prediction deviation change rate is calculated in the same way. These two change rates reflect whether the prediction deviation itself is gradually expanding or gradually narrowing, and together with the instantaneous value of the deviation, they constitute the complete input required for subsequent proportional-integral-differential calculations.

[0040] Normalization first converts each prediction deviation into a dimensionless relative deviation, then allocates the output ratio of the two channels based on this relative deviation. The absolute value of the temperature prediction deviation is divided by the preset temperature deviation benchmark value to obtain the normalized temperature deviation value; similarly, the absolute value of the conversion rate prediction deviation is divided by the preset conversion rate deviation benchmark value to obtain the normalized conversion rate deviation value. The preset temperature deviation benchmark value is 5 degrees Celsius, representing the difference between the preset maximum temperature of 45 degrees Celsius and the allowable stable operating temperature of 40 degrees Celsius for continuous exposure of ultra-thin plates; the preset conversion rate deviation benchmark value is 0.05, representing the difference between the preset target conversion rate of 0.92 and the acceptable lower limit of the plate image structure strength of 0.87. The two reference values ​​represent the permissible deviations of the process in their respective directions. After this conversion, the two deviations are freed from the dimensional difference between Celsius and dimensionless ratios. Their numerical values ​​reflect the proportion of each relative to the permissible process range in its direction, thus making them comparable. If the absolute values ​​of the two deviations are directly added together without this conversion, the magnitude of the temperature prediction deviation will be much higher than that of the conversion rate prediction deviation. The resulting weight will be constantly biased towards the speed channel and will not change with the operating conditions.

[0041] Add the normalized value of temperature deviation to the normalized value of conversion rate deviation to obtain the normalized total. Divide the normalized value of temperature deviation by the normalized total to obtain the weight of the speed channel. Divide the normalized value of conversion rate deviation by the normalized total to obtain the weight of the light intensity channel. The sum of the two weights obtained is always 1. Both channels output adjustment amounts in the same control cycle. The weights determine the proportion of each output amplitude rather than the choice between channels. When the normalized temperature deviation is relatively larger, the weight of the rotation speed channel is correspondingly greater. This is because the response speed of the conveyor roller actuator is faster than that of the light source actuator. Increasing the conveyor speed shortens the residence time of the exposure zone in the irradiation zone, and the total amount of accumulated light energy and reaction heat per unit area decreases accordingly. The response time scale of the plate surface temperature along this path is the residence time itself. After the light source irradiation intensity is reduced, the heat accumulated by the lamp source itself continues to radiate to the plate surface. The plate surface temperature cannot decrease synchronously with the light intensity command. Therefore, the temperature, which has an insurmountable upper limit and changes rapidly during the self-accelerating heat release stage, needs to be mainly compensated by the faster-responding rotation speed channel. When the normalized conversion rate deviation is relatively larger, the weight of the light intensity channel is greater. This is because the resin crosslinking conversion rate is an integral quantity that accumulates over time and cannot be reversed. It is only necessary to reach the preset target conversion rate when the exposure zone moves to the downstream detection station. A certain response lag is allowed for the compensation action. Moreover, the light source irradiation intensity is not constrained by the consistency of the rotation speed of adjacent zones and can continuously act on the deviation of the accumulated reaction degree with a large magnitude.

[0042] In the preset speed proportional-integral-derivative parameters, the proportional coefficient is set to 0.02, the integral coefficient to 0.005, and the derivative coefficient to 0.01. These three coefficients correspond to multiplying the instantaneous value of the temperature prediction deviation, the cumulative value of the temperature prediction deviation over a previous period, and the rate of change of the temperature prediction deviation by their respective coefficients, and then summing them to obtain the unweighted speed adjustment base value. The cumulative value of the temperature prediction deviation over a previous period is the sum of all historical temperature prediction deviations experienced by the plate corresponding to the exposure zone from the time the plate enters the exposure area until the current sampling time. When the exposure zone completes the exposure of the current plate and a new plate enters the area corresponding to the exposure zone, this cumulative value... The cumulative value of the conversion rate prediction deviation is obtained and reset in the same way after being cleared and restarted. The proportional coefficient in the preset light intensity proportional integral differential parameter is set to 1.2, the integral coefficient is set to 0.3, and the differential coefficient is set to 0.15. The conversion rate prediction deviation is applied in the same way to obtain the unweighted light intensity adjustment base quantity. The values ​​of these six coefficients are based on the response characteristics of the conveyor roller actuator and the light source actuator. The conveyor roller speed adjustment can follow the control signal change in a short time, so the proportional coefficient is set to be relatively small to avoid fluctuations caused by too fast response. The adjustment of the light source irradiance intensity has the thermal inertia of the lamp source itself, so the integral coefficient is set to be relatively large to compensate for the cumulative deviation caused by slow response. The weighted operation process multiplies the speed channel weight by the speed adjustment base value to obtain the speed adjustment value, and multiplies the light intensity channel weight by the light intensity adjustment base value to obtain the light intensity adjustment value. The channel weights are applied to the base values ​​obtained after the proportional-integral-differential operation is completed, rather than to the deviation value itself. This processing ensures that the proportional-integral-differential operation is always calculated based on complete deviation information. The weights only determine the proportion of the final adopted result, avoiding the weakening of the sensitivity of the derivative term to the deviation change trend due to the weights being applied to the deviation value in advance.

[0043] In one specific embodiment, step S4 includes: The difference between the speeds of two adjacent exposure zones is calculated by performing a difference operation. The speed difference between adjacent zones is compared based on a preset speed difference threshold to obtain the over-limit difference component that exceeds the preset speed difference threshold. The over-limit difference component is gradient-allocated according to the preset gradient allocation ratio to obtain the allocated speed adjustment amount. The allocated speed adjustment amount and light intensity adjustment amount are integrated to obtain the execution adjustment amount, and the conveyor roller actuator and the light source actuator are controlled to operate according to the execution adjustment amount.

[0044] Specifically, the rotation speed adjustment amount between two adjacent exposure zones refers to the rotation speed adjustment amount calculated by the previous steps for each of the two adjacent exposure zones in numerical order. The rotation speed adjustment amount corresponding to the zone with the earlier number is subtracted from the rotation speed adjustment amount corresponding to the zone with the later number to obtain the rotation speed difference between the adjacent zones. This difference can be positive or negative. The absolute value is then compared with a preset rotation speed difference threshold, which is 2 millimeters per second. This value is based on the fact that when the speed difference between adjacent zones exceeds this value during the transport of the ultra-thin resin plate, the plate will undergo observable tensile deformation locally due to the inconsistent traction force on both sides. When the speed difference is lower than this value, the plate can naturally adapt to the slight difference in transport speed without deformation due to its own flexibility.

[0045] When the absolute value of the speed difference between adjacent partitions does not exceed the preset speed difference threshold, the speed adjustment amount originally calculated for that partition is not changed and is directly used as the basis for subsequent execution. When the absolute value of the speed difference between adjacent partitions exceeds the preset speed difference threshold, the part of the value that exceeds the threshold is the over-limit difference component. The over-limit difference component is equal to the absolute value of the speed difference between adjacent partitions minus the preset speed difference threshold.

[0046] The gradient allocation process involves dividing the excess speed difference component into two parts: 60% to the next adjacent partition and 40% to the next level partition. The resulting values ​​are then added to the original speed adjustment of the corresponding partition to form a new speed adjustment. The speed difference between adjacent partitions is then recalculated. This comparison and allocation process is repeated until the absolute value of the speed difference between all adjacent partitions falls within the preset speed difference threshold. The final value obtained by each partition is the allocated speed adjustment. This step-by-step allocation method, rather than directly discarding or forcibly truncating the excess speed, ensures that the speed distribution along the width of the conveyor belt gradually transitions rather than abruptly changes. This avoids local stress concentration at the boundary due to simple limiting of speed differences between adjacent partitions.

[0047] The integrated processing involves taking the allocated speed adjustment amount for each exposure zone and the light intensity adjustment amount calculated in the previous step as two independent values ​​to form the execution adjustment amount. The speed part of the execution adjustment amount drives the corresponding conveyor roller actuator of the zone to adjust the actual operating speed, and the light intensity part drives the corresponding light source actuator of the zone to adjust the actual output intensity. The two parts of the value act on their respective actuators without mutual conversion, thereby completing the actual update of the current process conditions of the exposure zone.

[0048] The above describes the intelligent control method for the photopolymerization process of flexible resin plates in the embodiments of this application. The following describes the intelligent control device for the photopolymerization process of flexible resin plates in the embodiments of this application. One embodiment of the intelligent control device for the photopolymerization process of flexible resin plates in the embodiments of this application includes: The acquisition module is used to acquire the feedback amount of light source irradiation intensity, plate surface temperature and resin crosslinking conversion rate corresponding to the exposure zone, and obtain the feedback amount group; The calculation module is used to perform state calculation on the feedback quantity group based on the nominal model of the controlled object preset in the controller to obtain the theoretical conversion rate increment. The difference between the theoretical conversion rate increment and the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group is compared to obtain the model deviation. The model parameters in the nominal model of the controlled object are closed-loop tuned according to the model deviation to obtain the field model parameters corresponding to the exposure zone. The field model parameters are substituted into the nominal model of the controlled object to perform feedforward calculation on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone as it moves to the downstream detection station. The predicted control deviation is obtained based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature. The calculation module is used to calculate the speed adjustment amount and light intensity adjustment amount respectively for the conveyor roller actuator and the light source actuator based on the predicted control deviation. The control module is used to constrain the rotation speed adjustment amount based on the difference between the rotation speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount, and control the conveyor roller actuator and the light source actuator to operate according to the execution adjustment amount.

[0049] This invention also provides an intelligent control device for the photopolymerization process of flexible resin plates, which can be a server. The intelligent control device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor, designed as a computer, provides computing and control capabilities. The memory of the intelligent control device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the intelligent control device stores the data corresponding to this embodiment. The network interface of the intelligent control device is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0050] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the intelligent control method for the photopolymerization reaction process of the flexible resin plate.

[0051] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0052] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a flexible resin plate photopolymerization process intelligent control device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0053] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent control of the photopolymerization process of flexible resin plates, characterized in that, The method includes: Step S1: Obtain the feedback amount of light source irradiance intensity, plate surface temperature and resin crosslinking conversion rate corresponding to the exposure zone to obtain the feedback amount group; Step S2: Based on the nominal model of the controlled object preset in the controller, perform state calculation on the feedback quantity group to obtain the theoretical conversion rate increment. Compare the difference between the theoretical conversion rate increment and the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group to obtain the model deviation. Based on the model deviation, perform closed-loop tuning on the model parameters in the nominal model of the controlled object to obtain the field model parameters corresponding to the exposure zone. Substitute the field model parameters into the nominal model of the controlled object and perform feedforward calculation on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone as it moves to the downstream detection station. Based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature, obtain the predicted control deviation. Step S3: Calculate the speed adjustment amount and light intensity adjustment amount for the conveyor roller actuator and the light source actuator respectively based on the predicted control deviation; Step S4: Constrain the rotation speed adjustment amount according to the difference between the rotation speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount, and control the conveyor roller actuator and the light source actuator to operate according to the execution adjustment amount.

2. The intelligent control method for the photopolymerization process of flexible resin plates according to claim 1, characterized in that, Step S1 includes: Based on the irradiance sensor, surface temperature sensor and curing degree sensor set in the upstream detection station corresponding to the exposure zone, the exposure zone is synchronously detected to obtain the real-time light source irradiance value, real-time plate surface temperature value and real-time resin crosslinking conversion rate value corresponding to multiple consecutive sampling times. Based on the real-time light source irradiance value, the real-time plate surface temperature value, and the real-time resin crosslinking conversion rate value corresponding to two adjacent sampling times, differential calculation processing is performed on the exposure zone to obtain the light source irradiance change rate, the plate surface temperature change rate, and the resin crosslinking conversion rate change rate. Based on the real-time light source irradiance value, the real-time plate surface temperature value, the real-time resin crosslinking conversion rate value, the light source irradiance change rate, the plate surface temperature change rate, and the resin crosslinking conversion rate change rate, the exposure zones are classified and integrated to obtain the light source irradiance feedback, plate surface temperature feedback, and resin crosslinking conversion rate feedback corresponding to the exposure zones. The feedback quantities of the light source irradiance intensity, the plate surface temperature, and the resin crosslinking conversion rate corresponding to the exposure zone are input into the zone data association module for marking and encapsulation processing to obtain the feedback quantity group.

3. The intelligent control method for the photopolymerization reaction process of flexible resin plates according to claim 2, characterized in that, In step S2, based on the nominal model of the controlled object preset in the controller, the state of the feedback quantity group is calculated to obtain the theoretical conversion rate increment. The theoretical conversion rate increment is then compared with the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group to obtain the model deviation, including: Based on the light source irradiance feedback, plate surface temperature feedback, and preset initial values ​​of field model parameters in the feedback set, the nominal model of the controlled object is substituted and constructed to obtain the reaction rate differential equation. Based on the preset calculation time, the reaction rate differential equation is iteratively integrated to obtain the theoretical conversion rate increment corresponding to the exposure zone; Based on the real-time resin crosslinking conversion rate value at the start time of the preset estimated duration corresponding to the exposure zone, the real-time resin crosslinking conversion rate value at the end time of the preset estimated duration is processed by difference calculation to obtain the measured increment. The measured increment is divided by the theoretical conversion rate increment to obtain the model deviation.

4. The intelligent control method for the photopolymerization reaction process of flexible resin plates according to claim 3, characterized in that, In step S2, the model parameters in the nominal model of the controlled object are closed-loop tuned according to the model deviation to obtain the field model parameters corresponding to the exposure zone, including: The preset tuning learning rate is weighted based on the model deviation to obtain the parameter correction amount; The initial values ​​of the field model parameters are superimposed based on the parameter correction amount to obtain candidate model parameters; The candidate model parameters are subjected to boundary comparison processing based on a preset parameter threshold range to obtain the boundary-constrained model parameters. The model parameters in the nominal model of the controlled object are updated based on the model parameters after the boundary constraints to obtain the field model parameters corresponding to the exposure zone.

5. The intelligent control method for the photopolymerization reaction process of flexible resin plates according to claim 4, characterized in that, In step S2, the field model parameters are substituted into the nominal model of the controlled object, and feedforward calculation is performed on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone moving to the downstream detection station. Based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature, the predicted control deviation is obtained, including: Substitute the field model parameters into the nominal model of the controlled object for assignment processing to obtain the field reaction rate differential equation; Based on the conveyor belt speed corresponding to the exposure zone, the travel time required for the exposure zone to travel from the upstream detection station to the downstream detection station is calculated to obtain the predicted travel time. Based on the predicted travel time and the plate surface temperature feedback in the feedback group, the on-site reaction rate differential equation is iteratively integrated according to the preset calculation time to obtain the predicted conversion rate and predicted temperature corresponding to the exposure zone traveling to the downstream detection station. The difference between the predicted conversion rate and the preset target conversion rate is calculated, and the difference between the predicted temperature and the preset maximum temperature is calculated to obtain the conversion rate prediction deviation and the temperature prediction deviation. The conversion rate prediction deviation and the temperature prediction deviation are then categorized and integrated to obtain the prediction control deviation.

6. The intelligent control method for the photopolymerization reaction process of flexible resin plates according to claim 5, characterized in that, Step S3 includes: Based on the conversion rate prediction deviation and the temperature prediction deviation corresponding to two adjacent sampling times, differential calculation processing is performed on the exposure zone to obtain the conversion rate prediction deviation change rate and the temperature prediction deviation change rate. The conversion rate prediction deviation and the temperature prediction deviation are normalized to obtain the rotation speed channel weight and the light intensity channel weight. Based on the temperature prediction deviation, the rate of change of the temperature prediction deviation, and the speed channel weight, the preset speed proportional-integral-derivative parameters are weighted and processed to obtain the speed adjustment amount. Based on the conversion rate prediction deviation, the conversion rate prediction deviation change rate, and the light intensity channel weight, a weighted calculation is performed on the preset light intensity proportional integral differential parameters to obtain the light intensity adjustment amount.

7. The intelligent control method for the photopolymerization process of flexible resin plates according to claim 6, characterized in that, Step S4 includes: Based on the rotation speed adjustment amount of two adjacent exposure zones, a difference calculation is performed to obtain the rotation speed difference between adjacent zones; The speed difference between adjacent partitions is compared according to a preset speed difference threshold to obtain the over-limit difference component that exceeds the preset speed difference threshold. The over-limit difference component is gradient-allocated according to a preset gradient allocation ratio to obtain the allocated speed adjustment amount. The allocated rotational speed adjustment amount and the light intensity adjustment amount are integrated to obtain the execution adjustment amount, and the conveyor roller actuator and the light source actuator are controlled to operate according to the execution adjustment amount.

8. An intelligent control device for the photopolymerization process of flexible resin plates, characterized in that, For implementing the intelligent control method for the photopolymerization process of flexible resin plates as described in any one of claims 1-7, the intelligent control device for the photopolymerization process of flexible resin plates comprises: The acquisition module is used to acquire the feedback amount of light source irradiation intensity, plate surface temperature and resin crosslinking conversion rate corresponding to the exposure zone, and obtain the feedback amount group; The calculation module is used to perform state calculation on the feedback quantity group based on the nominal model of the controlled object preset in the controller to obtain the theoretical conversion rate increment. The difference between the theoretical conversion rate increment and the measured increment of the resin crosslinking conversion rate feedback quantity in the feedback quantity group is compared to obtain the model deviation. The model parameters in the nominal model of the controlled object are closed-loop tuned according to the model deviation to obtain the field model parameters corresponding to the exposure zone. The field model parameters are substituted into the nominal model of the controlled object to perform feedforward calculation on the feedback quantity group to obtain the predicted conversion rate and predicted temperature of the exposure zone as it moves to the downstream detection station. The predicted control deviation is obtained based on the difference between the predicted conversion rate and the preset target conversion rate, and the difference between the predicted temperature and the preset maximum temperature. The calculation module is used to calculate the speed adjustment amount and light intensity adjustment amount respectively for the conveyor roller actuator and the light source actuator based on the predicted control deviation. The control module is used to constrain the rotation speed adjustment amount based on the difference between the rotation speed adjustment amounts of adjacent exposure zones to obtain the execution adjustment amount, and control the conveyor roller actuator and the light source actuator to operate according to the execution adjustment amount.

9. An intelligent control device for the photopolymerization process of flexible resin plates, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the intelligent control method for the photopolymerization reaction process of flexible resin plates as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to execute the intelligent control method for the photopolymerization reaction process of flexible resin plates as described in any one of claims 1 to 7.

Images