Method and system for controlling thickness of wedge-shaped PVB interlayer film

Abstract

The present invention proposes a control method and system for the thickness of a wedge-shaped PVB intermediate film, in the field of plastic film control technology, the method comprising: obtaining target thicknesses of various positions of a PVB film sheet under different classifications through historical data; obtaining the actual thickness of various positions of the PVB film sheet, and obtaining the thickness adjustment amount of each position through the actual thickness and the target thickness; obtaining the process parameter adjustment amount and the adjustment priority upgrade according to the thickness adjustment amount of each position; adjusting the process parameters according to the process parameter adjustment amount and the adjustment priority; obtaining the actual thickness after parameter adjustment after a preset time interval, and determining whether to adjust again according to the actual thickness; and dynamically adjusting the process parameters by real-time monitoring and analysis of the actual thickness and the target thickness of various positions of the film sheet, and combining historical data to effectively improve the consistency and stability of the thickness of the PVB film sheet, realize precise control of the thickness, improve production efficiency and product quality, and meet different wedge degree requirements at the same time.

Description

Technical Field

[0001] The present application relates to the technical field of plastic film control, and in particular to a method and system for controlling the thickness of a wedge-shaped PVB intermediate film. Background Art

[0002] Polyvinyl Butyral (PVB) interlayer is an important safety glass interlayer material, widely used in the automotive, construction and other industries. In actual production, PVB film needs to meet certain wedge requirements to meet the needs of different application scenarios. Appropriate wedge can reduce visual distortion. PVB film thickness control is crucial to ensure the performance of the final product, because uneven thickness will affect the safety, optical quality and sound insulation of the glass. However, it is a challenge to achieve precise wedge control because it requires not only the average thickness of the entire film to meet the standard, but also the thickness distribution at each position along the width direction to meet specific design requirements.

[0003] In the traditional PVB film production process, due to various factors in the extrusion process such as changes in temperature, pressure, speed, and differences between raw material batches, it is difficult to control the consistency and stability of product thickness. Existing thickness control methods are usually based on manual experience or simple feedback control systems. The adjustment methods have significant shortcomings, which limits product quality and production efficiency.

[0004] First, manual adjustment is inefficient. Operators need to adjust multiple bolts one by one, which is a cumbersome and time-consuming process. For wide molds, manual adjustment requires repeated adjustments at multiple points, and the thickness and wedge adjustment cannot be completed quickly. The labor intensity of operators is high, the adjustment efficiency is low, and it is difficult to achieve a quick response during the production process. Secondly, the accuracy of manual adjustment is poor. Manual adjustment relies on the experience and skills of the operator, and it is difficult to achieve high-precision control. Due to the limited accuracy of manual adjustment, the uniformity of diaphragm thickness and wedge is difficult to guarantee, resulting in unstable product quality. Especially in high-demand applications, slight deviations in diaphragm thickness and wedge will significantly affect product performance. In addition, manual adjustment has large manual errors. The personal ability and working status of the operator directly affect the adjustment effect, and there are differences in the adjustment effect between different operators. The repeatability and consistency of manual adjustment are poor, and it is impossible to guarantee that the adjustment results are completely consistent each time, which increases the uncertainty of the production process and the instability of product quality. Summary of the invention

[0005] Based on the above problems, the present application provides a method and system for controlling the thickness of a wedge-shaped PVB intermediate film. By real-time monitoring and analyzing the actual thickness and target thickness of each position of the film, and combining historical data to dynamically adjust key process parameters such as extrusion temperature, extrusion volume, cooling speed, and roller gap distance, the thickness of the PVB film is effectively improved. The consistency and stability of the thickness are achieved, precise control of the thickness is achieved, and production efficiency and product quality are improved, while meeting different wedge degree requirements.

[0006] The purpose of this application is achieved by the following technical solutions:

[0007] In a first aspect, the present application provides a method for controlling the thickness of a wedge-shaped PVB interlayer, the method comprising:

[0008] Through historical data, the target thickness of PVB film at each position under different classifications is obtained;

[0009] Obtaining the actual thickness of each position of the PVB film, and obtaining the thickness adjustment amount of each position through the actual thickness and the target thickness;

[0010] According to the thickness adjustment amount of each position, the process parameter adjustment amount and the adjustment priority are obtained; according to the process parameter adjustment amount and the adjustment priority, the process parameter is adjusted; the process parameters include extrusion temperature, extrusion amount, cooling speed and roll gap distance;

[0011] The preset time interval is obtained through the system response time and stabilization time after the adjustment of each parameter at each position; after the preset time interval, the actual thickness after the parameter adjustment is obtained, and whether to adjust again is determined according to the actual thickness.

[0012] In a second aspect, the present application provides a control system for the thickness of a wedge-shaped PVB interlayer, the system comprising:

[0013] The target thickness determination module is used to obtain the target thickness of each position of the PVB film under different classifications through historical data;

[0014] An adjustment amount determination module is used to obtain the actual thickness of each position of the PVB film, and obtain the thickness adjustment amount of each position through the actual thickness and the target thickness;

[0015] An adjustment module, for obtaining a process parameter adjustment amount and an adjustment priority upgrade according to a thickness adjustment amount at each position; adjusting the process parameters according to the process parameter adjustment amount and the adjustment priority; the process parameters include extrusion temperature, extrusion amount, cooling speed and roll gap distance;

[0016] The feedback module is used to obtain a preset time interval through the system response time and stabilization time after each parameter at each position is adjusted; after the preset time interval, the actual thickness after the parameter adjustment is obtained, and whether to adjust again is determined according to the actual thickness.

[0017] The beneficial effects of the present invention include: accurately setting the target thickness of each position of the PVB film under different classifications, ensuring the accuracy of the benchmark for thickness control. The actual thickness of each position of the PVB film is obtained in real time, and compared with the target thickness, and the thickness adjustment amount of each position is calculated, so as to realize dynamic monitoring and adjustment of the thickness. Based on the machine learning model, the process parameter adjustment amount and adjustment priority are intelligently determined according to the thickness adjustment amount of each position, avoiding blind adjustment and improving the adjustment efficiency. Considering the influence of actual production environment factors such as network congestion on the system response time and stability time, through a reasonable preset time interval, it is ensured that the actual thickness after adjustment is obtained at the right time; while reducing erroneous judgments, it is also possible to reduce the amount of collected data, and at the same time, the thickness information can be obtained in time, and adjustments can be made according to the thickness information; reducing product failure rate; through accurate comparison and judgment, unnecessary frequent adjustments can be avoided, and thickness deviations can also be responded to in time to ensure the efficiency and accuracy of the adjustment; different product models, material ratios and mold models can be classified and processed, ensuring that the thickness control under different production conditions can achieve the best effect. By adjusting the process parameters in real time, it is possible to quickly respond to changes in the production process, improving the flexibility and adaptability of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of a method for controlling the thickness of a wedge-shaped PVB interlayer provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] Below, the present application is further described in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0020] The present application provides a method for controlling the thickness of a wedge-shaped PVB interlayer film, the method comprising:

[0021] Through historical data, the target thickness of PVB film at each position under different classifications is obtained;

[0022] Obtaining the actual thickness of each position of the PVB film, and obtaining the thickness adjustment amount of each position through the actual thickness and the target thickness;

[0023] According to the thickness adjustment amount of each position, the process parameter adjustment amount and the adjustment priority are obtained; according to the process parameter adjustment amount and the adjustment priority, the process parameter is adjusted; the process parameters include extrusion temperature, extrusion amount, cooling speed and roll gap distance;

[0024] The preset time interval is obtained through the system response time and stabilization time after the adjustment of each parameter at each position; after the preset time interval, the actual thickness after the parameter adjustment is obtained, and whether to adjust again is determined according to the actual thickness.

[0025] The working principle and effect of the above technical solution are:

[0026] First, obtain historical data, which records the actual thickness measurement results of PVB film at various locations (such as edge, center or other specific areas) under different classifications (which may be based on product type, production batch, equipment status, etc.), and pre-determine the target thickness based on product specifications and historical data.

[0027] Then, the actual thickness of each position of the PVB film is measured in real time or regularly; usually completed by high-precision measuring equipment, such as laser rangefinders and thickness sensors. The measured actual thickness is compared with the preset target thickness to calculate the thickness adjustment amount at each position. This adjustment amount reflects the difference between the current thickness and the target thickness and is the basis for subsequent adjustment of process parameters.

[0028] According to the thickness adjustment amount at each position, the process parameters that need to be adjusted and their adjustment amount are calculated. These process parameters include extrusion temperature, extrusion amount, cooling speed and roller gap distance, which together determine the thickness and shape of the PVB film.

[0029] At the same time, determine the priority of adjusting these parameters. This is usually based on factors such as the difficulty of adjustment, the impact on the production process, and the cost.

[0030] Then, the process parameters are adjusted accordingly according to the calculated adjustment amount and priority.

[0031] After adjusting the process parameters, the system needs a certain amount of time to respond to these changes and reach a new stable state. This time includes the system response time (i.e. the time from the start of parameter adjustment to the start of system response) and the stabilization time (i.e. the time from the system response to reaching a stable state).

[0032] By considering the system response time and stabilization time after each parameter adjustment at each position, a preset time interval is calculated. After this time interval, the actual thickness after parameter adjustment is measured again.

[0033] After a preset time interval, new actual thickness data is obtained and compared with the target thickness. If there is still a difference, the thickness adjustment amount is calculated again based on the new actual thickness data, and the above process parameter adjustment process is repeated.

[0034] This feedback mechanism ensures that the system can continuously approach the target thickness, improving product quality and production efficiency.

[0035] In summary, the control method of the thickness of the wedge-shaped PVB interlayer realizes the precise control of the thickness of the PVB film by measuring the actual thickness in real time, comparing it with the target thickness, calculating the adjustment amount, adjusting the process parameters, and feeding back the adjustment results. This method improves the flexibility and accuracy of the production process and helps meet the market demand for high-quality PVB products.

[0036] In some embodiments, the method of obtaining the target thickness of each position of the PVB film under different classifications through historical data includes:

[0037] Obtain historical data of the thickness of the PVB film, wherein the historical data includes thickness measurement results at various positions under different product models, material ratios, and mold models;

[0038] Classify the historical data according to product model, material ratio and mold model; obtain the average thickness measurement value of each position under each classification;

[0039] Obtaining thickness design standards for each product model; the design standards include thickness ranges for each position; obtaining median values ​​for each position of the product model based on the thickness ranges;

[0040] If the difference between the mean thickness measurement at each position under the same category and the corresponding median is within the corresponding first difference range, and the mean thickness measurement at each position meets the preset wedge requirement; then each mean is used as the target thickness for each position under the category; if the difference between the mean thickness measurement at one or more positions under the same category and the corresponding median exceeds the corresponding first difference range; or the mean thickness measurement at each position does not meet the preset wedge requirement; then the mean of the position with the smallest difference from the median is selected as the target thickness for the position under the category, and the target thickness for other positions under the category is obtained based on the target thickness and wedge requirement of the position.

[0041] The working principle and effect of the above technical solution are as follows: First, collect historical data on the thickness of PVB film from production records or databases. These data should cover the thickness measurement results of various positions (such as edge, center, etc.) under different product models, material ratios and mold models. Classify the historical data according to product model, material ratio and mold model; the purpose is to group data under similar conditions for subsequent analysis.

[0042] For each classification, calculate the mean of the thickness measurement results at each position to reflect the average thickness at each position under the classification. At the same time, obtain the thickness design standards for each product model, which usually include the thickness range at each position and the desired wedge degree (i.e., the degree of gradual change in the thickness of the diaphragm from one end to the other). According to the thickness range in the design standard, calculate the median value at each position of each product model as the theoretical ideal thickness.

[0043] Compare the difference between the mean thickness measurement value and the corresponding median value at each position under the same category. If these differences are within the set first difference range and the mean thickness measurement value at each position meets the preset wedge requirement, these means are considered reasonable and can be directly used as the target thickness at each position under the category.

[0044] If the difference between the mean thickness measurement of one or more positions under a certain category and the corresponding median value exceeds the first difference range, or the mean thickness measurement of each position does not meet the wedge requirement, special processing is required. In this case, the mean of the position with the smallest difference from the median value is selected as the target thickness of the position under the category. According to the determined target thickness and the preset wedge requirement, the target thickness of other positions under the category is calculated through mathematical calculation or empirical formula. After determining the target thickness of each position under each category, these target thicknesses will be used to guide the subsequent production process to ensure that the thickness of the PVB film meets the design requirements. At the same time, these target thicknesses can also be used as quality control standards in the production process, used to monitor and adjust production parameters in real time to ensure the stability and consistency of product quality.

[0045] By analyzing historical data and calculating the average thickness measurement at each location under each category, the target thickness can be determined more accurately, thereby achieving higher thickness control accuracy during the production process; the thickness of PVB films in different batches, different product models and material ratios can be ensured to remain consistent, thereby improving product consistency and quality stability.

[0046] Taking into account the preset wedge shape requirements, this method can ensure that the PVB film presents the desired gradient in thickness to meet specific design requirements. By accurately controlling the film thickness, waste and defective rate in the production process can be reduced, and production efficiency can be improved.

[0047] In summary, the working principle of this method is to determine the target thickness of each position of the PVB film under different classifications by collecting and analyzing historical data, combining design standards and wedge requirements. This method not only takes into account the differences in actual production, but also ensures the controllability and consistency of product quality.

[0048] In some embodiments, the actual thickness of each position of the PVB film is obtained, and the thickness adjustment amount of each position is obtained by comparing the actual thickness with the target thickness; including:

[0049] Obtain the current classification of the PVB film; obtain the corresponding target thickness at each position according to the current classification;

[0050] The actual thickness of each position of the PVB film is obtained through a non-contact online scanning device;

[0051] Obtaining a thickness adjustment amount at each position through the actual thickness at each position and the target thickness;

[0052] The thickness adjustment is:

[0053]

[0054]

[0055] Among them, H adi is the current thickness adjustment value of the i-th position; H ti is the target thickness at the ith position; H Li is the thickness adjustment amount of the previous i-th position; H ai is the current actual thickness of the i-th position; n1 is the number of positions where the previous thickness adjustment is positive, n2 is the number of positions where the previous thickness adjustment is negative; n is the total number of positions. Lj is the adjustment amount of the jth previous thickness with a positive value; H Lk is the adjustment amount for the kth previous negative thickness; β is a constant in the range of (0,1); γ is a constant in the range of (-1,1) and γ≠0.

[0056] The working principle of the above technical solution is as follows: First, determine the current classification of the PVB film, which is usually based on factors such as product model, material ratio and mold model. According to the current classification, retrieve the target thickness of each position from the pre-set data table. Use a non-contact online scanning device to measure the thickness of each position of the PVB film in real time or regularly to obtain the current actual thickness.

[0057] Next, calculate the thickness adjustment amount at each position according to the formula.

[0058] The formula takes into account multiple factors, including the target thickness, the current actual thickness, and the previous thickness adjustment.

[0059] For each location, the difference between the target thickness and the actual thickness is calculated.

[0060] α is a dynamic adjustment factor that calculates a proportionality based on the difference between the target thickness and the actual thickness at the current location and the sum of the square root of the sum of the squares of this difference at all locations. This coefficient is used to adjust the thickness adjustment at the current location to reflect the relative importance of the thickness differences at different locations; add 1 to avoid the denominator being zero.

[0061] β and γ are constants used to further adjust the thickness adjustment. The range of β is (0,1), which can scale the current adjustment to a certain extent according to the positive or negative situation of the previous thickness adjustment. The range of γ is (-1,1) and γ≠0. It adjusts the adjustment of the current position according to the difference of the thickness adjustment of all positions in the previous time (that is, the difference between the sum of the positive adjustment and the negative adjustment) to balance the uniformity of the overall thickness distribution.

[0062] In the formula, n1 and n2 represent the number of positions where the previous thickness adjustment is positive and negative, respectively. They are used to calculate the denominator in the γ adjustment term, which is the sum of the differences in the thickness adjustment of all positions in the previous time. According to the calculated thickness adjustment, the relevant process parameters in the production process (such as extrusion temperature, extrusion amount, cooling speed, roller gap distance, etc.) are adjusted to change the thickness distribution of the PVB film to make it closer to the target thickness. After the adjustment is implemented, the actual thickness of the PVB film is measured again, and the above calculation and adjustment process is repeated according to the new actual thickness and target thickness until a satisfactory thickness distribution is achieved.

[0063] In summary, the working principle calculates the thickness adjustment amount of each position by comprehensively considering the target thickness, the current actual thickness, the previous adjustment amount and two adjustment constants, and adjusts the production process parameters accordingly to achieve precise control of the thickness of the PVB film.

[0064] The effect of the above technical solution is: by obtaining the actual thickness of each position of the PVB film in real time and comparing it with the target thickness, the thickness adjustment amount of each position can be accurately calculated, thereby achieving higher thickness control accuracy in the production process. With the help of a non-contact online scanning device, the thickness of the film can be measured quickly and continuously, reducing the time and error of manual measurement and improving production efficiency. By accurately controlling the thickness of each position of the film, it can be ensured that products in different batches and under different production conditions have consistent quality and performance. The scrap rate and rework rate caused by uneven thickness are reduced, and production costs are reduced. The process provides a data-based feedback mechanism that can continuously monitor and analyze production data so that production parameters can be adjusted and optimized in a timely manner.

[0065] The introduction of the dynamic adjustment factor α enables the adjustment amount to be dynamically adjusted according to the ratio of the thickness difference at the current position to the overall difference. This helps to apply a larger adjustment amount at locations with larger thickness differences, thereby reaching the target thickness faster. Li The formula takes into account the impact of the previous adjustment. This helps avoid over-adjustment or under-adjustment and makes the adjustment process smoother and more controllable.

[0066] The term takes into account the positive and negative differences in the thickness adjustment of all positions in the previous time, which helps to balance the impact of the positive and negative adjustments as a whole and ensure the uniform distribution of the diaphragm thickness; β and γ are constants that can be adjusted within a certain range to adapt to different production conditions and needs. This increases the flexibility and adaptability of the adjustment.

[0067] The formula takes into account multiple factors, including target thickness, actual thickness, previous adjustment amount, and the difference between positive and negative adjustment amounts, thereby reducing errors and uncertainties in the adjustment process.

[0068] In summary, by obtaining the actual thickness of each position of the PVB film and comparing it with the target thickness, and using the above formula to calculate the thickness adjustment amount at each position, production accuracy can be significantly improved, production efficiency can be optimized, product consistency can be enhanced, production costs can be reduced, and continuous improvement can be supported. At the same time, the formula itself can be adjusted dynamically, taking into account the impact of the previous adjustment amount, balancing the impact of positive and negative adjustments, improving adjustment flexibility, and reducing adjustment errors.

[0069] In some embodiments, obtaining the process parameter adjustment amount and adjusting the priority upgrade according to the thickness adjustment amount of each position includes:

[0070] Acquiring historical process data, wherein the historical production data includes real-time thickness, extrusion volume, roll gap distance, casting speed, and environmental parameters at different time points;

[0071] Preprocessing the historical production data and extracting features;

[0072] Establishing a machine learning model of process parameters and thickness adjustment amounts through the historical data; the input of the machine learning model includes the thickness adjustment amount of each position, each process parameter, environmental parameters, and feedback speed and stability of adjusting each process parameter; the output of the machine learning model includes the adjustment amount of each process parameter and the adjustment priority;

[0073] According to the thickness adjustment amount at the current position, based on the machine learning model, the adjustment amount and adjustment priority of each process parameter are obtained.

[0074] The working principle of the above technical solution is:

[0075] Collect historical process data from production records, including real-time thickness, extrusion volume, roll gap distance, casting speed, and environmental parameters (such as temperature, humidity, etc.) at different time points.

[0076] The collected historical data is cleaned to remove outliers and missing values.

[0077] The data were normalized to ensure comparability between different parameters.

[0078] Extract features related to the thickness adjustment amount, such as the trend and periodicity of thickness change.

[0079] From the preprocessed data, the features most relevant to the thickness adjustment amount are selected as the input of the machine learning model.

[0080] The historical data is used to train a machine learning model whose input includes the thickness adjustment amount at each location, each process parameter, environmental parameters, and the feedback speed and stability of adjusting each process parameter.

[0081] The output of the model is the adjustment amount and adjustment priority of each process parameter. Here, the process parameter adjustment amount refers to the specific process parameter value (such as extrusion amount, roll gap distance, casting speed, etc.) that needs to be adjusted in order to achieve the target thickness; the adjustment priority refers to which parameters should be adjusted first among different process parameters.

[0082] The model is verified through cross-validation and other methods to ensure its accuracy and stability.

[0083] The model is optimized based on the validation results to improve its prediction performance.

[0084] During the production process, the thickness adjustment amount at each position is obtained in real time.

[0085] Input real-time data into the trained machine learning model to obtain the adjustment amount and adjustment priority of each process parameter. According to the adjustment amount and priority output by the model, the production process is adjusted accordingly. For example, if the model predicts that the extrusion volume needs to be increased to reduce the thickness deviation, the output of the extruder is automatically adjusted; if the model recommends adjusting the roll gap distance first, the roll gap is adjusted first. After the adjustment is implemented, the thickness change is continuously monitored and the new data is fed back into the model to further optimize and adjust the model.

[0086] Through the above steps, the process parameter adjustment amount and adjustment priority can be intelligently determined according to the thickness adjustment amount of each position, thereby achieving precise control of the thickness of the PVB film, which not only improves production efficiency but also ensures the stability and consistency of product quality.

[0087] The effect of the above technical solution is: this method can intelligently calculate the required process parameter adjustment amount according to the thickness adjustment amount of each position of the PVB film in real time, thereby realizing precise control of the production process, greatly improving production efficiency, and ensuring the uniformity and consistency of product thickness. By determining the adjustment priority, it can guide operators or automated equipment to preferentially adjust the process parameters that have the greatest impact on thickness or the most significant adjustment effect, which not only improves the pertinence of the adjustment, but also optimizes the allocation of production resources and avoids unnecessary waste.

[0088] The machine learning model built on historical data can automatically learn and adapt to the thickness variation patterns under different production conditions. This means that even in the face of new production environments or material changes, process parameters can be adjusted quickly to ensure the stability of product quality.

[0089] In summary, the system achieves precise control of the PVB film production process by intelligently obtaining the process parameter adjustment amount and adjustment priority according to the thickness adjustment amount of each position, improves production efficiency, product quality and resource utilization, and enhances the adaptability and continuous improvement capabilities of the production process.

[0090] In some embodiments, the system response time and stabilization time after adjusting each parameter at each position are used to obtain a preset time interval; after the preset time interval, the actual thickness after the parameter adjustment is obtained, and whether to adjust again is determined according to the actual thickness, including:

[0091] Obtain a preset time interval through the system response time and stabilization time after adjustment of each parameter at each position; obtain the actual thickness after adjustment after the preset time interval;

[0092] The preset time interval is:

[0093] T=p×max(Tr,Tw)

[0094] Tr=max(Trij)

[0095] Tw=max(Twij)

[0096]

[0097] Where T is the preset time interval, p is the adjustment coefficient, ranging from [0.8, 1.2], Tr is the system response time, Tw is the system more stable time; max() is the maximum value; Trij is the system response time of changing the jth process parameter at the i-th position; Twij is the system stable time of adjusting the jth process parameter at the i-th position; H adi is the thickness adjustment required for the i-th position; H adais the mean value of the historical thickness adjustment amount under the same classification; Trija is the mean response time for adjusting the j-th process parameter at the i-th position under this classification; Ned is the current network congestion coefficient, where 0 < Ned < 1; Neda is the mean value of the network congestion coefficient; Twija is the mean system stabilization time for adjusting the j-th process parameter at the i-th position under this classification; f1() and f2() are both function expressions;

[0098] f1(H adi ,H ada ) is the function expression of the first correspondence relationship between H adi and H ada , and f1(H adi ,H ada ) is the function expression of the second correspondence relationship between H adi and H ada . For example, where a1, b1, c1 are constants obtained from historical data; where a2, b2, c2 are constants obtained from historical data;

[0099] If the difference between the actual thickness and the target thickness at a certain position is within the second difference range and meets the wedge angle requirement, there is no need to adjust the process parameters at this position again, and the thickness at this position is continuously monitored;

[0100] If the difference between the actual thickness and the target thickness at a certain position exceeds the second difference range, or although it is within the range but does not meet the wedge angle requirement, the process parameters at this position and adjacent positions are adjusted according to the process parameter adjustment amount, adjustment priority, and wedge angle requirement.

[0101] The working principle of the above technical solution is as follows:

[0102] For each position and each process parameter, the system calculates the system response time and system stabilization time based on historical data and current network conditions

[0103] The response time is obtained by the function f1(H adi ,H ada ) and the adjusted mean response time, considering the influence of the current network congestion coefficient (Ned); the function can have multiple expressions. In this embodiment, the function expression adopts a quadratic function form, which can better fit the complex relationship between H adi and H ada ; by adjusting the coefficients a1, b1, c1 of the quadratic term, linear term, and constant term, the function can have different rates of change in different intervals, so as to more accurately reflect the actual situation. The function expression adopts a ratio form, which can reflect the relationship between H adi and H adaThis relative relationship helps to eliminate the difference in thickness adjustment between different batches and under different conditions, and improve the accuracy and stability of system response time.

[0104] The stabilization time is given by the function f2(H adi ,H ada ) and the adjusted mean stabilization time, also taking into account the impact of network congestion coefficient.

[0105] Use the adjustment coefficient p in the range of [0.8,1.2] multiplied by the larger value of the system response time and the stabilization time to obtain the preset time interval T.

[0106] After a preset time interval T, the actual thickness of each position is measured. For each position, the difference between the actual thickness and the target thickness is compared to see whether it is within a second difference range, and whether the wedge degree requirement is met.

[0107] If the difference between the actual thickness and the target thickness at a certain position is within the second difference range and meets the wedge requirement, the system considers that the thickness at this position is close enough to the target and there is no need to adjust the process parameters again, but the thickness change at this position will be continuously monitored.

[0108] If the difference between the actual thickness and the target thickness at a certain position exceeds the second difference range, or is within the range but does not meet the wedge requirement, the system needs to adjust the process parameters of the position and adjacent positions based on the process parameter adjustment amount, adjustment priority and wedge requirement.

[0109] When adjusting, the feedback from previous adjustments, the current network conditions, and the thickness changes at adjacent locations are comprehensively considered to ensure that the adjusted thickness not only meets the requirements of a single location, but also maintains the overall wedge shape. Continuously monitor the thickness changes at each location to ensure that the adjusted thickness remains within the target range and meets the wedge shape requirements. If it is detected that the thickness at any location deviates from the target or the wedge shape does not meet the requirements, repeat the above adjustment process until the thickness at all locations meets the requirements. Through this closed-loop feedback adjustment mechanism, precise control of the thickness and wedge shape during the production process of PVB diaphragms can be achieved, thereby improving product quality and production efficiency.

[0110] The effects of the above technical solution are as follows: the method of this embodiment can intelligently calculate the preset time interval according to the actual thickness adjustment requirements of each position, and ensure that the actual thickness after adjustment is obtained at the right time; while reducing erroneous judgments, it can also reduce the amount of collected data, and at the same time, it can timely obtain thickness information and make adjustments according to the thickness information; reduce product failure rate; through accurate comparison and judgment, it can avoid unnecessary frequent adjustments, and at the same time, it can timely respond to thickness deviations to ensure the efficiency and accuracy of adjustments. When adjusting process parameters, the process parameter adjustment amount, adjustment priority and wedge requirements will be comprehensively considered to achieve optimal allocation of production resources, which not only reduces resource waste, but also improves production efficiency, making the entire production process more economical and efficient. Through continuous monitoring and adjustment, it can ensure that the thickness of each position is always kept within the target range and meets the wedge requirements. When calculating the preset time interval, factors such as system response time and stabilization time, as well as network congestion coefficient are considered. This enables the system to maintain high stability and reliability in the face of different production environments and conditions, ensuring the continuity and stability of production.

[0111] In summary, this method achieves precise control and optimal adjustment of the PVB film production process through strategies such as accurate calculation of preset time intervals, intelligent judgment of adjustment needs, and optimization of resource allocation. It not only improves production efficiency, product quality, and resource utilization, but also enhances the stability and continuous improvement capabilities of the system.

[0112] The embodiment of the present application provides a control system for the thickness of a wedge-shaped PVB interlayer, the system comprising:

[0113] The target thickness determination module is used to obtain the target thickness of each position of the PVB film under different classifications through historical data;

[0114] An adjustment amount determination module is used to obtain the actual thickness of each position of the PVB film, and obtain the thickness adjustment amount of each position through the actual thickness and the target thickness;

[0115] An adjustment module, for obtaining a process parameter adjustment amount and an adjustment priority upgrade according to a thickness adjustment amount at each position; adjusting the process parameters according to the process parameter adjustment amount and the adjustment priority; the process parameters include extrusion temperature, extrusion amount, cooling speed and roll gap distance;

[0116] The feedback module is used to obtain a preset time interval through the system response time and stabilization time after each parameter at each position is adjusted; after the preset time interval, the actual thickness after the parameter adjustment is obtained, and whether to adjust again is determined according to the actual thickness.

[0117] In some embodiments, the target thickness determination module includes:

[0118] A first acquisition unit is used to acquire historical data of the thickness of the PVB film, wherein the historical data includes thickness measurement results of various positions under different product models, material ratios and mold models;

[0119] The second acquisition unit is used to classify the historical data according to product model, material ratio and mold model; and obtain the average thickness measurement value of each position under each classification;

[0120] A third acquisition unit is configured to acquire a thickness design standard for each product model; the design standard includes a thickness range for each position; and based on the thickness range, a median value for each position of the product model is acquired;

[0121] A first determination unit, if the difference between the mean thickness measurement value and the corresponding median value at each position under the same category is within the corresponding first difference range, and the mean thickness measurement value at each position meets the preset wedge requirement; then each mean value is used as the target thickness at each position under the category; a second determination unit, if the difference between the mean thickness measurement value and the corresponding median value at one or more positions under the same category exceeds the corresponding first difference range; or the mean thickness measurement value at each position does not meet the preset wedge requirement; then the mean value of the position with the smallest difference from the median value is selected as the target thickness at the position under the category, and the target thickness of other positions under the category is obtained according to the target thickness and wedge requirement of the position.

[0122] In some embodiments, the adjustment amount determination module includes:

[0123] The target thickness acquisition unit is used to acquire the current classification of the PVB film; and obtain the corresponding target thickness of each position according to the current classification;

[0124] An actual thickness acquisition unit is used to obtain the actual thickness of each position of the PVB film through a non-contact online scanning device;

[0125] An adjustment amount acquisition unit, used to obtain a thickness adjustment amount at each position according to the actual thickness and target thickness at each position;

[0126] The thickness adjustment is:

[0127]

[0128] Among them, H adi is the current thickness adjustment value of the i-th position; H ti is the target thickness at the ith position; H Li is the thickness adjustment amount of the previous i-th position; H aiis the current actual thickness of the i-th position; n1 is the number of positions where the previous thickness adjustment is positive, n2 is the number of positions where the previous thickness adjustment is negative; n is the total number of positions. Lj is the adjustment amount of the jth previous thickness with a positive value; H Lk is the adjustment amount for the kth previous negative thickness; β is a constant in the range of (0,1); γ is a constant in the range of (-1,1) and γ≠0.

[0129] In some embodiments, the adjustment module includes:

[0130] A fourth acquisition unit is used to acquire historical process data, wherein the historical production data includes real-time thickness, extrusion volume, roll gap distance, casting speed and environmental parameters at different time points;

[0131] A feature extraction unit, used for preprocessing the historical production data and extracting features;

[0132] A model building unit, used to build a machine learning model of process parameters and thickness adjustment amounts through the historical data; the input of the machine learning model includes the thickness adjustment amount of each position, each process parameter, environmental parameters, and feedback speed and stability of adjusting each process parameter; the output of the machine learning model includes the adjustment amount of each process parameter and the adjustment priority;

[0133] The adjustment amount acquisition unit is used to obtain the adjustment amount and adjustment priority of each process parameter based on the thickness adjustment amount at the current position and based on the machine learning model.

[0134] In some embodiments, the feedback module includes:

[0135] A fifth acquisition unit is used to obtain a preset time interval through the system response time and the stabilization time after the adjustment of each parameter at each position; and obtain the adjusted actual thickness after the preset time interval;

[0136] The preset time interval is:

[0137] T=p×max(Tr,Tw)

[0138] Tr=max(Trij)

[0139] Tw=max(Twij)

[0140]

[0141] Wherein, T is a preset time interval, p is an adjustment coefficient with a range of [0.8, 1.2], Tr is the system response time, Tw is the system more stable time; max() is to take the maximum value; Trij is the system response time for the i-th position to change the j-th process parameter; Twij is the system stable time for the i-th position to adjust the j-th process parameter; H adi is the thickness adjustment amount required for the i-th position; H ada is the average value of the historical thickness adjustment amounts under the same classification; Trija is the average response time for the i-th position to adjust the j-th process parameter under this classification; Ned is the current network congestion coefficient; 0 < Ned < 1; Neda is the average value of the network congestion coefficients; Twija is the average system stable time for the i-th position to adjust the j-th process parameter under this classification; f1() and f2() are both function expressions;

[0142] f1(H adi , H ada ) is the first correspondence function expression between H adi and H ada , f1(H adi , H ada ) is the second correspondence function expression between H adi and H ada , for example, wherein, a1, b1, c1 are constants obtained through historical data; wherein, a2, b2, c2 are constants obtained through historical data;

[0143] The third determination unit is used to, if the difference between the actual thickness and the target thickness at a certain position is within the second difference range and the wedge degree requirement is met, then there is no need to adjust the process parameters of this position again, and continuously monitor the thickness of this position;

[0144] The fourth determination unit is used to, if the difference between the actual thickness and the target thickness at a certain position exceeds the second difference range, or although it is within the range but does not meet the wedge degree requirement, then adjust the process parameters of this position and its adjacent positions according to the process parameter adjustment amount, adjustment priority and wedge degree requirement.

[0145] The principle and effect of the above technical solution are the same as those in the method embodiment of the present application, and will not be elaborated here.

[0146] The present application also provides an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of any method described in the present application are implemented.

[0147] This application is explained from the perspectives of purpose of use, effectiveness, progress and novelty, and has met the functional enhancement and usage requirements emphasized by the Patent Law. The above description and drawings of this application are only the preferred embodiments of this application, and are not intended to limit this application. Therefore, all structures, devices, features, etc. that are similar or identical to this application, that is, all equivalent replacements or modifications made in accordance with the scope of the patent application of this application, should fall within the scope of protection of the patent application of this application.

Claims

1. A method for controlling the thickness of a wedge-shaped PVB interlayer, characterized in that: The method comprises: S1. Obtain the target thickness of each position of the PVB film under different categories through historical production data; including: S11, obtaining historical production data of PVB film thickness; S12, classifying the historical production data according to product model, material ratio and mold model; obtaining the average thickness measurement value of each position under each classification; S13, obtaining the thickness design standard of each product model; the thickness design standard includes the thickness range of each position; according to the thickness range, obtaining the median value of each position of the product model; S14. If the difference between the thickness measurement mean value and the corresponding median value at each position under the same category is within the corresponding first difference range, and the thickness measurement mean value at each position meets the preset wedge degree requirement, each mean value is used as the target thickness at each position under the category; S15. If the difference between the thickness measurement mean value and the corresponding median value of one or more positions under the same category exceeds the corresponding first difference range; or the thickness measurement mean value of each position does not meet the preset wedge degree requirement, the mean value of the position with the smallest difference from the median value is selected as the target thickness of the position under the category, and the target thickness of other positions under the category is obtained according to the target thickness and wedge degree requirement of the position; S2, obtaining the actual thickness of each position of the PVB film, and obtaining the thickness adjustment amount of each position through the actual thickness and the target thickness; including: S21, obtaining the current classification of the PVB film; obtaining the target thickness of each corresponding position according to the current classification; S22, obtaining the actual thickness of each position of the PVB film by a non-contact online scanning device; S23, obtaining a thickness adjustment amount at each position according to the actual thickness at each position and the target thickness; The thickness adjustment is: in, is the current thickness adjustment amount of the i-th position; is the target thickness at the i-th position; is the thickness adjustment amount of the previous i-th position; is the current actual thickness at the i-th position; is the number of positions where the previous thickness adjustment amount is positive, is the number of positions where the previous thickness adjustment is negative; n is the total number of positions; is the adjustment amount of the jth previous thickness with a positive value; is the adjustment amount of the kth previous thickness with a negative value; is a constant, ranging from (0,1); is a constant in the range (-1,1) and ≠0; S3. Obtaining a process parameter adjustment amount and an adjustment priority according to the thickness adjustment amount at each position; adjusting the process parameters according to the process parameter adjustment amount and the adjustment priority; the process parameters include extrusion temperature, extrusion amount, cooling speed and roll gap distance; including: S31, acquiring historical production data, wherein the historical production data includes real-time thickness, extrusion volume, roller gap distance, casting speed and environmental parameters at different time points; S32, preprocessing the historical production data and extracting features; S33, establishing a machine learning model of process parameters and thickness adjustment amounts through the historical production data; the input of the machine learning model includes the thickness adjustment amount of each position, each process parameter, environmental parameters, and the feedback speed and stability of adjusting each process parameter; the output of the machine learning model includes the adjustment amount of each process parameter and the adjustment priority; S34, according to the current thickness adjustment amount of each position, based on the machine learning model, obtaining the adjustment amount and adjustment priority of each process parameter; S4, obtaining a preset time interval through the system response time and stabilization time after the adjustment of each parameter at each position; obtaining the actual thickness after the parameter adjustment after the preset time interval, and determining whether to adjust again according to the actual thickness; including: S41, obtaining a preset time interval through the system response time and stabilization time after adjustment of each parameter at each position; and obtaining the adjusted actual thickness after the preset time interval; The preset time interval is: Where T is the preset time interval, p is the adjustment coefficient, and the range is [0.8, 1.2]. is the system response time, is the system stabilization time; max() is the maximum value; The system response time for changing the jth process parameter at the ith position; System stabilization time for adjusting the jth process parameter for the i-th position; is the thickness adjustment required for the i-th position; It is the mean of historical thickness adjustment under the same classification; The mean value of the response time of adjusting the jth process parameter for the i-th position under this classification; is the current network congestion coefficient; 0< <1; is the mean value of the network congestion coefficient; is the mean value of the system stabilization time for adjusting the jth process parameter at the ith position under this classification; f1() and f2() are both function expressions; S42, if the difference between the actual thickness and the target thickness at a certain position is within the second difference range and meets the wedge degree requirement, there is no need to adjust the process parameters at the position again, and the thickness at the position is continuously monitored; S43. If the difference between the actual thickness and the target thickness at a certain position exceeds the second difference range, or is within the range but does not meet the wedge requirement, the process parameters of the position and adjacent positions are adjusted according to the process parameter adjustment amount, adjustment priority and wedge requirement.

2. The control system of the thickness of the wedge-shaped PVB interlayer is characterized by: The system comprises: The target thickness determination module is used to obtain the target thickness of each position of the PVB film under different categories through historical production data; including: A first acquisition unit is used to acquire historical production data of the thickness of the PVB film; The second acquisition unit is used to classify the historical production data according to product model, material ratio and mold model; and obtain the average thickness measurement value of each position under each classification; A third acquisition unit is configured to acquire a thickness design standard for each product model; the design standard includes a thickness range for each position; and based on the thickness range, a median value for each position of the product model is acquired; A first determination unit, if the difference between the thickness measurement mean value and the corresponding median value of each position under the same category is within the corresponding first difference range, and the thickness measurement mean value of each position meets the preset wedge degree requirement, then each mean value is used as the target thickness of each position under the category; The second determination unit selects the mean value of the position with the smallest difference from the median value as the target thickness of the position in the category, and obtains the target thickness of other positions in the category according to the target thickness of the position and the wedge degree requirement, if the difference between the mean value of the thickness measurement of one or more positions in the same category and the corresponding median value exceeds the corresponding first difference range; or the mean value of the thickness measurement of each position does not meet the preset wedge degree requirement. The adjustment amount determination module is used to obtain the actual thickness of each position of the PVB film, and obtain the thickness adjustment amount of each position through the actual thickness and the target thickness; including: The target thickness acquisition unit is used to acquire the current classification of the PVB film; and obtain the corresponding target thickness of each position according to the current classification; An actual thickness acquisition unit is used to obtain the actual thickness of each position of the PVB film through a non-contact online scanning device; An adjustment amount acquisition unit, used to obtain a thickness adjustment amount at each position according to the actual thickness and target thickness at each position; The thickness adjustment is: in, is the current thickness adjustment amount of the i-th position; is the target thickness at the i-th position; is the thickness adjustment amount of the previous i-th position; is the current actual thickness at the i-th position; is the number of positions where the previous thickness adjustment amount is positive, is the number of positions where the previous thickness adjustment is negative; n is the total number of positions; is the adjustment amount of the jth previous thickness with a positive value; is the adjustment amount for the kth previous thickness with a negative value; is a constant, ranging from (0,1); is a constant in the range (-1,1) and ≠0; An adjustment module is used to obtain a process parameter adjustment amount and an adjustment priority according to the thickness adjustment amount of each position; and to adjust the process parameters according to the process parameter adjustment amount and the adjustment priority; the process parameters include extrusion temperature, extrusion amount, cooling speed and roll gap distance; including: A fourth acquisition unit is used to acquire historical production data, wherein the historical production data includes real-time thickness, extrusion volume, roller gap distance, casting speed and environmental parameters at different time points; A feature extraction unit, used for preprocessing the historical production data and extracting features; A model building unit, used to build a machine learning model of process parameters and thickness adjustment amounts through the historical production data; the input of the machine learning model includes the thickness adjustment amount of each position, each process parameter, environmental parameters, and feedback speed and stability of adjusting each process parameter; the output of the machine learning model includes the adjustment amount of each process parameter and the adjustment priority; An adjustment amount acquisition unit, configured to obtain the adjustment amount and adjustment priority of each process parameter according to the current thickness adjustment amount at each position based on the machine learning model; The feedback module is used to obtain a preset time interval through the system response time and stabilization time after the adjustment of each parameter at each position; after the preset time interval, the actual thickness after the parameter adjustment is obtained, and whether to adjust again according to the actual thickness; including: A fifth acquisition unit is used to obtain a preset time interval through the system response time and the stabilization time after the adjustment of each parameter at each position; and obtain the adjusted actual thickness after the preset time interval; The preset time interval is: Where T is the preset time interval, p is the adjustment coefficient, and the range is [0.8, 1.2]. is the system response time, is the system stabilization time; max() is the maximum value; The system response time for changing the jth process parameter at the ith position; System stabilization time for adjusting the jth process parameter for the i-th position; is the thickness adjustment required for the i-th position; It is the mean of historical thickness adjustment under the same classification; The mean value of the response time of adjusting the jth process parameter for the i-th position under this classification; is the current network congestion coefficient; 0< <1; is the mean value of the network congestion coefficient; is the mean value of the system stabilization time for adjusting the jth process parameter at the ith position under this classification; f1() and f2() are both function expressions; A third determination unit is used for, if the difference between the actual thickness and the target thickness at a certain position is within a second difference range and meets the wedge degree requirement, there is no need to adjust the process parameters of the position again, and the thickness of the position is continuously monitored; The fourth determination unit is used to adjust the process parameters of a certain position and adjacent positions according to the process parameter adjustment amount, adjustment priority and wedge requirement if the difference between the actual thickness and the target thickness at a certain position exceeds the second difference range, or is within the range but does not meet the wedge requirement.

Images