Intelligent control method and system for uniformity of layer thickness of super-wide shed film co-extrusion

Abstract

The application provides an intelligent control method and system for layer thickness uniformity of super-wide shed film co-extrusion, relates to the technical field of multi-layer co-extrusion blow molding control, and comprises the following steps: after a bubble is formed into a multi-layer co-extrusion structure with three layers or more, and before cooling and setting, layer thickness measurement signals of each functional layer at multiple preset detection positions in the transverse direction of the bubble are collected in real time; the layer thickness values of each functional layer at each detection position are analyzed according to the layer thickness measurement signals, and the real-time layer thickness distribution profile of each functional layer in the transverse direction of the bubble is obtained. The application can effectively detect and control the uneven distribution of each functional layer in the transverse direction caused by the viscous encapsulation effect.

Description

Technical Field

[0001] This invention relates to the field of multi-layer co-extrusion blow molding control technology, and in particular to an intelligent control method and system for the uniformity of co-extruded layer thickness of ultra-wide greenhouse film. Background Technology

[0002] Ultra-wide multilayer co-extruded functional films are widely used for overall covering of large multi-span greenhouses. These films adopt a multilayer co-extrusion structure with three or more layers. Each functional layer is divided according to the functional properties of the melt supplied by the corresponding extruder, covering an outer functional layer that provides weather protection, a reinforcing functional layer that provides structural strength, and an inner functional layer that provides surface functional properties.

[0003] In multilayer co-extrusion blow molding, existing technologies mostly face the following drawbacks: Due to the viscosity differences in the rheological behavior of the melts of each functional layer, low-viscosity melts tend to flow and accumulate towards the lateral edge region of the bubble when flowing through the die runner, while high-viscosity melts are relatively concentrated in the central region of the bubble, resulting in a viscous encapsulation effect. This viscous encapsulation effect causes the layer thickness distribution of each functional layer along the lateral direction of the bubble to deviate from the design value, with the functional layers being thicker and the reinforcing functional layers thinner in the edge region, and the functional layers being thinner and the reinforcing functional layers thicker in the central region. These phenomena are particularly prominent in ultra-wide bubble structures.

[0004] This defect can cause spatial differences in functional performance. For example, the anti-drip layer in the edge area may precipitate and be lost early due to excessively high anti-drip agent concentration, thus losing its anti-drip function. In the central area, the infrared blocking rate is low due to a thinner insulation layer, causing nighttime temperature differences between the center and the sides of the greenhouse, affecting the uniformity of crop growth. Currently, the industry has mature solutions that control the total thickness deviation within a small range through technologies such as die gap adjustment and automatic air rings. However, for the systemic problem of uneven lateral distribution of functional layers caused by the adhesive encapsulation effect, there is currently a lack of effective online detection and real-time control methods. Summary of the Invention

[0005] This invention provides an intelligent control method and system for the uniformity of co-extruded layer thickness of ultra-wide greenhouse film. It can effectively detect and control the uneven lateral distribution of each functional layer caused by the adhesive encapsulation effect online, and further reduce the deviation of the layer thickness distribution of each functional layer from the design value along the lateral direction of the film bubble.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: Firstly, a method for intelligent control of the uniformity of co-extruded layer thickness in ultra-wide greenhouse films, the method comprising: Step 1: After the membrane bubble forms a multilayer co-extruded structure of three or more layers and before cooling and shaping, the layer thickness measurement signals of each functional layer at multiple preset detection positions along the transverse direction of the membrane bubble are collected in real time. Step 2: Based on the layer thickness measurement signal, analyze the layer thickness value of each functional layer at each detection position to obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble; Step 3: Compare the difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. Step 4: Analyze the variation trend and amplitude distribution of the layer thickness deviation along the transverse direction of the bubble, and extract the encapsulation state feature quantity characterizing the viscous encapsulation effect. The encapsulation state feature quantity includes the edge cohesion degree determined based on the layer thickness deviation envelope area of ​​each functional layer in the edge region of the bubble and the center cohesion degree determined based on the layer thickness deviation envelope depth of each functional layer in the center region of the bubble. Step 5: Based on the encapsulation state characteristic quantity and combined with the current viscosity parameters of the melt of each functional layer, determine the lip gap adjustment direction and lip gap adjustment amount of each layer flow channel of the die head at each transverse partition adjustment section; Step 6: Based on the determined lip gap adjustment direction and lip gap adjustment amount, adjust the outflow resistance of the melt in each functional layer of the corresponding transverse partition, and change the melt flow distribution of each functional layer in the transverse direction of the membrane bubble. Step 7: After completing this round of adjustment, reacquire the layer thickness measurement signal and use the updated layer thickness deviation as the input for the next adjustment cycle. Continue iterating until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range.

[0007] Secondly, the intelligent control system for ensuring uniformity of co-extruded layer thickness of ultra-wide greenhouse film includes: The layer thickness signal acquisition module is used to acquire layer thickness measurement signals at multiple preset detection positions along the transverse direction of the membrane bubble in real time after the membrane bubble has formed a multi-layer co-extrusion structure of three or more layers and before cooling and shaping. The profile analysis module is used to analyze the layer thickness value of each functional layer at each detection position based on the layer thickness measurement signal, and obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble. The deviation calculation module is used to compare the difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. The encapsulation feature extraction module is used to analyze the variation trend and amplitude distribution of the layer thickness deviation along the transverse direction of the bubble, and extract the encapsulation state feature quantity characterizing the viscous encapsulation effect. The encapsulation state feature quantity includes the edge cohesion degree determined based on the layer thickness deviation envelope area of ​​each functional layer in the edge region of the bubble and the center cohesion degree determined based on the layer thickness deviation envelope depth of each functional layer in the center region of the bubble. The lip gap adjustment calculation module is used to determine the lip gap adjustment direction and lip gap adjustment amount of each layer of the die head flow channel at each transverse partition adjustment section based on the encapsulation state characteristic quantity and the current viscosity parameters of the melt of each functional layer. The melt flow rate regulation execution module is used to adjust the outflow resistance of the melt in each functional layer of the corresponding transverse partition according to the determined lip gap adjustment direction and lip gap adjustment amount, thereby changing the melt flow rate distribution of each functional layer in the transverse direction of the membrane bubble. The closed-loop iterative control module is used to trigger a new round of layer thickness measurement signal acquisition after the current adjustment is completed. The updated layer thickness deviation is used as the input for the next adjustment cycle, and the iteration continues until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range.

[0008] The above-described solution of the present invention has at least the following beneficial effects: By employing real-time acquisition of layer thickness signals and profile analysis, the layer thickness distribution of each functional layer along the transverse direction of the membrane bubble is monitored online at each detection location and the deviation is quantified. The viscous encapsulation effect is incorporated into the closed-loop control system, enabling the control system to perceive the transverse distribution status of each functional layer in real time.

[0009] By decomposing the thickness deviation profile into symmetrical deviation components and antisymmetrical deviation components with the longitudinal centerline of the membrane bubble as the axis of symmetry, filtering out random disturbances, and fitting the bimodal envelope shape of the symmetrical deviation components, a symmetrical envelope model is constructed. Then, the edge eccentricity and center eccentricity are extracted as encapsulation state characteristics. In this way, the viscous encapsulation effect is transformed from a rheological phenomenon that cannot be directly measured into a calculable geometric characteristic parameter, and the control system can accurately perceive the real-time state and spatial distribution characteristics of the encapsulation effect.

[0010] Based on the encapsulation state characteristics and the current viscosity parameters of the melt of each functional layer, differentiated lip gap adjustment is implemented for each transverse partition. By changing the outflow resistance distribution of each layer of melt in the corresponding partition, the segregation trend is actively compensated, thereby adjusting the layer thickness distribution towards uniformity while retaining the original material formulation and functional characteristics of each functional layer.

[0011] By re-acquiring the layer thickness measurement signal after each adjustment cycle and using the updated layer thickness deviation as the input for the next adjustment cycle, the layer thickness deviation of each functional layer is gradually converged to the preset allowable deviation range through closed-loop iteration. This allows it to adapt to different process conditions and changes in membrane bubble width without relying on a preset static compensation model.

[0012] By making the thickness distribution of each functional layer along the transverse direction of the membrane bubble approach the design value, the drip retention period, heat preservation performance and weather resistance of each part of the membrane surface tend to be consistent, thereby eliminating the temperature deviation caused by the spatial difference in the functional performance of the membrane material in the greenhouse. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating the intelligent control method for the uniformity of co-extruded layer thickness of ultra-wide greenhouse film provided in an embodiment of the present invention.

[0014] Figure 2 This is a schematic diagram of an intelligent control system for ensuring uniformity of co-extruded layer thickness of ultra-wide greenhouse film provided in an embodiment of the present invention.

[0015] Figure 3 This is a schematic diagram of the process for determining the lip gap adjustment direction and lip gap adjustment amount of each layer of the die head at each transverse partition adjustment section based on the packaging state characteristic quantity and the current viscosity parameter of the melt of each functional layer. Detailed Implementation

[0016] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0017] like Figure 1 As shown, embodiments of the present invention propose an intelligent control method for the uniformity of co-extruded layer thickness of ultra-wide greenhouse films, the method comprising the following steps: Step 1: After the membrane bubble forms a multilayer co-extruded structure of three or more layers and before cooling and shaping, the layer thickness measurement signals of each functional layer at multiple preset detection positions along the transverse direction of the membrane bubble are collected in real time. Step 2: Based on the layer thickness measurement signal, analyze the layer thickness value of each functional layer at each detection position to obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble; Step 3: Compare the difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. Step 4: Analyze the variation trend and amplitude distribution of the layer thickness deviation along the transverse direction of the bubble, and extract the encapsulation state feature quantity characterizing the viscous encapsulation effect. The encapsulation state feature quantity includes the edge cohesion degree determined based on the layer thickness deviation envelope area of ​​each functional layer in the edge region of the bubble and the center cohesion degree determined based on the layer thickness deviation envelope depth of each functional layer in the center region of the bubble. Step 5: Based on the encapsulation state characteristic quantity and combined with the current viscosity parameters of the melt of each functional layer, determine the lip gap adjustment direction and lip gap adjustment amount of each layer flow channel of the die head at each transverse partition adjustment section; Step 6: Based on the determined lip gap adjustment direction and lip gap adjustment amount, adjust the outflow resistance of the melt in each functional layer of the corresponding transverse partition, and change the melt flow distribution of each functional layer in the transverse direction of the membrane bubble. Step 7: After completing this round of adjustment, reacquire the layer thickness measurement signal and use the updated layer thickness deviation as the input for the next adjustment cycle. Continue iterating until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range.

[0018] In this embodiment of the invention, by acquiring the layer thickness measurement signals of each functional layer in real time and analyzing them into a distribution profile, the deviation is obtained by comparing it position by position with the target design profile, so that the uneven distribution of layer thickness caused by the viscous encapsulation effect can be quantitatively characterized. On this basis, the changing trend and amplitude distribution of the deviation are analyzed, and the edge eccentricity and center eccentricity are extracted as encapsulation state characteristic quantities, transforming the rheological phenomenon that cannot be directly measured into geometric characteristic parameters that the control system can perceive, providing a precise decision basis for subsequent differentiated adjustment.

[0019] Based on the encapsulation state characteristics and the viscosity parameters of the melt of each functional layer, the lip gap adjustment scheme of each transverse partition is determined. By changing the outflow resistance distribution, the segregation trend is actively compensated, and the layer thickness distribution is adjusted towards homogenization while retaining the original material formulation. After each round of adjustment, the signal is re-acquired and the updated deviation is used as the input for the next cycle to form a closed loop iteration, which gradually converges to the preset deviation allowable range. It can adapt to different process conditions and changes in bubble width without relying on a static compensation model.

[0020] This embodiment uses a multilayer co-extruded bubble structure with three or more layers as the application object. The total number of layers in the bubble is denoted as N, where N is an integer greater than or equal to 3. Each functional layer is divided according to the functional properties of the melt supplied by the corresponding extruder, including an outer functional layer (denoted as the first layer) that provides weather protection, an reinforcing functional layer (denoted as the middle layer) that provides structural strength, and an inner functional layer (denoted as the Nth layer) that provides surface functional properties, including dripping and defogging. In actual products, N is typically 5 layers, in which case, from the outside to the inside, they are the outer functional layer (the first layer), the thermal insulation functional layer (the second layer), the reinforcing functional layer (the third layer), the adhesive layer (the fourth layer), and the inner functional layer (the fifth layer). The outer and inner functional layers use a low-viscosity melt formulation, while the reinforcing functional layer uses a high-viscosity melt formulation. The difference in melt viscosity among the functional layers is the physical source of the viscous encapsulation effect.

[0021] In a preferred embodiment of the present invention, step 1 above may include: The overall execution period of this step is after the membrane bubble is extruded from the die and forms a stable multilayer structure, and before the membrane bubble enters the cooling air ring and traction clamping roller. During this period, the melt of each functional layer is still in a flowable high-temperature state, and the layer thickness information truly reflects the effect of the viscous encapsulation effect in the die flow channel. It has not yet been disturbed by subsequent cooling shrinkage and traction stretching, making it the best window for online layer thickness detection. K preset detection positions are set along the transverse direction of the membrane bubble. The typical value of K is 20 to 60. The detection positions are evenly distributed from one edge of the membrane bubble to the opposite edge. The spacing between adjacent detection positions is determined according to the actual width W of the membrane bubble and the accuracy requirements.

[0022] With the longitudinal centerline of the membrane vesicle as the origin of the lateral coordinate system ( The horizontal coordinates of the two sides are respectively and The horizontal coordinates of each detection location are: ， The total number of interlayer interfaces in the membrane vesicle is N-1, and the interlayer interfaces are numbered sequentially from the outside to the inside as follows: ,interface i Located in the i Layer and first i Between +1 floors.

[0023] Step 1.1: At each preset detection position in the transverse direction of the membrane bubble, a detection signal is emitted towards the surface of the membrane bubble, so that the detection signal penetrates each functional layer and generates reflected echoes at the interlayer interfaces between adjacent functional layers; wherein, each functional layer is divided according to the functional attributes of the melt supplied by the corresponding extruder, including an outer functional layer providing weather protection, a reinforcing functional layer providing structural strength, and an inner functional layer providing surface functional properties; by receiving the reflected echoes, a multi-interface echo sequence carrying interlayer interface position information at the corresponding detection position is obtained, specifically including: at each preset detection position in the transverse direction of the membrane bubble... At this location, a detection probe installed on the outside of the membrane bubble emits a detection signal perpendicularly to the surface of the membrane bubble. In this embodiment, the detection signal is an ultrasonic pulse signal, and the frequency of the ultrasonic wave is selected in the range of 2MHz to 10MHz. The specific frequency is selected according to the acoustic characteristics of each functional layer material and the minimum layer thickness requirement, so as to ensure that the axial resolution can distinguish the reflected echoes of the interface between adjacent layers.

[0024] When ultrasound passes through different media, it produces significant echo reflections at interfaces where the media density changes abruptly. The detection probe includes an ultrasonic transmitting unit and an ultrasonic receiving unit. The transmitting unit emits short ultrasonic pulses at a preset frequency onto the surface of the membrane bubble. The pulses pass through the functional layers of the membrane bubble and are reflected at the interlayer interfaces between every two adjacent functional layers. iAt the detection location, due to the difference in acoustic impedance between adjacent material layers, some ultrasonic energy is reflected back to the detection probe, forming reflected echoes. The interfaces between each layer generate corresponding reflected echoes sequentially from the outer surface to the inner surface of the membrane bubble. These echoes are arranged according to the order in which they arrive at the detection probe, collectively forming the reflected echoes at the detection location. The multi-interface echo sequence; each echo pulse in the multi-interface echo sequence corresponds to a specific interlayer interface in the multilayer structure of the membrane bubble, and its arrival time encodes the spatial depth information of the corresponding interface relative to the surface of the membrane bubble.

[0025] Before transmitting the detection signal, an ultrasonic coupling medium is coated between the detection probe and the outer surface of the membrane bubble. In this embodiment, a water-based or oil-based coupling agent is used to establish good acoustic coupling between the detection probe and the membrane bubble surface, eliminating the attenuation of ultrasonic wave transmission due to air gaps, thereby improving the transmission efficiency of the detection signal at the interfaces between layers and the signal-to-noise ratio of the reflected echo; by receiving all reflected echoes, the corresponding detection position is obtained. The multi-interface echo sequence is as follows: here, each functional layer is divided according to the functional properties of the melt supplied by the corresponding extruder, including the outer functional layer (layer 1) that provides weather protection, the reinforcing functional layer that provides structural strength, and the inner functional layer (layer N) that provides surface functional properties. Each functional layer is supplied with melt by an independent extruder, and the interlayer interface is the physical boundary formed after the different melts supplied by adjacent extruders merge in the die.

[0026] Step 1.2: Based on the multi-interface echo sequence, identify the echo characteristic peaks corresponding to each inter-layer interface, extract the time position parameters of each echo characteristic peak, and obtain the interface position characteristic parameters of each inter-layer interface. Specifically, this includes: after obtaining the multi-interface echo sequence, performing signal preprocessing on the echo sequence, including removing baseline drift and suppressing high-frequency noise; the removal of baseline drift is achieved by subtracting its moving average value from the original echo sequence, with the length of the sliding window being twice the duration of the ultrasonic pulse; the suppression of high-frequency noise is achieved by bandpass filtering, with the lower cutoff frequency of the filter being 50% of the ultrasonic transmission center frequency and the upper cutoff frequency being 200% of the center frequency.

[0027] After signal preprocessing, the echo characteristic peaks corresponding to each interlayer interface are identified based on the time-domain waveform of the multi-interface echo sequence. The echo characteristic peak refers to the local amplitude maxima formed by reflection from the interlayer interface in the echo sequence. The peak position indicates the time when the reflected echo arrives at the detection probe. The identification of the echo characteristic peak is achieved by setting two constraints: amplitude threshold and minimum peak spacing. The amplitude threshold is used to eliminate spurious peaks caused by noise. In this embodiment, the amplitude threshold is taken as 10% to 20% of the maximum echo amplitude in the echo sequence. The minimum peak spacing is used to distinguish adjacent interlayer interface echoes and prevent the side lobes of the same echo from being mistakenly identified as independent interlayer interfaces. The value of the minimum peak spacing is determined based on the full width at half maximum (FWHM) of the ultrasonic pulse, with a typical value of 0.5 μs to 1.0 μs.

[0028] For each identified echo characteristic peak, its corresponding peak time is extracted as the time position parameter of that echo characteristic peak; the time position parameter is recorded in microseconds (μs) with the start time of the ultrasonic pulse emission as zero, denoted as . , indicating the detection location First The temporal position parameters corresponding to each interlayer interface; the temporal position parameters of all identified echo characteristic peaks are arranged in order of depth from the outer surface to the inner surface of the corresponding interlayer interface to obtain the interface position characteristic parameters of each interlayer interface. The interface position characteristic parameters are essentially a time value sequence, and each element in the sequence... The round-trip time of an ultrasonic wave at a layer interface.

[0029] The time position parameter The scaling factor between the ultrasonic wave and the corresponding interface depth depends on the propagation speed of the ultrasonic wave in each functional layer material; the ultrasonic wave in the first... Propagation speed in functional layer materials (Unit: m / s), obtained by calibrating the sound velocity of a standard sample of the functional layer material beforehand. The calibration process is as follows: Prepare the first... The standard sample of the functional layer material, the thickness of the standard sample is denoted as . (Unit: mm) The average value was determined by taking three precise measurements with a micrometer; the ultrasonic probe was coupled to the surface of the standard sample block, an ultrasonic pulse was emitted and the reflected echo from the bottom surface was received, and the round-trip propagation time of the ultrasonic wave in the standard sample block was recorded as . (Unit: μs), the average value is taken after five repeated measurements.

[0030] Ultrasound in the Propagation speed in functional layer materials The coefficient 2000 in this formula comes from unit conversion. The unit is mm, which needs to be converted to m; The unit is μs, which needs to be converted to s; the round-trip path is divided by 2; the overall conversion factor is 2000 (i.e., ...). ), sound velocity calibration values ​​of each functional layer material The material parameter table stored in the system is automatically retrieved based on the material type of the functional layer at each detection location during subsequent layer thickness calculations; for example, when the standard sample thickness of the outer functional layer... Average round-trip time of five measurements hour, .

[0031] Step 1.3: Based on the interface position feature parameters of adjacent interlayer interfaces, calculate the layer thickness value of each functional layer at the corresponding detection position to obtain the independent layer thickness measurement value of each functional layer at the corresponding detection position. Specifically, this includes obtaining the interface position feature parameters of each interlayer interface. and the corresponding ultrasonic propagation speed Then, based on the interface position feature parameters of two adjacent interlayer interfaces, the corresponding detection positions are calculated. The thickness values ​​of each functional layer; for the outer functional layer (layer 1) located between the outer surface of the membrane bubble and the interface between the first layer, its thickness value is... The time position parameter corresponding to the first interlayer interface is multiplied by the propagation speed of the ultrasonic wave in the outer functional layer material, and then divided by 2. This is because the time position parameter records the round-trip time, and the one-way time is half of the round-trip time. In the formula The unit is μm. The unit is m / s. The unit is μs.

[0032] For the interface between two adjacent layers and Between Intermediate functional layer (of which) , ), its layer thickness value The difference in time position parameters between the interface between the next and previous layers is multiplied by the ultrasonic wave at the [number]th layer. The propagation velocity in the functional layer material is divided by 2, and the calculation formula is: ; The formula means: multiply the time difference between two adjacent interfaces by the speed of sound and divide by 2 to obtain the physical distance between the two interfaces, i.e., the first... The layer thickness value of the functional layer; for the interface between the innermost layers. The inner functional layer between the inner surface of the membrane vesicle (the first layer) (layer), its layer thickness value According to the detection location Total thickness of the membrane bubble The calculation formula is derived by working backwards from the already calculated thickness of each outer functional layer: ; In the formula For detection location The total thickness of the membrane bubble (in μm) at the location can be obtained by converting the time position parameters and average sound velocity of the ultrasonic bottom echo (i.e., the echo reflected from the inner surface of the membrane bubble), or it can be obtained by an independent laser thickness sensor. Indicates from the 1st floor to the 2nd floor. Layer at detection location The sum of the layer thickness values ​​at each location.

[0033] In the above indirect estimation method, the first The independent measurement errors of each outer functional layer will accumulate and be transmitted to the thickness of the inner functional layer. In the calculation, the accuracy of ultrasonic time difference measurement is used. Sound velocity calibration accuracy Estimate the standard deviation of single-layer thickness measurement Approximately ;for The five-layer structure The cumulative standard deviation is calculated according to the error propagation formula. The estimate is approximately To control the cumulative error within an acceptable range, the sound velocity of each functional layer material was measured during the system calibration phase. Perform no fewer than 10 independent and repeated calibrations to reduce the uncertainty of sound velocity calibration; during the operation phase, continuously acquire 3 echo signals at each detection location, and take the median of the time and position parameters to suppress random noise; when detected... The estimated standard deviation exceeds the first When the target layer thickness reaches 5% of the design value, an inner layer accuracy warning signal is issued, prompting the operator to check the coupling status of the ultrasonic probe and the effectiveness of the sound velocity calibration of each outer layer.

[0034] The corresponding detection position is obtained through the above calculation process. Independent layer thickness measurements of each functional layer Independent layer thickness measurements refer to the layer thickness values ​​measured independently for each functional layer at a single preset detection location—distinguished from the total thickness measurement value obtained by combining multiple layers. The independent layer thickness measurements preserve the individual thickness information of each functional layer. The unit is uniformly taken as micrometer (um).

[0035] Step 1.4: After completing the signal acquisition and layer thickness calculation at all preset detection positions in the transverse direction of the membrane bubble, the independent layer thickness measurements of each functional layer at each detection position are integrated into a multi-channel layer thickness measurement signal. Specifically, this includes: at all preset detection positions in the transverse direction of the membrane bubble... After completing signal acquisition and layer thickness calculation at each preset detection location according to steps 1.1 to 1.3 above, the independent layer thickness measurements of each functional layer at each detection location are then processed. Numbered according to the detection location and functional layer number Two-dimensional arrangement and encapsulation integration are performed to obtain a multi-channel layer thickness measurement signal; the multi-channel layer thickness measurement signal is the final output data body of step 1, and its data structure is a... OK A two-dimensional matrix with columns, the matrix's first column... Line 1 The column element is , indicating the detection location First Independent layer thickness measurements for functional layers; during integration, assigning a corresponding timestamp to the layer thickness calculation results at each detection location, recording the data acquisition time at that location; for all transverse layers of the membrane bubble... After data integration is completed at all preset detection locations, the multi-channel layer thickness measurement signal is transmitted to the processing module in step 2 as the raw input data for subsequent profile analysis.

[0036] In a preferred embodiment of the present invention, step 2 above may include: Step 2.1: Extract layer thickness values ​​from the multi-channel layer thickness measurement signal according to functional layer category and lateral detection position of the membrane bubble, to obtain discrete layer thickness data sequences for each functional layer at each preset detection position in the lateral direction of the membrane bubble. Specifically, this includes: extracting layer thickness values ​​from the output multi-channel layer thickness measurement signal according to functional layer category (… ) and transverse detection position of membrane vesicles ( Extract layer thickness values ​​from two dimensions respectively; first, by functional layer category. j The first dimension is split, which separates the layer thickness data of all N functional layers contained in the multi-channel layer thickness measurement signal into N independent data channels. For example, for a five-layer structure with N=5, after splitting, 5 independent data channels are obtained, corresponding to layers 1 to 5 respectively.

[0037] Further, for each functional layer data channel, the lateral detection positions of the membrane bubbles are numbered. Extract the layer thickness values ​​of the functional layer at each detection location in ascending order. The data are arranged into an ordered numerical sequence along the transverse direction of the membrane bubble, resulting in a discrete layer thickness data sequence for each functional layer at each preset detection position along the transverse direction of the membrane bubble. The discrete layer thickness data sequence contains the first... The element is The sequence elements are arranged in the same direction as the transverse direction of the membrane vesicle -- the first element ( The corresponding edge region on one side of the membrane vesicle, the Kth element ( The corresponding element is the edge region on the opposite side of the membrane vesicle, and the middle element corresponds to the central region.

[0038] Step 2.2 involves spatially reconstructing the discrete layer thickness data sequences of each functional layer along the transverse direction of the membrane bubble to obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble. Specifically, this includes: spatially reconstructing the discrete layer thickness data sequences of each functional layer along the transverse direction of the membrane bubble. Since the discrete layer thickness data sequences only provide... Layer thickness values ​​at each preset detection location The transverse direction of the membrane vesicle is actually a continuous physical space, so that it can be positioned at any transverse position. You can get the first place anywhere The thickness value of the functional layer needs to be obtained by interpolating the discrete data sequence to transform it into a continuous function representation along the transverse direction of the membrane bubble.

[0039] In this embodiment, spatial continuity reconstruction is achieved using cubic spline interpolation; cubic spline interpolation uses adjacent detection positions... and Transverse segments of membrane vesicles For each subinterval, construct a cubic polynomial. This ensures that adjacent subintervals have continuous zeroth, first, and second derivatives at the connection point, thereby guaranteeing the smoothness of the reconstructed profile curve; cubic polynomial over subintervals The expression is: ; in For the first Four undetermined coefficients for each subinterval This represents the horizontal coordinate of the left endpoint of the sub-interval. Let be any horizontal coordinate of the layer thickness to be determined. That is, the first Functional layer in horizontal position The layer thickness value at the location, in units of .

[0040] Cubic spline interpolation needs to satisfy the following boundary and continuity conditions: the value of the interpolation function at each subinterval is equal to the measured layer thickness at that endpoint, i.e. , The first derivative of the interpolation function of adjacent subintervals is continuous at the connection point, i.e. The first derivative The first derivative characterizes the layer thickness profile in The rate of change of slope at point , in units of The second derivatives of adjacent subintervals are continuous at the connection point, i.e. The second derivative The second derivative characterizes the layer thickness profile in... The change in curvature at that point, in units of .

[0041] Add natural boundary conditions at the edges of the membrane vesicles. This uniquely determines the coefficients of all subintervals. The cubic spline interpolation described above ensures a smooth profile while avoiding Runge oscillations in the edge regions that occur with higher-order polynomial interpolation. It exhibits good numerical stability and engineering practicality. After spatial continuum reconstruction, the real-time thickness distribution profile of each functional layer along the transverse direction of the membrane bubble is obtained. This profile serves as the input data for step 3.

[0042] In a preferred embodiment of the present invention, step 3 above may include: Step 3.1: Retrieve the target layer thickness design profile for each functional layer from the preset process parameter storage unit. The target layer thickness design profile defines the layer thickness design value of the corresponding functional layer at each detection position in the transverse direction of the membrane bubble. Specifically, the process parameter storage unit is a non-volatile memory in the system used to store preset process data. In this embodiment, a flash memory chip (Flash EEPROM) is used, which stores a complete set of process design parameters corresponding to each membrane bubble product model, including the target layer thickness design profile for each functional layer and the sound velocity calibration value of each layer material. Standard viscosity parameters for each layer of melt, and preset allowable deviation ranges, etc.

[0043] The target layer thickness design profile defines the layer thickness design value for the corresponding functional layer at each detection position in the transverse direction of the membrane bubble; the first layer thickness design profile defines the layer thickness design value for the corresponding functional layer. Functional layer at detection location The target layer thickness design value at the location is denoted as The horizontal coordinate system of the target layer thickness design profile is completely consistent with the real-time layer thickness distribution profile, both based on the lateral position of the membrane bubble. The horizontal axis is defined by the longitudinal centerline of the vesicle as the origin. ), in terms of layer thickness (unit) The target layer thickness design profile is set as the vertical axis, and both layers share the same spatial reference frame. The target layer thickness design profile is pre-defined based on the membrane bubble product specifications and greenhouse application requirements. The target profiles for different functional layers are independently defined and stored. The specific definition method is as follows: For the outer functional layer (layer 1), its target layer thickness design profile is... It is usually set to be uniformly distributed along the transverse direction of the membrane vesicles, that is... ;in The design values ​​are determined based on the effective concentration of UV light stabilizer required for weather protection and the expected service life: that is, in typical applications of ultra-wide greenhouse films, the target design thickness of the outer functional layer. Take as to This thickness range ensures that the effective release concentration of the UV stabilizer throughout the entire lifespan of the film surface is not lower than the critical protection threshold. The outer functional layer is designed with a uniform thickness based on the following engineering considerations: insufficient weather resistance at any location on the film surface will become a shortcoming in the overall service life. Therefore, the outer functional layer must cover the entire film surface uniformly in the lateral direction.

[0044] For the reinforcing layer, its target layer thickness design profile is formulated based on the mechanical requirements of the overall tensile strength and puncture resistance of the membrane bubble; in ultra-wide greenhouse film applications, the reinforcing layer undertakes the main structural load-bearing function of the film surface, and its target design layer thickness... Take as to The specific values ​​are determined based on the span of the greenhouse, the local wind load, and the snow load level; the target profile of the enhanced functional layer is usually designed to be evenly distributed in the transverse direction to ensure that the mechanical properties of the film surface are consistent and to avoid local weak areas.

[0045] For the internal functional layer (the first (layer), its target layer thickness design profile The target design thickness of the inner functional layer is determined based on the combined requirements of anti-drip effect and anti-fogging effect; in the application of ultra-wide greenhouse film, the thickness is determined accordingly. Take as to This thickness range allows the dripping agent to form a continuously distributed effective concentration gradient on the membrane surface, maintaining the dripping function for at least 12 to 24 months. Unlike the uniform design values ​​for each layer, in certain specific application scenarios, such as when there are differences in light conditions between the greenhouse roof and the sidewalls, the target profile of the inner functional layer can also be designed as a non-uniform distribution, i.e., different detection positions... place Different values ​​can be selected to adapt to the differentiated functional needs of different areas of the film surface; for example, due to the large angle of direct sunlight and high temperature in the central area of ​​the greenhouse top, the dripping agent precipitates quickly, so the value of the central area can be adjusted accordingly. The design value should be moderately increased by 10% to 15% to compensate for accelerated loss; the target layer thickness design profile for all functional layers should be adjusted accordingly. The process parameters are entered into the storage unit once before membrane bubble production and remain unchanged during normal production unless product specifications are changed or the process formula is adjusted; step 3 is executed each time by directly retrieving them from the storage unit. This is for use in subsequent difference calculations and does not need to be set repeatedly.

[0046] Step 3.2: Spatially align the real-time layer thickness distribution profile with the target layer thickness design profile at each detection position in the transverse direction of the membrane bubble. Calculate the layer thickness difference between the real-time layer thickness value and the target layer thickness design value at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. Specifically, this includes: aligning the real-time layer thickness distribution profile output in step 2... The target layer thickness design profile retrieved in step 3.1 First, check each transverse position of the membrane bubble. Spatial alignment is performed at each detection location point on the real-time profile; the purpose of spatial alignment is to ensure that each detection location point is aligned with the real-time profile. Corresponding position points on the target profile To ensure precise one-to-one correspondence and avoid deviations in difference calculations due to misalignment, the specific method for spatial alignment is as follows: the longitudinal centerline of the membrane bubble is used as the alignment reference, which is already set in step 2.2. Using the center line as the reference point, the origin of the lateral coordinates of the real-time profile and the target profile are unified at [location]. Because the real-time profile and the target profile share the same set of detection position coordinate systems; Since the position of the detection probe remains fixed during the membrane bubble formation process, spatial alignment is naturally achieved.

[0047] After spatial alignment is completed, the lateral detection position is determined for each membrane bubble. The real-time layer thickness distribution profile will be displayed at [location]. Layer thickness at the location (Right now (or interpolation result), subtract the target layer thickness to design the profile at the same location. Layer thickness design value The difference obtained is the value of the first detection position. functional layer thickness deviation Perform the above difference calculation on all functional layers and all detection locations one by one, and include all layer thickness deviations. Numbered by functional level and detection location number The layers are arranged and the thickness deviation of each functional layer along the transverse direction of the membrane bubble is obtained. The unit is ,when When, it indicates the detection location. First The real-time layer thickness of the functional layer is greater than the design value (too thick); when When the time is less than the design value, it indicates that the real-time layer thickness is less than the design value (too thin). The absolute value reflects the degree of deviation; the thickness deviation of this layer. As input data for step 4; from continuous profile and continuous layer thickness deviation profile This is used for the symmetry decomposition operation in step 4.1, where the continuous layer thickness design profile... By discrete The same cubic spline interpolation as in step 2.2 is performed to obtain the result.

[0048] In a preferred embodiment of the present invention, step 4 above may include: Step 4.1: Using the longitudinal centerline of the membrane bubble as the axis of symmetry, decompose the thickness deviation of each functional layer along the transverse spatial distribution profile of the membrane bubble into symmetrical deviation components and antisymmetrical deviation components. Specifically, this includes: using the longitudinal centerline of the membrane bubble as the axis of symmetry, decomposing the thickness deviation of each functional layer along the transverse spatial distribution profile of the membrane bubble into symmetrical deviation components and antisymmetrical deviation components. Decomposed into symmetrical deviation components and antisymmetric deviation components The longitudinal centerline of the bubble refers to the virtual axis extending vertically upward from the center of the die during the blow molding process. The bubble forming process has rotational symmetry about this axis. Therefore, the layer thickness segregation caused by the viscous encapsulation effect naturally has even symmetry characteristics about this centerline under ideal conditions: the segregation of low viscosity functional layers towards the two sides is symmetrical in amplitude and spatial position, and the aggregation of high viscosity functional layers towards the center is also symmetrical.

[0049] Symmetrical deviation components This describes the even-symmetric portion of the layer thickness deviation about the centerline, representing the deterministic influence of the systematic viscous encapsulation effect on the layer thickness distribution, i.e., satisfying... Antisymmetric bias components This describes the odd-symmetric portion of the layer thickness deviation about the centerline, i.e., satisfying... The main sources are non-systematic factors such as detection noise, random process disturbances, and instantaneous oscillation of the membrane bubble; the specific decomposition operation is as follows: for a membrane bubble located at a certain lateral position to the right of the centerline in the lateral direction. Layer thickness deviation value Take its left side position symmetrical about the center line. Layer thickness deviation value These two deviation values ​​form a symmetrical position pair.

[0050] The value of the symmetric deviation component at the symmetric position pair (and also equal to) (Due to its symmetry) it is calculated according to the following formula The value of the antisymmetric deviation component at this symmetrical position. Calculate according to the following formula In the formula and All units are ,therefore and The unit is also .

[0051] For all symmetrical positions of the membrane vesicle in the transverse direction ( from to Repeat the above decomposition operation (within the specified range) to obtain the complete profile of the symmetrical deviation components. and antisymmetric deviation component profile ; where the symmetrical deviation component The structural information carrying the viscous encapsulation effect, its profile exhibits a typical bimodal geometry, and it is the main object of subsequent envelope fitting analysis; antisymmetric deviation components In step 4.1, these components are separated and excluded, and no longer participate in subsequent steps 4.2 and 4.3, thus avoiding interference from random perturbations and asymmetry factors on the extraction of encapsulated features. It should be noted that in practical engineering, the current antisymmetric bias component can be further refined. Perform energy analysis: calculate its root mean square value and symmetry deviation component. The root mean square (RMS) values ​​are compared. If the RMS value of the antisymmetric component exceeds 30% of the RMS value of the symmetric component, it indicates that there is a significant asymmetric interference, such as obvious swaying of the bubble or uneven cooling on one side. At this time, an early warning signal is sent to the system to prompt the operator to check the symmetry of the bubble forming state.

[0052] As a supplementary explanation, for high-viscosity functional layers, i.e., apparent viscosity... For functional layers with viscosity higher than the median viscosity of all functional layers, the viscous encapsulation effect causes their deviation profile to be thinner at the edges (negative deviation valleys) and thicker at the center (positive deviation peaks), i.e., a double-valley, single-peak shape that is the mirror image of the low-viscosity functional layers. To ensure that the symmetric envelope model in step 4.2 is uniformly applicable to all functional layers, the symmetric deviation component of the high-viscosity functional layers is analyzed before fitting the double-peak envelope shape in step 4.2. Perform inversion preprocessing, Multiply by -1 to reverse the bi-valley-peak morphology of the high-viscosity layer to the bi-peak-valley morphology consistent with that of the low-viscosity layer.

[0053] After inversion preprocessing, the edge peak search algorithm and center valley search algorithm in step 4.2 are uniformly applicable to all functional layers, eliminating the need to maintain two separate sets of envelope fitting logic for functional layers with different viscosity types; the edge clustering extracted after inversion preprocessing and central segregation Unified representation of the first j The relative degree to which the thickness of the functional layer deviates from the design value in the edge and center regions of the membrane bubble. The larger the layer, the more severe the deviation at the edge region; The larger the value, the more severe the deviation of the layer in the central region; both feature values ​​are decoupled from the viscosity properties of the functional layer itself, and are used uniformly in step 5.2 to determine the adjustment strategy for edge partitioning and center partitioning.

[0054] Step 4.2: Fit the bimodal envelope morphology of the symmetrical deviation component to identify the peak positions and peak amplitudes of the lateral edge regions of the membrane bubble, obtaining the geometric characteristic parameters of the edge peaks; based on the geometric characteristic parameters of the edge peaks, determine the valley or plateau values ​​of the lateral central region of the membrane bubble, obtaining the geometric characteristic parameters of the central valley; based on the geometric characteristic parameters of the edge peaks and the central valley, construct a symmetrical envelope model with the two lateral edge peaks and the central valley as basic morphological templates, specifically including: fitting the obtained symmetrical deviation component... Perform bimodal envelope morphology fitting; due to the viscous encapsulation effect, low-viscosity functional layers tend to cluster towards the edges, while high-viscosity functional layers tend to cluster towards the center, with symmetrical deviation components. In the transverse direction of the membrane vesicle ( from to The spatial distribution profile (of the positive half-axis or equivalent full range) exhibits a typical bimodal geometry with a central valley: in the region near the edges of the membrane vesicles, The presence of two positive peaks indicates that the thickness of this functional layer in the edge region is significantly higher than the design value; in the central region of the membrane bubble ( (approximately equal to 0) The appearance of a negative valley or plateau value indicates that the thickness of the functional layer in the central region is significantly lower than the design value; this bimodal geometry is a direct mapping of the viscous encapsulation effect in the deviation data.

[0055] The process of fitting the bimodal envelope shape is as follows: first, for the first... Symmetrical deviation component profile of functional layers Only the positive half-axis portion is taken interval Perform analysis (due to) Due to the even symmetry, the analysis results of the positive semi-axis can directly yield the full-width profile characteristics through symmetry mapping. interval Perform point-by-point scanning within the edge search area to search. The local maximum value of the peak is located at the edge peak position, denoted as . The corresponding deviation amplitude is the edge peak amplitude, denoted as . , The unit is Due to symmetry, the position of the edge peak on the other side is... The amplitude is also The search employs a peak identification algorithm that locks the rising and falling edges: that is, during the scanning process, when three consecutive sampling points... When the value exhibits a rising, rising, falling pattern, the midpoint is marked as a candidate peak.

[0056] Further verification is needed to determine whether the amplitude of the candidate peak exceeds the preset peak identification threshold, which is set to [value missing]. ,in for The maximum value across the entire range, For the first The peak identification threshold coefficient for the functional layer is typically between 0.20 and 0.35. The specific calibration method is as follows: Under stable production operation conditions, i.e., no significant adhesive encapsulation effect occurs or the layer thickness deviation has been confirmed through offline measurement. Continuous data collection under the following operating conditions The profile data of the symmetrical deviation components of the complete cycle. No fewer than 30 rounds; for each round of data, calculate the standard deviation of the symmetric deviation component profile across the entire width of the membrane bubble. and maximum value The noise level is defined as follows: ; Take as and The ratio, i.e. This ratio represents the proportion of normal random fluctuations (3 times the standard deviation) relative to the maximum deviation.

[0057] For different functional layers Perform the above calibration separately, and take the upper and lower limits of the 95% confidence interval from multiple independent calibrations (no less than 5 times) as... The lower and upper limits of the value range; for low-viscosity functional layers such as outer and inner functional layers, due to the significant viscous encapsulation effect, Larger Typically, the smaller value is chosen, and the calibration result falls within the range of 0.20 to 0.28; for thicker functional layers such as enhancement layers, Typically, the larger value is chosen, and the calibration result falls within the range of 0.28 to 0.35; the functional layers are then... The calibration values ​​are stored in the system's characteristic parameter table and are numbered according to the functional layer during runtime. It can be directly retrieved for real-time use by the peak identification algorithm.

[0058] Extracting geometric feature parameters of edge peaks and Afterwards, according to Define the transverse center search region of the membrane vesicle; the center search region is taken as... interval That is, within the range from the center line to halfway to the edge peak, within the central search area, the search... The lowest value or flat area is used to determine the location of the transverse central valley of the membrane vesicle. and central valley amplitude For typical viscous encapsulation effects, the location of the central valley Approximately equal to 0 (at the center line). Typically a negative value (indicating thinness), the unit is... If the central area It presents itself as a flat platform rather than a sharp valley, which is not uncommon in actual production because the central concentration acts continuously over a wider lateral range. Take the center location of the platform area. Take as platform area The average value.

[0059] Based on the extracted edge peak geometric feature parameters and the geometric feature parameters of the central valley A symmetrical envelope model is constructed using the two edge peaks and the central valley as basic morphological templates. This symmetrical envelope model is a parameterized geometric description model that fully characterizes the spatial distribution of the thickness deviation caused by the viscous encapsulation effect along the transverse direction of the membrane bubble. Its envelope line is formed by the connection point... To the point The smooth, monotonic curve is constructed by using cubic spline interpolation to generate a smooth, continuous tangent curve between the two points. The slope of the tangent at the edge peak endpoint is set to 0 (horizontal tangent), and the slope of the tangent at the central valley endpoint is also set to 0, thus ensuring a natural transition of the envelope at the peaks and valleys. Due to symmetry, on the other side... The envelope of the positive semi-axis is obtained by passing the envelope of the positive semi-axis about the center line. The complete envelope is obtained by performing a mirror-symmetric mapping, meaning it satisfies even symmetry; the generated symmetric envelope model consists of the positive semi-axis envelope curve, the negative semi-axis mirror curve, and the horizontal coordinate axis ( The enclosed area (enclosed by the axis) visually represents the systematic deviation of the spatial distribution of the functional layer material caused by the adhesive encapsulation effect.

[0060] Step 4.3: Based on the symmetrical envelope model, calculate the edge eccentricity and center eccentricity respectively to obtain the encapsulation state characteristic quantities. The edge eccentricity is determined based on the ratio of the envelope area of ​​the two edge peaks to the total area of ​​the full-width deviation profile, and the center eccentricity is determined based on the ratio of the envelope depth of the central region to the design value of the target layer thickness of the corresponding functional layer. Specifically, this includes: based on the obtained symmetrical envelope model, calculating the edge eccentricity and center eccentricity respectively. Edge cohesion of functional layers and central segregation By merging the two feature quantities, we obtain the first... Encapsulation state features of functional layers; edge clustering In the formula The left edge peak (located in) The area of ​​the envelope of a point (e.g., the area enclosed by the left envelope curve segment, the horizontal coordinate axis, and the perpendicular lines between the intersections of the envelope lines on both sides of the peak and the coordinate axis), is expressed in units of 1. ; The right edge peak (located in) The envelope area of ​​(location), defined symmetrically to the left, with the same unit (integer units). .

[0061] in The total area of ​​the full-width deviation profile, i.e., the profile curve of the symmetrical deviation component. In the full range of the membrane bubble ( from to The area of ​​the entire region enclosed by the horizontal and vertical coordinate axes is taken as the absolute value and then integrated or summed, with the units being 1200 Å. In practical calculations, the composite trapezoidal rule is used to numerically integrate the envelope curve; the positive semi-axis envelope curve is set at the horizontal discrete position. (total) One sampling point, Position of the edge peak 、 Central Valley Location The corresponding envelope curve ordinate values ​​are respectively Sampling point spacing Taken as the distance between adjacent detection positions (i.e., the width of the membrane bubble). Divide by ), then the area of ​​the one-sided envelope Calculate according to the following formula: ; Total area of ​​full-width deviation profile By analyzing the full-amplitude symmetrical deviation component profile Using the same composite trapezoidal rule, from Points to get; It is a dimensionless ratio, with values ​​between 0 and 1. The closer to 1, the more severe the deviation of the layer thickness of the functional layer from the design value in the edge region of the membrane bubble. The encapsulation effect dominates the layer thickness deviation of the functional layer. The closer the value is to 0, the weaker the edge cohesion phenomenon, and the thickness deviation is mainly caused by other non-encapsulation factors; under typical operating conditions, when the viscous encapsulation effect is significant, The value ranges from 0.6 to 0.9.

[0062] Central segregation In the formula The envelope depth of the central region is the vertical distance from the horizontal coordinate axis (zero deviation line) to the lowest point of the central valley envelope. Unit take ; For the first The target layer thickness design value of the functional layer at the centerline of the membrane bubble, in units of 1. , It is a dimensionless ratio. The larger the value, the more severe the deviation of the functional layer's thickness from the design value in the central region of the bubble; under typical operating conditions, when the adhesive encapsulation effect is significant, The value is in the range of 0.15 to 0.30, which means that the deviation of the thickness of the functional layer in the central area from the design value is 15% to 30%.

[0063] Edge cohesion and central segregation Constituting the first Encapsulated state characteristics of functional layers; for all functional layers Perform steps 4.1 to 4.3 above respectively to obtain the set of encapsulated state feature quantities for all functional layers. The encapsulation state characteristic quantity serves as the input for step 5, providing a quantitative decision-making basis for determining the lip gap adjustment scheme for each lateral partition; from a physical perspective, and The spatial morphology and intensity of the viscous encapsulation effect are fully and quantitatively described from two complementary geometric dimensions: the proportion of total enrichment at the edge and the relative value of depth at the center.

[0064] In a preferred embodiment of the present invention, step 5 above may include: The membrane vesicle is divided laterally into Each zone adjustment segment The typical value is 6 to 12; each zone adjustment section corresponds to a transverse section in the die head flow channel where the lip gap can be adjusted independently, and the zone number is recorded as follows. ,in and These correspond to the left and right edge regions of the membrane vesicle, respectively. Approximately equal to The region corresponds to the central partition of the membrane vesicle.

[0065] Step 5.1: Obtain the current viscosity parameters of each functional layer melt. These current viscosity parameters include the melt index or apparent viscosity value of each functional layer melt at the current process temperature. Specifically, viscosity parameters are core physical quantities describing the flow resistance characteristics of each functional layer melt under real-time process conditions. In this embodiment, the viscosity parameter uses the apparent viscosity value. (Unit: Pa∙s), i.e., the first digit of the first digit of the second digit of the third digit of the fourth digit of the fifth digit of the sixth ... Functional layer melt at a specific shear rate and current process temperature The ratio of shear stress to shear rate at (degrees Celsius); there are two ways to obtain the apparent viscosity value, and one should be selected according to the sensor configuration at the production site.

[0066] Method 1 is the online direct measurement method, in which an online rheometer or online viscosity sensor is installed in the melt delivery pipeline of the extruder corresponding to each functional layer. In this embodiment, a slit-type online rheometer is selected and installed on the melt delivery pipeline between the metering section outlet and the die inlet of the extruder. The online rheometer has a built-in known geometric dimension (slit height). and slit width Precision machining ensures that the typical size is , The rectangular cross-section of the flow channel is used to measure the length of the flow channel. Typical value is 50mm; melt flow rate is set. As the flow passes through the measuring channel, pressure sensors at both ends of the rheometer collect the pressure difference between the inlet and outlet of the measuring channel in real time. According to the principle of slit rheology, the shear stress of the melt at the measuring channel wall is... Apparent shear rate of melt Apparent viscosity ;Will Compared with the current process temperature Associated record, as the first The current viscosity parameter of the functional layer melt can reflect the actual fluctuation of the melt viscosity in real time with high accuracy.

[0067] Method two is the temperature lookup table method, suitable for production sites without online viscosity sensors. This method involves pre-testing the functional layer melts of different material grades offline to establish a viscosity-temperature correspondence data table for each material grade at common shear rates in each functional layer, which is stored in the process parameter storage unit. During operation, the temperature is adjusted according to the current barrel temperature setpoint of each extruder. Based on the nominal shear rate calculated from the set extrusion flow rate, the current apparent viscosity value is obtained by looking up the corresponding viscosity-temperature data table through linear interpolation. The viscosity-temperature correlation data table is a two-dimensional table structure: the horizontal axis represents temperature (°C), with a temperature range of 180°C to 250°C, and a calibration point every 5°C, for a total of 15 columns; the vertical axis represents the nominal shear rate (°C). ), the range is to Ten points were evenly selected on a logarithmic coordinate system (10, 16, 25, 40, 63, 100, 160, 250, 400, 630). (10 rows in total; each cell stores the apparent viscosity value at the corresponding temperature and shear rate). (unit ),total Data points.

[0068] Offline rheological testing was performed using a high-pressure capillary rheometer (model Goettfert RG25), with a barrel diameter of 15mm and a test die length-to-diameter ratio of [missing information]. The die diameter is 1 mm. The test is conducted under a nitrogen protective atmosphere to eliminate the influence of melt thermal oxidation degradation on the measurement results. For each functional layer material, after being held at each temperature point for 15 minutes, the measurement is carried out sequentially at 10 shear rate levels. After stabilizing at each level for 30 seconds, the pressure sensor reading and melt temperature are recorded. The measurement is repeated 3 times and the average value is taken. The true apparent viscosity value is calculated by Bagley inlet pressure correction and Rabinowitsch non-Newton correction. This method does not require additional hardware investment, but the accuracy is slightly lower than that of online measurement methods.

[0069] Melt index can also be used as a supplementary or alternative indicator to viscosity parameters. (Unit: g / 10min) The mass of melt flowing out through a standard die (2.095 mm in diameter, 8.000 mm in length) within 10 minutes under standard temperature (typically 190℃ or 230℃) and standard load (typically 2.16 kg or 5.0 kg); there is an empirical inverse relationship between melt index and apparent viscosity. The higher, The lower the value, the better the melt flowability; the melt index is obtained by: according to the... The grade and batch number of the materials used in the functional layer are obtained from the quality inspection report provided by the material supplier, along with the standard MI value of that batch of materials; simultaneously, the current process temperature is considered. The deviation from the standard test temperature was corrected for the MI value using the Arrhenius temperature-viscosity relationship. Mainly as Auxiliary verification reference, when online measurement According to A viscosity parameter anomaly warning is triggered when the calculated expected viscosity value deviates by more than 20%; the current viscosity parameter is obtained. (and auxiliary verification indicators) ), and the material type and current process temperature of each functional layer. Record these together and transmit them to step 5.2 for use in formulating traffic allocation and adjustment strategies.

[0070] Step 5.2: Based on the encapsulation state characteristic quantity and the current viscosity parameter, determine the flow distribution adjustment strategy for each functional layer at each transverse partition adjustment segment of the membrane bubble. The flow distribution adjustment strategy indicates which functional layers require increased outflow resistance and which require decreased outflow resistance in each transverse partition. Specifically, this includes: based on the output encapsulation state characteristic quantity... and the obtained current viscosity parameters Determine the transverse position of each functional layer in the vesicle. The traffic allocation adjustment strategy at each partition adjustment segment; the traffic allocation adjustment strategy specifies the traffic allocation adjustment strategy at each horizontal partition adjustment segment. Within this framework, which functional layers need to increase outflow resistance, which functional layers need to decrease outflow resistance, and which functional layers need to maintain their current resistance?

[0071] The core principle for determining the strategy is: the essence of the viscous encapsulation effect is that the low-viscosity functional layer melt agglomerates towards the edge, squeezing the flow space of the high-viscosity functional layer melt in the edge region. Therefore, in the edge zone, it is necessary to suppress the excessive flow of the low-viscosity functional layer to release the flow space for the high-viscosity functional layer; in the center zone, it is necessary to suppress the excessive flow of the high-viscosity functional layer to release the flow space for the low-viscosity functional layer. The specific determination steps are as follows.

[0072] Identify low-viscosity and high-viscosity functional layer sets based on the viscosity parameters of each functional layer; Apparent viscosity values ​​of each functional layer Arranged in ascending order, the first... The functional layer with the lowest viscosity is grouped into the low viscosity functional layer set, and then... The functional layer with the highest viscosity is grouped into the high-viscosity functional layer set, among which and The value is determined based on the actual clustering of viscosity values ​​in each layer, typically... It consists of 1 to 2 layers (corresponding to the outer functional layer and / or the inner functional layer). It consists of 1 to 2 layers (corresponding to the reinforcing functional layer and / or adhesive layer); the functional layer with intermediate viscosity does not show significant segregation in the encapsulation effect and is not marked as an adjustment target for the time being.

[0073] According to edge cohesion Develop a regulation strategy for the edge partitions, specifically for the edge partitions on both sides of the membrane vesicle ( and and may be included as appropriate and (Check the edge segregation of each low-viscosity functional layer one by one) ,like Exceeding the preset edge clustering threshold , The value ranges from 0.25 to 0.35. The calibration process is as follows: during the initial stage of production, the system operates in an open-loop manner (without closed-loop regulation), and data is collected over multiple complete cycles. Data, calculation of normal random fluctuations The statistical upper bound is the mean plus three standard deviations. If the upper limit of this statistical value is set to 1.1 to 1.3 times, then the low viscosity functional layer is determined to be... If the traffic in the edge partition is too high, reduce it to the edge partition. The middle section is marked as requiring increased outflow resistance; simultaneously, due to the excessive flow of this low-viscosity functional layer crowding out the flow space of the high-viscosity functional layer in the same edge zone, the same zone... The high-viscosity functional layer inside is marked as requiring reduced outflow resistance; if If the threshold is not exceeded, the flow rate of the low-viscosity functional layer in the edge partition is still within an acceptable range and no adjustment mark is made.

[0074] According to the central segregation Develop a regulatory strategy for the central region; for the central region of the membrane vesicle ( Approximately equal to (One or two nearby partitions), examine the center cohesion of each high-viscosity functional layer one by one. ,like Exceeding the preset center-segregation threshold ,in The value ranges from 0.08 to 0.12, and the calibration method is the same as... Similarly, the high-viscosity functional layer is then determined. The traffic in the central partition is too high, so it should be redirected to the central partition. The middle section is marked as requiring increased outflow resistance; at the same time, the same section... The low-viscosity functional layer inside is marked as requiring reduced outflow resistance; if If the threshold is not exceeded, no adjustment mark is made; additionally, the total extrusion flow rate of each functional layer is independently controlled and kept constant by the screw speed of its corresponding extruder. The lip gap adjustment in steps 5.3 to 6.2 only changes the flow distribution ratio of the melt in each transverse zone of the bubble, and does not change the total extrusion amount of the melt in each functional layer. This flow conservation premise ensures that the adjustment action of this method only affects the transverse uniformity of the layer thickness, without interfering with the total material consumption of each functional layer and the overall extrusion balance of the bubble.

[0075] Fourth, for partitions where none of the encapsulation state characteristics exceed the corresponding threshold, typically intermediate transitional partitions between edge and center partitions, the current flow state is maintained unchanged, and no resistance adjustment is performed; through the above four-step strategy, the spatial clustering information of the encapsulation effect is determined. This is converted into indications for increasing or decreasing resistance in each zone, resulting in a complete flow distribution regulation strategy. This strategy is based on... OK The decision matrix is ​​organized in columns, with matrix elements taking values ​​of +1 (reduce resistance), -1 (increase resistance), and 0 (remain unchanged), for use in step 5.3.

[0076] Step 5.3, based on the flow distribution adjustment strategy, determine the lip gap adjustment direction and lip gap adjustment amount at each transverse adjustment section of the flow channel in each layer of the die head. Specifically, the lip gap refers to the opening gap between the upper and lower die lips at the outlet of the die head flow channel. The laminar flow channel corresponding to the functional layer is in the first layer. The current lip-mouth gap value at the partition is recorded as follows: (Unit: μm); The size of the lip gap directly determines the flow resistance of the melt flowing out of the corresponding flow channel. The smaller the gap, the greater the flow resistance and the smaller the melt flow rate; the larger the gap, the smaller the flow resistance and the greater the melt flow rate. The direction of lip gap adjustment is determined according to the flow distribution adjustment strategy. For functional layers marked in the strategy as needing to increase outflow resistance (decision matrix value: -1), In partition At this location, the direction of lip-mouth gap adjustment is determined to be the direction of decreasing gap, and the corresponding adjustment amount is... Take negative values; for functional layers marked as needing to reduce outflow resistance (decision matrix value +1). In partition At this location, the direction of lip-mouth gap adjustment is determined to be the direction of increasing the gap. Take positive values; for partitions marked as remaining unchanged (decision matrix value is 0), No gap adjustment is performed.

[0077] Lip-mouth gap adjustment The specific value depends on the current apparent viscosity of the corresponding functional layer. The mapping function expression used in this embodiment is determined together with the size of the encapsulation state feature of the partition; In the formula For the first The overall calibration coefficient for the functional layer is dimensionless and ranges from 0.02 to 0.20. Through prior experimental calibration, it was determined that on the test mold, with a known viscosity... Single-factor adjustment experiments were conducted on the functional layer melt: under fixed encapsulation state characteristic conditions, different... Candidate values ​​are adjusted with a gap, and the change in the thickness of each functional layer of the membrane bubble is measured after adjustment. The value that maximizes the approach of the layer thickness to the design value after a single adjustment without causing overshoot oscillation is selected. The value is used as the calibration coefficient for this functional layer; different functional layer materials differ due to their rheological properties. They vary, for example, the outer functional layer (LDPE / LLDPE blend system) Typically calibrated to 0.08 to 0.15, the enhanced functional layer (mLLDPE) It is typically calibrated to be between 0.04 and 0.10.

[0078] Each functional layer The calibration values ​​are stored in the system's calibration parameter table in the form of a lookup table. The calibration parameter table has a two-dimensional table structure and includes the following fields: Functional layer number. Functional layer material type and corresponding melt apparent viscosity range ( ), calibrated Value, calibration date, calibration operator number; each line corresponds to a functional layer, and the process is performed according to the functional layer number during runtime. Retrieve directly.

[0079] The detailed procedure for the single-factor adjustment experiment is as follows: Use a test die with geometry similar to the production die (lip diameter reduced to...). Number of partitions ), for the first The functional layer material, within a process temperature range of 190°C to 210°C, will exhibit encapsulation state characteristics. Fixed to typical operating conditions (e.g., edge segregation) Central segregation ), traversing candidates with a step size of 0.01 Values ​​range from 0.01 to 0.30; in each Apply gap adjustment amount under candidate value After waiting for a 3-fold stabilization delay, the layer thickness change of each functional layer was measured, and the changes in edge cohesion and center cohesion were recorded. and Define approximation efficiency The overshoot criterion is defined as the adjustment after... or The sign reversed (i.e., the value changed from positive to negative, indicating over-adjustment), select... Maximum and without overshoot The value is used as the calibration value for this functional layer under the specified viscosity conditions.

[0080] It should be noted that, since the calibration experiment was conducted on a real test mold, the interlayer coupling effect between the lip gap adjustments of each flow channel is naturally reflected in the calibration results. Therefore, there is no need to separately model and compensate for interlayer coupling effects during runtime.

[0081] For the first Functional layer in the first The encapsulation characteristic value at the partition is dimensionless and determined based on the spatial location of the partition and the encapsulation state characteristic; for the edge partitions on both sides of the membrane bubble ( For close to 1 or (partition) That is, the driving force for the adjustment of the edge partition comes from the edge segregation of the layer; for the central partition of the vesicle ( Approximately equal to (partition) That is, the adjustment driving force of the central partition comes from the central cohesion of that layer; for the intermediate transition partition... Pick and Partitioning based on the results of linear interpolation with distance weights When close to the edge, Closer When the distance from the center is relatively close, Closer The specific linear interpolation method is as follows: Set the partition... The horizontal center position is The edge of the membrane vesicle is The central position is ,but .

[0082] For the first The functional layer corresponds to the flow channel in the first layer. The current lip gap value (in μm) at the partition is obtained in real time by a gap sensor mounted on the die head; Multiplying the base value by the encapsulation characteristic value and the calibration coefficient, the physical meaning is: the magnitude of the adjustment is proportional to the current gap value; the larger the gap, the larger the absolute adjustment corresponding to the same proportion, which is consistent with the sensitivity law of melt flow rate to gap changes; after integrating the encapsulation characteristic value and direction sign, The complete formula is denoted as: ;in To adjust the direction sign, values ​​are +1 (increase the gap, corresponding to decreased resistance), -1 (decrease the gap, corresponding to increased resistance), or 0 (no adjustment). To avoid excessively large single adjustments causing instability in membrane bubble formation, [further adjustments are needed]. Set the upper limit of the adjustment amount. , Pick 20% to 30%; when calculated Exceed At that time, cut it to ; after confirmation (Includes information on adjustment direction and adjustment amount) , ) By partition number and laminar flow channel number Organize accordingly.

[0083] In a preferred embodiment of the present invention, step 6 above may include: Step 6.1: Based on the determined lip gap adjustment direction and adjustment amount, map the adjustment scheme to lip gap control commands corresponding to the lateral adjustment sections of the membrane bubble. Each lip gap control command specifies the target lip opening value for the corresponding laminar flow channel in the corresponding section, specifically including: based on the determined lip gap adjustment amount... Its symbol already includes the adjustment direction ( To reduce the gap, To increase the gap, (For non-adjustment), its absolute value The magnitude of the adjustment has already been included. The abstract adjustment scheme is mapped to specific lip gap control commands corresponding to the adjustment segments of each transverse zone of the membrane vesicle; first, for each zone that needs adjustment... and laminar flow channel Combinations (i.e.) (Items not equal to 0), read from the gap sensor mounted on the die head. Laminar flow channel in the first Actual opening value of the current lip gap at the partition (unit Then calculate the target opening value of the lip of the laminar flow channel at that section. This formula indicates that the target opening value equals the current opening value plus the increment of this adjustment. When the value is negative, the target opening value is less than the current value (the gap is closed). When the value is positive, the target opening value is greater than the current value (opening the gap wider).

[0084] Adjust each partition that needs adjustment and laminar flow channel The combinations are packaged into a single lip-mouth gap control instruction; each lip-mouth gap control instruction contains the following fields: partition number. Laminar flow channel number Current gap value Target gap value Adjusting the increment And the adjustment priority; the adjustment priority is based on The size of the partition and the weight of its impact on the encapsulation effect are combined to determine the overall effect. Larger instructions located on critical edge or center partitions have higher priority, ensuring that partitions with the greatest impact on encapsulation effects are adjusted first. After all lip gap control instructions are generated, they are sorted by adjustment priority from high to low and stored in the instruction queue, awaiting execution in step 6.2. The combination of partitions and laminar flow channels does not generate control commands.

[0085] Step 6.2: According to the lip gap control command corresponding to each zone adjustment section, adjust the lip opening of the corresponding layer flow channel to the specified target lip opening value to change the outflow resistance of the melt in each functional layer of each transverse zone, so that the melt flow distribution of each functional layer in the transverse direction of the membrane bubble is adjusted towards homogenization. Specifically, this includes: according to the generated lip gap control command corresponding to each zone adjustment section, retrieving commands from the command queue in descending order of priority, and adjusting the lip opening of the corresponding layer flow channel to the specified target lip opening value. The specific method of adjustment is as follows: the gap adjustment execution unit is installed on the mold head. In this embodiment, a piezoelectric ceramic micro-displacement actuator is used, and its resolution can reach [missing information]. Full itinerary Alternatively, a precision ball screw mechanism driven by a stepper motor can be used, with a single-step displacement of... Upon receiving control commands, the drive mold lip generates a micrometer-level displacement along the vertical direction.

[0086] During adjustment, the gap sensor operates at a speed not lower than... The sampling rate provides real-time feedback of the current gap opening value, and the control system compares the feedback value with... Perform closed-loop PID comparison, when Once the adjustment is determined to be in place, the drive is stopped and the current position is locked. The threshold for determining when the gap adjustment is in place is taken as follows: to After the lip gap of one laminar flow channel in a zone adjustment section is adjusted, a short stabilization waiting time (approximately) is allowed. to (to stabilize the local melt flow state), continue executing the next instruction in the instruction queue until all zones and laminar flow channels that need adjustment have completed gap adjustment.

[0087] As the lip gaps of each zone change, the outflow resistance of the melt in each functional layer of the corresponding transverse zone undergoes corresponding physical changes: The reduced gap in the partition increases the flow resistance of the melt in that layer, correspondingly reducing the melt flow rate and thus suppressing excessive outflow of the low-viscosity functional layer in that partition; The increased gap between the partitions reduces the flow resistance of the melt in that layer, resulting in a corresponding increase in melt flow rate. This releases the flow space for the high-viscosity functional layer in that partition. The differential adjustment of outflow resistance has a combined effect on the melt flow rate distribution of each functional layer in the transverse direction of the bubble. The melt flow rate of the low-viscosity functional layer, which originally tended to concentrate in the edge region, is specifically suppressed, while the melt flow rate of the high-viscosity functional layer, which originally tended to concentrate in the center region, is redistributed to the edge region. Ultimately, this adjusts the melt flow rate distribution of each functional layer in the transverse direction of the bubble towards uniformity, gradually eliminating the layer thickness segregation caused by the viscous encapsulation effect.

[0088] In a preferred embodiment of the present invention, step 7 above may include: Step 7.1: After completing this round of adjustment, trigger a new round of layer thickness measurement signal acquisition to obtain a new round of layer thickness measurement signal. Specifically, this includes: after completing all lip gap adjustment actions in step 6, waiting for a preset stabilization delay. This allows the melt flow state within the membrane bubble to reach a new steady-state equilibrium after gap adjustment; stabilization delay The extrusion time is determined by a combination of factors including the bubble size, melt viscosity, and extrusion line speed. A typical selection principle is as follows: This equals the time required for the membrane bubble to travel from the die exit to the detection position (i.e., the distance from the die exit to the detection position divided by the extrusion line speed) plus... to The flow field stability margin; for a typical ultra-wide greenhouse film production line, the width of the film bubble is equal to Extrusion line speed Detection position distance from the die head exit ;in In actual deployment, The value is stored in the system's process parameter table and is automatically retrieved based on the current product specifications and linear velocity. The value range is [range missing]. to .

[0089] After the stabilization delay ends, the closed-loop iterative control module issues a new round of acquisition start command. The command is sent to the layer thickness signal acquisition module in step 1, driving the detection probe to re-detect each preset position. The ultrasonic wave transmission and echo reception operations are performed sequentially; the operation procedure for the new round of acquisition is exactly the same as that for the first round of acquisition, at each detection location. The system sequentially performs detection signal transmission, reflected echo reception, echo characteristic peak identification, and layer thickness calculation to obtain a new round of layer thickness measurement signals. The new round of layer thickness measurement signals contains the latest status information of the layer thickness distribution of each functional layer after the adjustment of the lip gap in this round.

[0090] Step 7.2: Based on the new round of layer thickness measurement signals, analyze the layer thickness value of each functional layer at each detection position to obtain a new round of real-time layer thickness distribution profile along the transverse direction of the membrane bubble for each functional layer; compare the difference between the new round of real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the updated layer thickness deviation, specifically including: based on the obtained new round of layer thickness measurement signals, sequentially perform layer thickness value analysis and deviation calculation, first following the same process as in Step 2, sorting the new round of layer thickness measurement signals by functional layer category. and detection location Discrete layer thickness values ​​are extracted, and cubic spline interpolation is performed on the discrete layer thickness data sequences of each functional layer along the transverse direction of the membrane bubble to reconstruct a new round of real-time layer thickness distribution profiles for each functional layer along the transverse direction of the membrane bubble, denoted as […]. Cubic spline interpolation coefficients used in profile analysis and The interpolation coefficients need to be recalculated using the same method as in step 2.2, since the layer thickness distribution has changed after this round of adjustment. The interpolation coefficients need to be re-solved based on the new discrete data points.

[0091] A new round of real-time layer thickness distribution profiles were obtained. After that, At the detection location value at The target layer thickness design profile is in the same position as the corresponding functional layer. Design value at the location (This has already been retrieved and stored in system memory in step 3.1; if the process remains unchanged, it does not need to be retrieved again.) Calculate the difference for each detection position using the following formula: For all functional layers and all detection locations Performing this calculation yields the updated layer thickness deviation. The updated layer thickness deviation reflects the actual improvement effect of this round of lip gap adjustment on layer thickness distribution. Compared to the previous round If the adjustment shrinks, it indicates that the direction of adjustment is correct and effective; conversely, if... An increase or no change indicates that the adjustment strategy for this round needs to be adjusted in the next iteration, for example, by appropriately increasing the calibration coefficient. Or adjust the packaging feature values The weight of the value.

[0092] Step 7.3: Determine whether the updated layer thickness deviations all fall within the preset allowable deviation range. If the layer thickness deviations of each functional layer along the transverse direction of the membrane bubble all fall within the preset allowable deviation range, terminate the iterative closed-loop adjustment. Otherwise, use the updated layer thickness deviations as the input for the next adjustment cycle, and re-execute the encapsulation state feature extraction, lip gap adjustment scheme determination, and outflow resistance adjustment. Specifically, this includes: obtaining the updated layer thickness deviations. Then, make a judgment Whether all deviations fall within the preset allowable range; the preset allowable range is a pre-defined allowable range for layer thickness deviation, defined as a percentage of the target layer thickness design value for each functional layer.

[0093] In this embodiment, the upper and lower limits of the preset allowable deviation range are taken as positive and negative. , The thickness should be 3% to 5% of the target layer design value, with the specific value determined comprehensively based on the functional performance tolerance of the membrane bubble product and the economics of process control; for high-end greenhouse film products with high requirements for functional uniformity... Take 3% (i.e.) (±3%); for general-purpose agricultural films, 5% is acceptable; The calibration process is as follows: Using the target layer thickness design value as a benchmark, offline experiments are conducted to determine the dripping duration and thermal insulation performance degradation curves under different deviation amplitudes. The deviation amplitude corresponding to a functional performance decrease of no more than 10% is taken as the baseline. The upper limit of the possible values.

[0094] The specific process of convergence judgment is as follows: traverse all functional layers. and all detection locations Calculate the absolute value of each deviation. Relative to its target design value The percentage; if for all and All conditions are met If the layer thickness distribution has reached the homogenization requirement, the iterative closed-loop adjustment is terminated, and the current lip gap setting for each partition is maintained. If the system remains unchanged, it exits the closed-loop control mode and enters the steady-state monitoring mode; in the steady-state monitoring mode, the system still operates at a low frequency (e.g., every...). to (Once) Continue to collect layer thickness signals. If the deviation is detected to exceed the allowable range again, for example, due to changes in ambient temperature or fluctuations in raw material batches causing a drift in melt viscosity, the closed-loop regulation will be automatically restarted.

[0095] If any functional layer exists any detection location If the layer thickness deviation at a certain point does not meet the above convergence conditions, it is determined that the current adjustment has not yet fully converged the layer thickness distribution; at this time, the updated layer thickness deviation will be... As input for the next adjustment cycle, first re-execute the encapsulation state feature extraction in step 4: for the updated deviation profile A new round of symmetric envelope decomposition (step 4.1), bimodal envelope shape fitting (step 4.2), and feature quantization calculation (step 4.3) is performed to obtain the updated encapsulation state feature quantities. Based on Repeat step 5 to determine the lip-mouth gap adjustment scheme; further execute step 6 to adjust the lip-mouth gap and change the outflow resistance; after the adjustment is completed, return to step 7.1 to trigger a new round of signal acquisition.

[0096] Through the complete closed-loop iterative process described above—including data acquisition, analysis, deviation quantification, encapsulation feature extraction, gap adjustment, execution adjustment, re-acquisition, and convergence judgment—the thickness deviation of each functional layer along the transverse direction of the membrane bubble is determined. Gradually reduce until all deviations fall within the preset allowable range. Within; in actual production testing, for the initial deviation amount In typical operating conditions ranging from 20% to 40%, where the sticky encapsulation effect has resulted in the functional layer being 20% ​​to 40% thicker at the edges and 20% to 30% thinner at the center, the closed-loop iteration typically requires 4 to 8 rounds to converge, with a total adjustment time of approximately [missing information]. Iteration rounds, where The time required for a single round of signal acquisition and data processing is approximately ( to ), The time required for adjusting the clearance of a single wheel is approximately to .

[0097] During iterative closed-loop regulation, the current viscosity parameter The data can be reacquired in each iteration to track potential fluctuations in process temperature, or it can be reused from the previous iteration once the process temperature has stabilized. The value is determined to reduce system overhead, and the specific method is based on the actual fluctuation range of the process temperature: if the barrel setting temperature fluctuation between two consecutive cycles does not exceed ±2℃, then the value of the previous cycle is used. Otherwise, repeat step 5.1 to obtain the updated version. .

[0098] like Figure 2 As shown, embodiments of the present invention also provide an intelligent control system for the uniformity of co-extruded layer thickness of ultra-wide greenhouse film, comprising: The layer thickness signal acquisition module is used to acquire layer thickness measurement signals at multiple preset detection positions along the transverse direction of the membrane bubble in real time after the membrane bubble has formed a multi-layer co-extrusion structure of three or more layers and before cooling and shaping. The profile analysis module is used to analyze the layer thickness value of each functional layer at each detection position based on the layer thickness measurement signal, and obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble. The deviation calculation module is used to compare the difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. The encapsulation feature extraction module is used to analyze the variation trend and amplitude distribution of the layer thickness deviation along the transverse direction of the bubble, and extract the encapsulation state feature quantity characterizing the viscous encapsulation effect. The encapsulation state feature quantity includes the edge cohesion degree determined based on the layer thickness deviation envelope area of ​​each functional layer in the edge region of the bubble and the center cohesion degree determined based on the layer thickness deviation envelope depth of each functional layer in the center region of the bubble. The lip gap adjustment calculation module is used to determine the lip gap adjustment direction and lip gap adjustment amount of each layer of the die head flow channel at each transverse partition adjustment section based on the encapsulation state characteristic quantity and the current viscosity parameters of the melt of each functional layer. The melt flow rate regulation execution module is used to adjust the outflow resistance of the melt in each functional layer of the corresponding transverse partition according to the determined lip gap adjustment direction and lip gap adjustment amount, thereby changing the melt flow rate distribution of each functional layer in the transverse direction of the membrane bubble. The closed-loop iterative control module is used to trigger a new round of layer thickness measurement signal acquisition after the current adjustment is completed. The updated layer thickness deviation is used as the input for the next adjustment cycle, and the iteration continues until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range.

[0099] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0100] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent control of the uniformity of co-extruded layer thickness of ultra-wide greenhouse film, characterized in that, The method includes: After the membrane bubble forms a multi-layer co-extrusion structure of three or more layers and before cooling and shaping, the layer thickness measurement signals of each functional layer are collected in real time at multiple preset detection positions along the transverse direction of the membrane bubble. Based on the layer thickness measurement signal, the layer thickness value of each functional layer at each detection position is analyzed to obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble; The difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer is compared at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. The variation trend and amplitude distribution of the thickness deviation along the transverse direction of the bubble were analyzed, and the encapsulation state characteristics that characterize the viscous encapsulation effect were extracted. The encapsulation state characteristics include the edge cohesion determined based on the envelope area of ​​the thickness deviation of each functional layer in the edge region of the bubble and the center cohesion determined based on the envelope depth of the thickness deviation of each functional layer in the center region of the bubble. Based on the encapsulation state characteristics, and combined with the current viscosity parameters of the melt of each functional layer, the lip gap adjustment direction and lip gap adjustment amount of each layer flow channel of the die head at each transverse partition adjustment section are determined. Based on the determined lip gap adjustment direction and lip gap adjustment amount, adjust the outflow resistance of the melt in each functional layer of the corresponding transverse partition, and change the melt flow distribution of each functional layer in the transverse direction of the membrane bubble. After completing this round of adjustment, the layer thickness measurement signal is reacquired, and the updated layer thickness deviation is used as the input for the next adjustment cycle. The process continues until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range.

2. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 1, characterized in that, After the membrane bubble forms a multilayer co-extruded structure with three or more layers, but before cooling and setting, the layer thickness measurement signals of each functional layer at multiple preset detection positions along the transverse direction of the membrane bubble are collected in real time, including: At each preset detection position in the transverse direction of the membrane bubble, a detection signal is emitted to the surface of the membrane bubble so that the detection signal penetrates each functional layer and generates reflected echoes at the interlayer interfaces between adjacent functional layers. The functional layers are divided according to the functional properties of the melt supplied by the corresponding extruder, including an outer functional layer that provides weather protection, an reinforcing functional layer that provides structural strength, and an inner functional layer that provides surface functional properties. By receiving the reflected echoes, a multi-interface echo sequence carrying interlayer interface position information at the corresponding detection position is obtained. Based on the multi-interface echo sequence, the echo characteristic peaks corresponding to each inter-layer interface are identified, and the time position parameters of each echo characteristic peak are extracted to obtain the interface position characteristic parameters of each inter-layer interface. Based on the interface position feature parameters between adjacent layers, the layer thickness value of each functional layer at the corresponding detection position is calculated to obtain the independent layer thickness measurement value of each functional layer at the corresponding detection position. After completing the signal acquisition and layer thickness calculation at all preset detection positions in the transverse direction of the membrane bubble, the independent layer thickness measurement values ​​of each functional layer at each detection position are integrated into a multi-channel layer thickness measurement signal.

3. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 2, characterized in that, Based on the layer thickness measurement signal, the layer thickness value of each functional layer at each detection position is analyzed to obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble, including: From the multi-channel layer thickness measurement signal, the layer thickness value is extracted according to the functional layer category and the transverse detection position of the membrane bubble, so as to obtain the discrete layer thickness data sequence of each functional layer at each preset detection position in the transverse direction of the membrane bubble. The discrete layer thickness data sequence of each functional layer is spatially continuous and reconstructed along the transverse direction of the membrane bubble to obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble.

4. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 3, characterized in that, The difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer is compared at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble, including: The target layer thickness design profile of each functional layer is retrieved from the preset process parameter storage unit. The target layer thickness design profile defines the layer thickness design value of the corresponding functional layer at each detection position in the transverse direction of the membrane bubble. The real-time layer thickness distribution profile and the target layer thickness design profile are spatially aligned at each detection position in the transverse direction of the membrane bubble. The layer thickness difference between the real-time layer thickness value and the target layer thickness design value is calculated at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble.

5. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 4, characterized in that, The variation trend and amplitude distribution of the layer thickness deviation along the transverse direction of the bubble are analyzed to extract encapsulation state characteristics that characterize the viscous encapsulation effect. These encapsulation state characteristics include edge cohesion determined based on the envelope area of ​​the layer thickness deviation of each functional layer at the bubble edge and center cohesion determined based on the envelope depth of the layer thickness deviation of each functional layer at the bubble center. With the longitudinal centerline of the membrane bubble as the axis of symmetry, the thickness deviation of each functional layer along the transverse spatial distribution profile of the membrane bubble is decomposed into symmetric deviation components and antisymmetric deviation components. The bimodal envelope morphology of the symmetrical deviation component is fitted to identify the peak position and peak amplitude of the lateral edge regions of the membrane bubble, and the geometric feature parameters of the edge peaks are obtained. Based on the geometric feature parameters of the edge peaks, the valley or plateau value of the lateral central region of the membrane bubble is determined, and the geometric feature parameters of the central valley are obtained. Based on the geometric feature parameters of the edge peaks and the geometric feature parameters of the central valley, a symmetrical envelope model with the two lateral edge peaks and the central valley as the basic morphological templates is constructed. Based on the symmetric envelope model, edge eccentricity and center eccentricity are calculated to obtain the encapsulation state characteristic quantities. The edge eccentricity is determined based on the ratio of the envelope area of ​​the two edge peaks to the total area of ​​the full-width deviation profile, and the center eccentricity is determined based on the ratio of the envelope depth of the central region to the design value of the target layer thickness of the corresponding functional layer.

6. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 5, characterized in that, Based on the aforementioned encapsulation state characteristics and combined with the current viscosity parameters of the melt in each functional layer, the lip gap adjustment direction and lip gap adjustment amount at each transverse partition adjustment section of the die head flow channel are determined, including: Obtain the current viscosity parameters of each functional layer melt, wherein the current viscosity parameters include the melt index or apparent viscosity value of each functional layer melt at the current process temperature; Based on the encapsulation state characteristic quantity and the current viscosity parameter, the flow distribution adjustment strategy of each functional layer at each transverse partition adjustment segment of the membrane bubble is determined. The flow distribution adjustment strategy indicates the functional layers that need to increase outflow resistance and the functional layers that need to decrease outflow resistance in each transverse partition. Based on the flow distribution adjustment strategy, the adjustment direction and adjustment amount of the lip gap at each transverse partition adjustment section of each layer of the die head flow channel are determined.

7. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 6, characterized in that, Based on the determined lip gap adjustment direction and adjustment amount, adjust the outflow resistance of the melt in each functional layer of the corresponding transverse zone, thereby changing the melt flow distribution of each functional layer in the transverse direction of the membrane bubble, including: Based on the determined lip gap adjustment direction and lip gap adjustment amount, the adjustment scheme is mapped to the lip gap control instructions corresponding to each transverse partition adjustment segment of the membrane bubble. Each lip gap control instruction specifies the target opening value of the lip of the corresponding laminar flow channel in the corresponding partition. According to the lip gap control command corresponding to each zone adjustment section, adjust the lip opening of the corresponding layer flow channel to the specified lip target opening value, so as to change the outflow resistance of the melt of each functional layer in each transverse zone, and adjust the melt flow distribution of each functional layer in the transverse direction of the membrane bubble towards the direction of homogenization.

8. The intelligent control method for uniformity of co-extruded layer thickness of ultra-wide greenhouse film according to claim 7, characterized in that, After completing this round of adjustment, the layer thickness measurement signal is reacquired, and the updated layer thickness deviation is used as the input for the next adjustment cycle. This process is continued iteratively until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range, including: After completing this round of adjustment, a new round of layer thickness measurement signal acquisition is triggered to obtain a new round of layer thickness measurement signal; Based on the new round of layer thickness measurement signals, the layer thickness value of each functional layer at each detection position is analyzed to obtain a new round of real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble; the new round of real-time layer thickness distribution profile is compared with the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the updated layer thickness deviation. Determine whether the updated layer thickness deviation falls within the preset allowable deviation range. If the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range, terminate the iterative closed-loop adjustment. Otherwise, use the updated layer thickness deviation as the input for the next adjustment cycle, and re-execute the encapsulation state feature extraction, lip gap adjustment scheme determination, and outflow resistance adjustment.

9. An intelligent control system for ensuring uniformity of co-extruded layer thickness of ultra-wide greenhouse film, wherein the system implements the method as described in any one of claims 1 to 8, characterized in that, include: The layer thickness signal acquisition module is used to acquire layer thickness measurement signals at multiple preset detection positions along the transverse direction of the membrane bubble in real time after the membrane bubble has formed a multi-layer co-extrusion structure of three or more layers and before cooling and shaping. The profile analysis module is used to analyze the layer thickness value of each functional layer at each detection position based on the layer thickness measurement signal, and obtain the real-time layer thickness distribution profile of each functional layer along the transverse direction of the membrane bubble. The deviation calculation module is used to compare the difference between the real-time layer thickness distribution profile and the target layer thickness design profile of the corresponding functional layer at each detection position to obtain the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble. The encapsulation feature extraction module is used to analyze the variation trend and amplitude distribution of the layer thickness deviation along the transverse direction of the bubble, and extract the encapsulation state feature quantity characterizing the viscous encapsulation effect. The encapsulation state feature quantity includes the edge cohesion degree determined based on the layer thickness deviation envelope area of ​​each functional layer in the edge region of the bubble and the center cohesion degree determined based on the layer thickness deviation envelope depth of each functional layer in the center region of the bubble. The lip gap adjustment calculation module is used to determine the lip gap adjustment direction and lip gap adjustment amount of each layer of the die head flow channel at each transverse partition adjustment section based on the encapsulation state characteristic quantity and the current viscosity parameters of the melt of each functional layer. The melt flow rate regulation execution module is used to adjust the outflow resistance of the melt in each functional layer of the corresponding transverse partition according to the determined lip gap adjustment direction and lip gap adjustment amount, thereby changing the melt flow rate distribution of each functional layer in the transverse direction of the membrane bubble. The closed-loop iterative control module is used to trigger a new round of layer thickness measurement signal acquisition after the current adjustment is completed. The updated layer thickness deviation is used as the input for the next adjustment cycle, and the iteration continues until the layer thickness deviation of each functional layer along the transverse direction of the membrane bubble falls within the preset allowable deviation range.

Images