A method for preparing a drawing film composite packaging material and an automatic preparation device thereof

Abstract

The invention relates to the technical field of material preparation, and discloses a method for preparing a brushed film composite packaging material and an automatic preparation device thereof, wherein a method for preparing a brushed film composite packaging material comprises the following steps: selecting a polymer monomer, dissolving the monomer in a suitable solvent, heating to a suitable temperature, and using a stirring device to stir and mix, so that the monomer undergoes a polymerization reaction to form a polymer solution; dehydrating the polymer solution to remove a small amount of low-molecular impurities in the solution to obtain a relatively pure polymer; drying the polymer to a suitable moisture content to solidify it into solid particles; subjecting the dried polymer particles to a film-forming process to make a soft and tough film; a high-purity polymer product can be obtained, and a small amount of low-molecular impurities in the solution can be removed by stirring and mixing with a stirrer, and various types of polymers can be conveniently prepared by selecting different polymer monomers.

Description

Technical Field

[0001] The present invention relates to the technical field of material preparation, and in particular to a method for preparing a wire drawing film composite packaging material and an automatic preparation device thereof. Background Art

[0002] "Drawn film" refers to a film formed by mixing plastic particles or fibers with various additives through a specific manufacturing process, and then heating, pressurizing or adding a catalyst. This film usually has good flexibility and tensile resistance, and can be used to make various packaging materials, such as drawn bags, hand bags, boxes and roll paper. The main advantages of drawn film are high transparency, good toughness, and good softness. It can withstand a certain degree of stretching and tearing without breaking. In addition, the drawn film also has good printing properties, and various patterns and texts can be printed according to needs to improve the appearance and added value of the product. Therefore, drawn film is widely used in the packaging industry, especially in food packaging and pharmaceutical packaging.

[0003] When preparing the drawing film composite packaging material, it is necessary to mix a variety of raw materials to obtain a polymer solution, then dehydrate and dry the polymer solution, and finally make the dried polymer into a film before cutting and molding. In the process of mixing the raw materials, usually a variety of materials with good proportions are put into the solvent at one time for mixing. In the mixing process, if too much raw material is poured in, it will not have enough time to react with the solvent, resulting in insufficient reaction. In addition, when stirring, it is usually stirred manually and cannot be stirred for a long time, resulting in insufficient stirring and mixing, which leads to insufficient mixing of the polymer solution and poor molding quality of the packaging material. Summary of the invention

[0004] In view of the problem in the prior art that when polymers are mixed, the stirring and mixing is insufficient and the stirring efficiency is low, resulting in the problem of poor material molding quality, a method for preparing a drawing film composite packaging material and an automatic preparation device thereof are proposed.

[0005] The purpose is to improve the efficiency of raw material mixing and stirring, so that each raw material can be fully and evenly mixed with the solvent to improve the material preparation efficiency.

[0006] The technical solution of the present invention is a method for preparing a drawing film composite packaging material, comprising the following steps:

[0007] Select polymer monomers, dissolve the monomers in a suitable solvent, heat to a suitable temperature, and stir and mix using a stirring device to allow the monomers to undergo a polymerization reaction to form a polymer solution;

[0008] The polymer solution is dehydrated to remove a small amount of low molecular weight impurities in the solution to obtain a relatively pure polymer;

[0009] Dry the polymer to an appropriate moisture content so that it solidifies into solid particles;

[0010] The dried polymer particles are subjected to film-forming treatment to form soft and tough films;

[0011] As needed, the film is cut, formed, and processed to produce the desired packaging material.

[0012] By adopting the above technical scheme, a high-purity polymer product can be obtained. By stirring and mixing with an agitator, a small amount of low-molecular impurities in the solution can be removed. The conditions of the polymerization reaction, such as temperature and stirring speed, can be controlled to ensure that the desired type of polymer product is obtained. By selecting different polymer monomers, various types of polymers can be conveniently prepared.

[0013] Furthermore, the polymer monomer may be vinyl acetate, polyvinyl chloride, methyl methacrylate and acetyl methyl methacrylate.

[0014] By adopting the above technical scheme, vinyl acetate has good flexibility and wear resistance, and polyvinyl chloride has excellent thermal stability and chemical corrosion resistance, thereby improving the quality of the packaging material after preparation.

[0015] Furthermore, the polymer can be dried by vacuum freezing and spray drying.

[0016] By adopting the above technical scheme, the polymer is dried by vacuum freezing and spray drying, which can maintain the integrity of the polymer molecular chain, avoid degradation or damage caused by wet treatment, reduce the loss of polymer during the drying process, improve product purity, reduce energy consumption, reduce environmental pollution, improve production efficiency, and shorten the production cycle.

[0017] Furthermore, the film forming process can be achieved by casting, calendering and the like.

[0018] By adopting the above technical solution, a film-like product can be obtained by casting, calendering and other methods to achieve filmization, improve the processability of the product, make it easier to form, print and coat, improve the mechanical properties of the product, make it stronger, stretch-resistant, tear-resistant, etc., and improve the stability and durability of the product, reduce deformation, cracks and aging problems, reduce production costs and improve production efficiency.

[0019] Another object of the present invention is to provide a drawing film composite packaging material preparation device, the purpose of which is to improve the stirring and mixing effect of the raw materials, thereby improving the material molding quality.

[0020] To achieve the above-mentioned object, the present invention provides the following technical solutions: a drawing film composite packaging material preparation device, comprising a stirring barrel, and also comprising a stirring device arranged in the stirring barrel;

[0021] The stirring device comprises a stirring unit arranged in the stirring barrel, and a feeding unit arranged in a circular array on the top of the stirring barrel;

[0022] The stirring unit includes a rotating assembly arranged at the top of the stirring barrel, and a stirring assembly arranged at the bottom of the motor;

[0023] The rotating assembly includes a motor arranged at the top of the mixing barrel, a rotating plate arranged at the bottom of the motor output shaft, a rotating rod arranged at the bottom of the rotating plate, and the rotating rod is fixedly connected to the middle of the rotating plate, a rotating ball arranged at the bottom of the rotating rod, and a rotating base arranged at the bottom of the rotating ball, and the rotating base is arranged at the bottom of the inner cavity of the mixing barrel.

[0024] By adopting the above technical scheme, when stirring the polymer solution, multiple polymer monomers are poured into the feeding unit respectively, and the polymer solution in the stirring barrel is stirred by the stirring unit to make it fully react; when the stirring unit is working, the rotating component drives the stirring component to rotate for stirring, the motor is started, the motor drives the rotating plate to rotate, the rotation of the rotating plate drives the rotating rod to rotate, and the rotating rod drives the rotating ball at the bottom to rotate in the rotating base. The setting of the rotating base improves the stability of the rotation of the rotating rod. The rotating plate is in the shape of a staircase, so that the rotating rod can rotate around the central axis of the motor, so that the rotating rod drives the stirring component to rotate obliquely, which can better stir the solution.

[0025] Furthermore, the stirring assembly includes a stirring plate arranged in the middle of the rotating rod, and stirring blades arranged in a ring array on the stirring plate.

[0026] By adopting the above technical solution, when the rotating rod rotates, the stirring plate and the stirring blade rotate obliquely around the central axis of the stirring barrel. During the rotation process, the three stirring plates will rotate obliquely up and down respectively, so that the stirring plates will move up and down continuously. The area of ​​the stirring plate is larger than that of the stirring blade, so that the stirring plate will compress and push the upper and lower solutions when rotating, so that the solution in the stirring barrel will move downward, which can quickly allow the polymer monomers flowing on the top of the polymer solution to quickly enter the solution, thereby accelerating the mixing speed. The area of ​​the stirring blade is small, and the solution can pass through when rotating, and the solution is stirred and mixed in the horizontal position, thereby accelerating the fusion reaction of the polymer monomers entering the middle of the solution and improving the reaction rate.

[0027] Furthermore, the feeding unit includes a feeding assembly arranged at the top of the mixing barrel, and a feeding assembly arranged at the bottom of the feeding assembly;

[0028] The material discharge assembly comprises a material discharge barrel arranged on the top of the mixing barrel, a material discharge trough opened at the bottom of the material discharge barrel, and a material discharge hole opened at the bottom of the material discharge trough.

[0029] By adopting the above technical scheme, the discharge component is used to place the raw materials to be fused. The raw materials placed in the discharge barrel will flow to the lower trough under the action of the discharge component, and flow into the stirring barrel through the discharge hole to mix with the solvent.

[0030] Furthermore, the unloading component includes a unloading piece arranged at the bottom of the unloading trough, a unloading plate arranged at the top of the unloading piece, a unloading column arranged at the bottom of the unloading plate, and the bottom of the unloading column extends out of the bottom of the unloading hole, and an extrusion surface opened at the top of one side of the rotating plate, and the extrusion surface is slidably connected to the bottom of the unloading column.

[0031] By adopting the above technical solution, when the rotating plate rotates, the extrusion surface on one side of the rotating plate will squeeze the material discharge column, so that the material discharge column pulls the material discharge piece upward. The material discharge piece can be a spring, so that the material discharge plate extends out of the material discharge trough, so that the raw materials can flow down from the material discharge trough and the material discharge hole. The setting of the material discharge component of the annular array enables the rotating plate to intermittently release different raw materials when rotating, so that different raw materials flow down in sequence and in a quantitative manner for mixing, which is convenient for sufficient mixing reaction during stirring, and avoids too much at one time, resulting in uneven mixing, insufficient reaction, and reduced mixing and reaction efficiency.

[0032] Furthermore, a liquid inlet and a liquid outlet are respectively provided on both sides of the mixing barrel.

[0033] By adopting the above technical solution, the liquid inlet is used to input the solvent, and the liquid outlet is used to discharge the mixture after the reaction.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. High-purity polymer products can be obtained. A small amount of low-molecular impurities in the solution can be removed by stirring and mixing with an agitator. The conditions of the polymerization reaction, such as temperature and stirring speed, can be controlled to ensure that the desired type of polymer product is obtained. Various types of polymers can be easily prepared by selecting different polymer monomers.

[0036] 2. Film-forming can be achieved through casting, calendering and other methods to obtain film-like products, improve the processability of the product, make it easier to shape, print and coat, improve the mechanical properties of the product, make it stronger, stretch-resistant, tear-resistant, etc., and improve the stability and durability of the product, reduce deformation, cracks and aging problems, reduce production costs and improve production efficiency.

[0037] 3. When the rotating plate rotates, the extrusion surface on one side of the rotating plate will squeeze the material discharge column, so that the material discharge column pulls the material discharge piece upward. The material discharge piece can be a spring, so that the material discharge plate extends out of the material discharge trough, so that the raw materials can flow down from the material discharge trough and the material discharge hole. The setting of the material discharge component of the annular array enables the rotating plate to intermittently release different raw materials when rotating, so that different raw materials flow down quantitatively in sequence for mixing, which is convenient for sufficient mixing reaction during stirring, and avoids too much at one time, resulting in uneven mixing, insufficient reaction, and reduced mixing and reaction efficiency.

[0038] 4. By introducing the second-order sliding film control algorithm based on genetic algorithm optimization, the preparation method of drawing film composite packaging material significantly improves the reaction efficiency and product quality. The global search capability of the genetic algorithm optimizes the sliding film control parameters, so that the system reaches the optimal state under different reaction conditions. The intelligent control algorithm can respond to system state changes in real time, accurately adjust the heating temperature and stirring speed, ensure that the polymerization reaction is carried out under the optimal conditions, and significantly improve the reaction speed and efficiency. In addition, the optimization of the control strategy reduces energy consumption and improves the economic benefits of the system. At the same time, the intelligent optimization algorithm enhances the system's robustness to external disturbances and parameter changes, ensures the stability and reliability of the production process, realizes the high degree of automation and intelligence of the production process, reduces manual intervention, and improves production efficiency and operational safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the overall preparation process structure of the present invention;

[0040] Figure 2 It is a schematic diagram of the overall structure inside the stirring device of the present invention;

[0041] Figure 3 It is a schematic diagram of the overall structure of the rotating assembly and the stirring assembly of the present invention;

[0042] Figure 4 It is a schematic diagram of the positional relationship between the rotating plate and the discharge barrel of the present invention;

[0043] Figure 5 It is a schematic diagram of the overall structure of the discharge assembly and the blanking assembly of the present invention.

[0044] Figure 6 This is a flow chart of the integration and application of the intelligent optimization polymerization reaction control system of the present invention.

[0045] In the figure:

[0046] 1. Mixing barrel; 2. Rotating assembly; 21. Motor; 22. Rotating plate; 23. Rotating rod; 24. Rotating ball; 25. Rotating base; 3. Stirring assembly; 31. Stirring plate; 32. Stirring blade; 4. Discharging assembly; 41. Discharging barrel; 42. Discharging trough; 43. Discharging hole; 5. Discharging assembly; 51. Discharging part; 52. Discharging plate; 53. Discharging column; 54. Extrusion surface; 6. Liquid inlet; 7. Liquid outlet. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0048] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a method for preparing a brushed film composite packaging material, comprising the following steps: selecting a polymer monomer, dissolving the monomer in an appropriate solvent, heating to 50 degrees Celsius, heating the bottom of the stirring barrel 1, and using a stirring device to stir and mix, so that the monomer undergoes a polymerization reaction to form a polymer solution; dehydrating the polymer solution to remove a small amount of low-molecular impurities in the solution to obtain a relatively pure polymer; drying the polymer to an appropriate moisture content to solidify it into solid particles; subjecting the dried polymer particles to a film-forming process to make a soft and tough film; and cutting, molding, and other processes on the film as needed to make the desired packaging material.

[0049] Specifically, a high-purity polymer product can be obtained, and a small amount of low-molecular impurities in the solution can be removed by stirring and mixing with an agitator. The conditions of the polymerization reaction, such as temperature and stirring speed, can be controlled to ensure that the desired type of polymer product is obtained. Various types of polymers can be conveniently prepared by selecting different polymer monomers.

[0050] Reference Figure 1 , the polymer monomers can be vinyl acetate, polyvinyl chloride, methyl methacrylate and acetyl methyl methacrylate.

[0051] Specifically, vinyl acetate has good flexibility and wear resistance, and polyvinyl chloride has excellent thermal stability and chemical corrosion resistance, which improves the quality of the packaging material after preparation.

[0052] Reference Figure 1-Figure 6 , the polymer can be dried by vacuum freezing and spray drying.

[0053] Specifically, the polymer is dried by vacuum freezing and spray drying, which can maintain the integrity of the polymer molecular chain, avoid degradation or damage caused by wet treatment, reduce polymer loss during the drying process, improve product purity, reduce energy consumption, reduce environmental pollution, improve production efficiency, and shorten the production cycle.

[0054] Reference Figure 1 , film-forming treatment can be achieved through casting, calendering and other methods.

[0055] Specifically, by film-forming through methods such as casting and calendering, a film-like product can be obtained, which improves the processability of the product, makes it easier to form, print and coat, and can improve the mechanical properties of the product, making it stronger, more resistant to stretching and tearing, etc., as well as improve the stability and durability of the product, reduce deformation, cracking and aging problems, reduce production costs and improve production efficiency.

[0056] Example 2, reference Figure 1-Figure 3 , which is the second embodiment of the present invention, provides: a drawing film composite packaging material preparation device, including a stirring barrel 1, and also including a stirring device installed in the stirring barrel 1; the stirring device includes a stirring unit installed in the stirring barrel 1, and a feeding unit installed in a ring array on the top of the stirring barrel 1; the stirring unit includes a rotating component 2 installed on the top of the stirring barrel 1, and a stirring component 3 installed at the bottom of the motor 21; the rotating component 2 includes a motor 21 installed on the top of the stirring barrel 1, a rotating plate 22 fixedly connected to the bottom of the output shaft of the motor 21, a rotating rod 23 fixedly connected to the bottom of the rotating plate 22, and the rotating rod 23 is fixedly connected to the middle of the rotating plate 22, a rotating ball 24 fixedly connected to the bottom of the rotating rod 23, and a rotating base 25 rotatably connected to the bottom of the rotating ball 24, and the rotating base 25 is arranged at the bottom of the inner cavity of the stirring barrel 1.

[0057] Specifically, when stirring the polymer solution, multiple polymer monomers are poured into the feeding unit respectively, and the polymer solution in the stirring barrel 1 is stirred by the stirring unit to make it fully react; when the stirring unit is working, the rotating component 2 drives the stirring component 3 to rotate for stirring, and the motor 21 is started. The motor 21 drives the rotating plate 22 to rotate, and the rotation of the rotating plate 22 drives the rotating rod to rotate. The rotating rod 23 drives the rotating ball 24 at the bottom to rotate in the rotating base 25. The setting of the rotating base 25 improves the stability of the rotation of the rotating rod 23. The rotating plate 22 is in a staircase shape, so that the rotating rod 23 can rotate around the central axis of the motor 21, so that the rotating rod 23 drives the stirring component 3 to rotate obliquely, which can better stir the solution.

[0058] Reference Figure 1-Figure 3The stirring assembly 3 includes a stirring plate 31 fixedly connected to the middle of the rotating rod 23, and stirring blades 32 fixedly connected to the stirring plate 31 in an annular array.

[0059] Specifically, when the rotating rod 23 rotates, the stirring plate 31 and the stirring blade 32 rotate obliquely around the central axis of the stirring barrel 1. During the rotation process, the three stirring plates 31 will rotate obliquely up and down respectively, so that the stirring plates 31 will continuously move up and down. The area of ​​the stirring plate 31 is larger than that of the stirring blade 32, so that the stirring plate 31 will compress and push the upper and lower solutions when rotating, so that the solution in the stirring barrel 1 will move downward, which can quickly allow the polymer monomers flowing on the top of the polymer solution to quickly enter the solution, thereby accelerating the mixing speed. The stirring blade 32 has a small area, and can allow the solution to pass through when rotating, and stir and mix the solution in the horizontal position, thereby accelerating the fusion reaction of the polymer monomers entering the middle of the solution and improving the reaction rate. The rest of the structure is the same as that of Example 1.

[0060] Example 3, reference Figure 1-Figure 5 , which is the third embodiment of the present invention. This embodiment is different from the second embodiment in that: the feeding unit includes a discharge component 4 installed on the top of the mixing barrel 1, and a discharge component 5 installed at the bottom of the discharge component 4; the discharge component 4 includes a discharge barrel 41 fixedly connected to the top of the mixing barrel 1, a discharge trough 42 opened at the bottom of the discharge barrel 41, and a discharge hole 43 opened at the bottom of the discharge trough 42.

[0061] Specifically, the discharge component 4 is used to place the raw materials to be fused. The raw materials placed in the discharge barrel 41 will flow to the lower trough 42 under the action of the discharge component 5, and flow into the mixing barrel 1 through the discharge hole 43 to be mixed with the solvent.

[0062] Reference Figure 1-Figure 5 The unloading assembly 5 includes a unloading piece 51 fixedly connected to the bottom of the unloading trough 42, a unloading plate 52 fixedly connected to the top of the unloading piece 51, a unloading column 53 fixedly connected to the bottom of the unloading plate 52, and the bottom of the unloading column 53 extends out of the bottom of the unloading hole 43, and an extrusion surface 54 opened at the top of one side of the rotating plate 22, and the extrusion surface 54 is slidably connected to the bottom of the unloading column 53.

[0063] Specifically, when the rotating plate 22 rotates, the extrusion surface 54 on one side of the rotating plate 22 will squeeze the material discharge column 53, so that the material discharge column 53 pulls the material discharge piece 51 upward. The material discharge piece 51 can be a spring, so that the material discharge plate 52 extends out of the material discharge trough 42, so that the raw materials can flow down from the material discharge trough 42 and the material discharge hole 43. The setting of the annular array material discharge component 5 enables the rotating plate 22 to intermittently release different raw materials when rotating, so that different raw materials flow down in sequence and are mixed in a quantitative manner, which is convenient for sufficient mixing reaction during stirring, and avoids too much at one time resulting in uneven mixing, insufficient reaction, and reduced mixing and reaction efficiency.

[0064] Reference Figure 2 A liquid inlet 6 and a liquid outlet 7 are respectively provided on both sides of the mixing barrel 1.

[0065] Specifically, the liquid inlet 6 is used to input the solvent, and the liquid outlet 7 is used to discharge the mixture after the reaction. The rest of the structure is the same as that of Example 2.

[0066] like Figure 6 In the above-mentioned steps of selecting and dissolving the polymer monomers, intelligent optimization algorithms and deep learning algorithms are introduced to monitor and adjust the solution temperature and stirring speed in real time to optimize the polymerization reaction efficiency; the specific implementation process can be summarized as follows:

[0067] Step A: Sensor deployment and data collection:

[0068] Objective: Deploy sensors to collect solution temperature, viscosity, and color data, and ensure that data is accurately and in real time fed back to the control system;

[0069] 1. Sensor selection

[0070] Temperature sensor: Use RTD (resistance temperature detector) sensor, which has high accuracy and fast response characteristics; the calculation formula after acquisition is as follows:

[0071]

[0072] Where T is the temperature, R is the resistance of the temperature sensor, and R 0 is the resistance of the sensor at 0°C, and α represents the temperature coefficient of the sensor;

[0073] Viscosity sensor: Using a rotational viscometer, this sensor can monitor the viscosity changes of the solution in real time; the viscosity is calculated as:

[0074] η=K·(f 0 -f)

[0075] Where η is the calculated viscosity value, f represents the current vibration frequency, and f 0is the vibration frequency of the sensor under no-load condition, K is the calibration factor of the sensor;

[0076] Color sensor: using a spectrometer and a CMOS image sensor. The spectrometer can capture the color changes of the solution, while the CMOS image sensor can take images of the solution for subsequent image analysis. For the color sensor, its output is an RGB image, which can be directly used for subsequent image analysis and processing.

[0077] 2. Sensor installation

[0078] Temperature sensor installation: Fix the temperature sensor at an appropriate position in the mixing barrel to ensure that the temperature of the solution can be accurately measured. The installation position is the bottom and middle of the mixing barrel, avoiding direct contact with the stirring device to prevent damage;

[0079] Installation of viscosity sensor: Install the viscosity sensor on the side wall of the mixing barrel and contact the solution through a dedicated interface to ensure real-time collection of viscosity data;

[0080] Color sensor installation: Install the color sensor and camera on the outside of the transparent window of the mixing tank to monitor the color change of the solution in real time through the light-transmitting window;

[0081] 3. Data Acquisition System

[0082] Hardware:

[0083] Equipment selection: Choose suitable industrial computer data acquisition equipment, which needs to have sufficient processing power and interfaces, be compatible with various sensors, and support real-time data transmission;

[0084] Equipment configuration: Ensure that the industrial computer is installed with the necessary drivers and interface modules to communicate data with the sensor, connect the sensor to the corresponding interface of the data acquisition device, and ensure that the connection is stable and reliable;

[0085] Software: Develop data acquisition software and use Python to write data acquisition programs. The following functions must be implemented:

[0086] Initialize the sensor: configure the initialization parameters of the sensor to ensure that the sensor operates in the correct working state;

[0087] Read sensor data: Read data from sensors regularly and receive data from temperature, viscosity and color sensors through the interface;

[0088] Data processing: preliminary processing of raw data, including noise filtering and data smoothing, to ensure the accuracy and stability of the data;

[0089] Real-time monitoring: The processed data is displayed in real time on the monitoring interface and saved to the local database or cloud server for subsequent analysis;

[0090] Alarm system: Set alarm thresholds. When sensor data exceeds the preset range, an alarm prompt is triggered to notify operators in time for processing;

[0091] 4. Data calibration and testing

[0092] Calibration: Use standard reference materials to calibrate the sensor. By comparing the sensor reading with the standard value, adjust the sensor's calibration factor to ensure the accuracy of the reading;

[0093] Testing: Conduct preliminary tests to ensure that the sensor can collect data stably and accurately. Record the data during the test and analyze the accuracy and consistency of the data to ensure that the sensor can work properly under different environmental conditions.

[0094] Step B: Building and training a multimodal convolutional neural network model with an attention mechanism: Implement a multimodal convolutional neural network (CNN) structure with an attention mechanism, which combines humidity sensors, temperature sensors, and image data to identify and determine the progress of the polymerization reaction and analyze the color and viscosity of the solution;

[0095] Data preparation;

[0096] Image data: Images of the solution at different stages of the reaction process are captured by a color sensor or camera;

[0097] Temperature and humidity data: The temperature and humidity changes during the reaction process are recorded in real time through temperature sensors and humidity sensors;

[0098] Data annotation;

[0099] Three different color ranges are defined, corresponding to different reaction periods and viscosity categories:

[0100] Initial stage of reaction: color is pale yellow to light yellow; viscosity is low;

[0101] Description: In the early stage of the reaction, the solution is lighter in color and has a lower viscosity; the polymer monomers have just started to polymerize, and most of the solution is unreacted monomers;

[0102] Mid-reaction period: the color is between light yellow and dark yellow; the viscosity is medium;

[0103] Description: In the middle of the reaction, the color of the solution gradually deepens and the viscosity begins to increase; as the polymerization reaction gradually proceeds, the concentration of polymer molecules in the solution increases, and the viscosity rises accordingly;

[0104] Late stage of reaction: color is dark yellow to amber; viscosity is high;

[0105] Description: In the late stage of the reaction, the solution is darkest in color and has the highest viscosity; the polymerization reaction is basically completed, and most of the solution is high molecular weight polymer molecules;

[0106] Match each image with the corresponding temperature and humidity data to build a multimodal dataset;

[0107] Data preprocessing;

[0108] Image data preprocessing:

[0109] Resizing: resize all images to a uniform size, at 224x224 pixels;

[0110] Normalization: Normalize the image pixel values ​​and scale the pixel values ​​to the range of [0, 1];

[0111] Data enhancement: Perform data enhancement operations such as rotation, flipping, and scaling to increase the diversity of the data set and prevent overfitting;

[0112] Sensor data preprocessing: Standardization: Standardize the temperature and humidity data to make them have the same scale to facilitate fusion with image data;

[0113] Network implementation;

[0114] Image feature extraction module: The input image data passes through a series of convolutional layers and pooling layers to extract low-level and high-level features; the convolution calculation formula is:

[0115]

[0116] In the formula, I is the input image pixel matrix, K represents the filter matrix, M corresponds to the height of the filter, and N corresponds to the width of the filter; O represents the corresponding output matrix after convolution; i and j correspond to the horizontal and vertical coordinate positions of the output matrix O respectively; m and n correspond to the horizontal and vertical coordinate positions of the filter matrix K respectively;

[0117] Pooling calculation formula:

[0118]

[0119] Among them, Q(x, y) is the output feature map after pooling, x, y represent the horizontal and vertical position indexes of the output feature map, u and v are the horizontal and vertical position indexes in the pooling window, h and w represent the height and width of the pooling window, s h and w Represent the strides of the pooling operation in the vertical and horizontal directions respectively; P is the input feature map;

[0120] Sensor data processing module: Input temperature data T and humidity data H are extracted through the fully connected layer; the calculation formula of the fully connected layer is:

[0121] F senor =σ(W sensor ·[T,H]+b sensor )

[0122] Among them, F sensor is the calculated sensor feature, W senor represents the weight of the fully connected layer for calculating temperature and humidity features, b sensor represents the corresponding bias term, and σ() represents the activation function;

[0123] Attention mechanism module: Calculates the attention weights between image features and sensor features through the following steps:

[0124] Calculate attention energy: attention energy e i It is calculated by linearly transforming and adding the image features and sensor features through a nonlinear activation function. The calculation formula is as follows:

[0125] e i =tanh(W a F image，i +U a F sensor +b a )

[0126] Among them, e i is the attention energy corresponding to the i-th image feature, F image，i represents the feature vector of the i-th image feature, F sensor represents the sensor feature vector, W a is the weight matrix related to the linear transformation of image features, U a is the weight matrix related to the linear transformation of sensor features, tanh is the hyperbolic tangent function, as a nonlinear activation function, b a is the bias vector of the nonlinear transformation in the attention mechanism, and a is the abbreviation of attention, indicating that the matrices involved are all parameters in the attention operation;

[0127] Calculate attention weight: attention weight α i By focusing on the energy i Normalize the softmax function to get it;

[0128]

[0129] In the formula, α iis the attention weight corresponding to the i-th image feature, e i is the attention energy corresponding to the i-th image feature, N is the number of image features, exp() represents an exponential function used to map the attention energy to a positive range, and j is an index variable used to traverse all input features; e j Represents the attention score of the jth element

[0130] Calculate the context vector: The context vector C is the image feature according to the attention weight α i Obtained by weighted summation;

[0131]

[0132] C represents the context vector, α i is the attention weight corresponding to the i-th image feature, N is the number of image features, and F image，i Represents the feature vector of the i-th image feature;

[0133] Feature fusion: concatenate the context vector C with the sensor features to obtain fused features;

[0134] F fused =concat(C,F sensor )

[0135] F fused Indicates fusion features, concat() indicates feature concatenation operation, which concatenates two feature vectors together;

[0136] Classification module: Input the fusion features into the fully connected layer and calculate the classification result through the activation function. The calculation formula is as follows:

[0137] F fc1 =σ(W fc1 ·F fused +b fc1 )

[0138]

[0139] In the formula, F fc1 is the fully connected layer feature in the classification module, W fc1 , W out is the weight of the fully connected layer in the classification module, b fc1 , b out is the corresponding bias vector in the classification module, is the output classification probability, softmax() represents an activation function;

[0140] Model training: Train a multimodal convolutional neural network model with an attention mechanism to achieve real-time analysis and judgment of solution color and viscosity;

[0141] Data preparation: Divide the sample set into training set, validation set, and test set;

[0142] Hyperparameter settings: learning rate, batch size, training rounds, optimizer, loss function, activation function, convolution kernel size, number of convolution layers, number of convolution kernels per layer, pooling layer size, number of neurons in fully connected layers, attention mechanism dimension, dropout probability;

[0143] Model evaluation metrics:

[0144] Accuracy: The proportion of samples predicted correctly;

[0145] Precision: The ratio of correctly predicted positive samples to all predicted positive samples.

[0146] Recall rate (recal 1): the proportion of correctly predicted positive samples to all actual positive samples;

[0147] F1 score: the harmonic mean of precision and recall;

[0148] Step C: Implementation of the control algorithm combined with the intelligent optimization algorithm: Design a second-order synovial control algorithm combined with a genetic algorithm, and use the genetic algorithm to optimize the parameters of the synovial control algorithm to improve the control performance of the system; the input of the synovial control algorithm is determined by the classification output of the convolutional neural network model in step B, and finally the output of the synovial control is used to adjust the heating temperature and stirring speed to ensure the optimization of the reaction conditions; the implementation steps are as follows:

[0149] Convolutional neural network output processing: Use the convolutional neural network model in step B to classify and judge the current reaction period and viscosity state to obtain the current system state;

[0150] Determine the system model:

[0151]

[0152] Where x(t) represents the system state, u(t) is the control input, f() represents the system dynamic equation, and the nonlinear dynamic part of the system describes the behavior of the system when there is no control input u(t); It represents the first-order derivative of the system state, t is the time variable, g() is the control gain of the system, and describes how the control input u(t) affects the dynamics of the system;

[0153] Sliding surface design:

[0154]

[0155]

[0156] Where λ is the synovial parameter, s(x, t) represents the synovial surface function, is the derivative of the synovial surface function, is the second-order derivative of the system state;

[0157] Design of synovial control law: Where u(t) is the control input, K is the control gain, and sign() is the compliance function;

[0158] Genetic algorithm to optimize sliding mode control parameters: Genetic algorithm is used to optimize the sliding mode control parameters λ and control gain K to minimize the control error;

[0159] Initialize the population: determine the population size N; randomly generate the initial population, each individual i contains the synovial surface parameter λ and the control gain K;

[0160] Fitness function: Where J is the value of the optimization objective function, e(t) is the tracking error, defined as the difference between the actual system state and the desired state; p is a positive weight factor, and t represents the termination time;

[0161] Selection: Use the roulette wheel selection algorithm to select excellent individuals based on fitness values;

[0162] Calculate the selection probability for each individual: Where P i represents the probability of each individual's choice; J i The fitness function value corresponding to the i-th individual, N is the population size;

[0163] Crossover method: Use single-point crossover to perform genetic recombination on selected individuals;

[0164] Mutation method: Use uniform mutation to perform genetic mutation on newly generated individuals; the formula is as follows:

[0165] λ new =λ old + mutation_rate · (rand() - 0.5)

[0166] K new =K old + mutation_rate · (rand() - 0.5)

[0167] Among them, λ new represents the new synovial surface parameter after the mutation operation, λ old is the synovial surface parameter before mutation operation, K new is the new control gain after the mutation operation, Kold is the new control gain before the mutation operation, mutation_rate represents the mutation rate parameter, and rand() is the random function;

[0168] Repeat the above steps iteratively until the fitness function value converges to the preset number of iterations;

[0169] Step D: Real-time parameter adjustment based on the output of the control algorithm

[0170] Objective: Use the output of the control algorithm obtained in step C, i.e., the control input, to adjust the heating temperature and stirring speed in real time to ensure the optimization of reaction conditions;

[0171] Control algorithm output processing: according to the synovial control parameters λ and K optimized in step C, the control input u(t) is calculated using the control algorithm;

[0172] According to the value range of the control input, the corresponding adjustment method is:

[0173] Heating temperature adjustment: When u(t)>0, increase the heating temperature; the adjustment method is When u(t)<0, reduce the heating temperature; the adjustment method is Where T represents temperature, T min Represents the minimum temperature, T max represents the maximum temperature, u max Represents the maximum value of the control input;

[0174] Stirring speed adjustment: When u(t)>0, increase the stirring speed. The adjustment method is When u(t)<0, reduce the stirring speed and the adjustment method is Among them, S(t) represents the speed, S min Indicates the minimum value of the speed, S max Indicates the maximum value of the rotation speed;

[0175] The specific application cases are as follows:

[0176] Step A: Sensor deployment and data collection

[0177] Sensor selection and parameters:

[0178] Temperature sensor: PT100 RTD sensor, model: Omega PT100, measurement range: -200℃ to 850℃, accuracy: ±0.1℃;

[0179] Viscosity sensor: Brookfield rotational viscometer, model: Brookfield DV2T, measuring range: 1cP (centipoise, unit of dynamic viscosity) to 6,000,000cP, accuracy: ±1%;

[0180] Color sensor: Adafruit TCS34725 RGB color sensor, model: TCS34725, measurement range: full spectrum visible light, accuracy: 16-bit ADC;

[0181] Data Acquisition System

[0182] Hardware:

[0183] Data acquisition device: Raspberry Pi 4, model: Raspberry Pi 4 Model B

[0184] Processor: Broadcom BCM2711, Quad-core Cortex-A72 (ARM v8) 64-bit SoC@1.5GHz, Memory: 4GB LPDDR4-3200 SDRAM, Interface: 2×micro-HDMI ports, 2×USB 3.0 ports, 2×USB 2.0 ports, GPIO pins;

[0185] Data collection volume:

[0186] Data is collected once per second (1Hz), and each collection includes three types of data: temperature, viscosity and color;

[0187] Data collection duration: 1 hour (3600 seconds);

[0188] Total data volume: 3600 acquisitions;

[0189] The data examples are shown in the table

[0190] Time (seconds) Temperature(℃) Viscosity(cP) Color (RGB) 0 25.3 1500 (120，85，90) 1 25.4 1520 (121，86，91) 2 25.5 1535 (122，87，92) … … … … 3599 30.0 2000 (130，100，110)

[0191] Step B: Building and training a multimodal convolutional neural network model with an attention mechanism;

[0192] Data preparation: Use the data collected in the first step to construct a multimodal dataset, including temperature data, viscosity data, and color image data;

[0193] The overall structure of the network model:

[0194] Image feature extraction module:

[0195] Input layer (Image Input): shape (224, 224, 3);

[0196] Convolutional layer 1 (Conv2D): 64 3x3 convolution kernels, ReLU activation function;

[0197] MaxPooling2D: pooling size 2x2;

[0198] Convolutional layer 2 (Conv2D): 128 3x3 convolution kernels, ReLU activation function;

[0199] MaxPooling2D: pooling size 2x2;

[0200] Convolutional layer 3 (Conv2D): 256 3x3 convolution kernels, ReLU activation function;

[0201] MaxPooling2D: pooling size 2x2;

[0202] Fully connected layer 1 (Dense): 1024 neurons, ReLU activation function;

[0203] Fully connected layer 2 (Dense): 512 neurons, ReLU activation function;

[0204] Output layer (Dense): 256 neurons, ReLU activation function;

[0205] Sensor data processing module:

[0206] Input layer (Temperature Input): shape (1);

[0207] Fully connected layer (Dense): 64 neurons, ReLU activation function;

[0208] Input layer (Viscosity Input): shape (1);

[0209] Fully connected layer (Dense): 64 neurons, ReLU activation function;

[0210] Multimodal data fusion layer:

[0211] Image feature (256) + temperature feature (64) + viscosity feature (64);

[0212] Concatenate layer: output shape (256+64+64);

[0213] Fully connected layer (Dense): 512 neurons, ReLU activation function;

[0214] Attention mechanism module:

[0215] Attention Weights:

[0216] Fully connected layer 1 (Dense): 128 neurons, ReLU activation function;

[0217] Fully connected layer 2 (Dense): 1 neuron, softmax activation function;

[0218] Weighted Sum: Output shape (512);

[0219] Output layer:

[0220] Fully connected layer (Dense): 128 neurons, ReLU activation function;

[0221] Output layer (Dense): number of output categories (3-stage classification), softmax activation function;

[0222] Training hyperparameter settings

[0223] Learning Rate: 0.001, Batch Size: 32, Epochs: 50, Optimizer: Adam, Loss Function: CategoricalCrossentropy, Activation Functions: ReLU (hidden layer), softmax (output layer), Data Augmentation: Random Rotation, Horizontal Flip, Vertical Flip, Scaling, Regularization: Dropout (0.5);

[0224] Step C: Implementation of control algorithm combined with intelligent optimization algorithm

[0225] The state equation of the system is:

[0226] Sliding surface design: follow the formula described above and set the sliding surface parameter λ = 1;

[0227] Control law design: According to the formula described above, set the control gain K = 10;

[0228] Genetic algorithm optimizes synovial control parameters:

[0229] Setting parameters: fitness function weight ρ = 0.1, population size N = 50, number of iterations G = 100, crossover probability Pc = 0.8, mutation probability P m =0.01;

[0230] Initialize the population: randomly generate N = 50 individuals, each of which represents a sliding surface parameter λ and a control gain K, and the range is set between [0.1, 5];

[0231]

[0232]

[0233] After iteration, the best parameters obtained by genetic algorithm optimization are λ = 2.1, K = 12.1;

[0234] According to the obtained optimal parameters, the control input u(t)=15 is calculated;

[0235] The initial heating temperature and stirring speed were 25 °C and 300 rpm (rotation speed), respectively;

[0236] New heating temperature:

[0237] New stirring speed:

[0238] Therefore, the control parameters of the synovial membrane obtained by genetic algorithm optimization can be used to adjust the heating temperature and stirring speed in real time using the control input output by the control algorithm.

[0239] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for preparing a drawing film composite packaging material, characterized in that: The steps include: Select polymer monomers, dissolve the monomers in a suitable solvent, heat to a suitable temperature, and stir and mix using a stirring device to allow the monomers to undergo a polymerization reaction to form a polymer solution; The polymer solution is dehydrated to remove a small amount of low molecular weight impurities in the solution to obtain a relatively pure polymer; Dry the polymer to an appropriate moisture content so that it solidifies into solid particles; The dried polymer particles are subjected to film-forming treatment to form soft and tough films; According to the needs, the film is cut and formed into the required packaging material; The steps of selecting polymer monomers and dissolving the monomers in appropriate solvents are as follows: Intelligent optimization algorithms and deep learning algorithms are introduced to monitor and adjust the solution temperature and stirring speed in real time to optimize the polymerization reaction efficiency; the specific implementation process is as follows: Step A: Sensor deployment and data collection: Deploy sensors to collect solution temperature, viscosity and color data to ensure accurate and real-time data feedback to the control system; Temperature sensor: Using a resistance temperature detector, the calculation formula after acquisition is as follows: Where T is the temperature, R is the resistance of the temperature sensor, R0 is the resistance of the sensor at 0°C, and α represents the temperature coefficient of the sensor; Viscosity sensor: Using a rotational viscometer, this sensor can monitor the viscosity changes of the solution in real time; the viscosity is calculated as: η=K·(f0-f) Where η is the calculated viscosity value, f represents the current vibration frequency, f0 is the vibration frequency of the sensor under no-load condition, and K is the calibration coefficient of the sensor; Color sensor: using a spectrometer and a CMOS image sensor. The spectrometer can capture the color changes of the solution, while the CMOS image sensor can take images of the solution for subsequent image analysis. For the color sensor, its output is an RGB image, which can be directly used for subsequent image analysis and processing. Step B: Building and training a multimodal convolutional neural network model with an attention mechanism: This structure combines viscosity sensors, temperature sensors, and image data to identify and determine the progress of the polymerization reaction and analyze the color and viscosity of the solution; Three different color ranges are defined, corresponding to different reaction periods and viscosity categories: Initial stage of reaction: color is pale yellow to light yellow; viscosity is low; Mid-reaction period: the color is between light yellow and dark yellow; the viscosity is medium; Late stage of reaction: color is dark yellow to amber; viscosity is high; Match each image with the corresponding temperature and viscosity data to construct a multimodal dataset; Image data preprocessing: Resizing: resize all images to a uniform size, at 224x224 pixels; Normalization: Normalize the image pixel values ​​and scale the pixel values ​​to the range of [0, 1]; Sensor data preprocessing: Standardization: Standardize the temperature and viscosity data to make them have the same scale to facilitate fusion with image data; Network Implementation: Image feature extraction module: The input image data X passes through a series of convolutional layers and pooling layers to extract low-level and high-level features; the convolution calculation formula is: In the formula, I is the input image pixel matrix, K represents the filter matrix, M corresponds to the height of the filter, and N corresponds to the width of the filter; O represents the corresponding output matrix after convolution; i and j correspond to the horizontal and vertical coordinate positions of the output matrix 0 respectively; m and n correspond to the horizontal and vertical coordinate positions of the filter matrix K respectively; Pooling calculation formula: Among them, Q(x, y) is the output feature map after pooling, x, y represent the horizontal and vertical position indexes of the output feature map, u and v are the horizontal and vertical position indexes in the pooling window, h and w represent the height and width of the pooling window, s h and w Represent the strides of the pooling operation in the vertical and horizontal directions respectively; P is the input feature map; Sensor data processing module: Input temperature data T and viscosity data H are extracted through the fully connected layer; the fully connected layer calculation formula is: F sensor =σ(W sensor ·[T,H]+b sensor ) Among them, F sensor is the calculated sensor feature, W sensor represents the weight of the fully connected layer for calculating temperature and viscosity features, b sensor represents the corresponding bias term, and σ() represents the activation function; Attention mechanism module; calculates the attention weights between image features and sensor features through the following steps: Calculate attention energy: attention energy e i It is calculated by linearly transforming the image features and sensor features, adding them together, and then using a nonlinear activation function. The calculation formula is as follows: e i =tanh(W a F image，i +U a F sensor +b a ) Among them, e i is the attention energy corresponding to the i-th image feature, F image，i represents the feature vector of the i-th image feature, F sensor represents the sensor feature vector, W a is the weight matrix related to the linear transformation of image features, U a is the weight matrix related to the linear transformation of sensor features, tanh is the hyperbolic tangent function, as a nonlinear activation function, b a is the bias vector of the nonlinear transformation in the attention mechanism, a is the attention, indicating that the matrices involved are all parameters in the attention operation; Calculate attention weight: attention weight α i By focusing on the energy i Normalize the softmax function to get it; In the formula, α i is the attention weight corresponding to the i-th image feature, e i is the attention energy corresponding to the i-th image feature, N is the number of image features, exp() represents an exponential function used to map the attention energy to a positive range, and j is an index variable used to traverse all input features; e j represents the attention score of the jth element; Calculate the context vector: The context vector C is the image feature according to the attention weight α i Obtained by weighted summation; C represents the context vector, α i is the attention weight corresponding to the i-th image feature, N is the number of image features, and F image，i Represents the feature vector of the i-th image feature; Feature fusion: concatenate the context vector C with the sensor features to obtain fused features; F fused =concat(C,F sensor ) F fused Represents fusion features, concat() represents feature concatenation operation, which concatenates two feature vectors together; Classification module: Input the fusion features into the fully connected layer, and calculate the classification results through the activation function. The calculation formula is as follows: F fc1 =σ(W fc1 ·F fused +b fc1 ) In the formula, F fc1 is the fully connected layer feature in the classification module, W fc1 , W out is the weight of the fully connected layer in the classification module, b fc1 , b out is the corresponding bias vector in the classification module, is the output classification probability, σ and softmax() represent the activation function; Model training: Train a multimodal convolutional neural network model with an attention mechanism to achieve real-time analysis and judgment of solution color and viscosity; Data preparation: Divide the sample set into training set, validation set, and test set; Hyperparameter settings: learning rate, batch size, training rounds, optimizer, loss function, activation function, convolution kernel size, number of convolution layers, number of convolution kernels per layer, pooling layer size, number of neurons in fully connected layers, attention mechanism dimension, dropout probability; Step C: Implementation of the control algorithm combined with the intelligent optimization algorithm: Design a second-order synovial control algorithm combined with a genetic algorithm, and use the genetic algorithm to optimize the parameters of the synovial control algorithm to improve the control performance of the system; the input of the synovial control algorithm is determined by the classification output of the convolutional neural network model in the second step, and finally the output of the synovial control is used to adjust the heating temperature and stirring speed to ensure the optimization of the reaction conditions; the implementation steps are as follows: Convolutional neural network output processing: Use the convolutional neural network model in step B to classify and judge the current reaction period and viscosity state to obtain the current system state; Determine the system model: Where x(t) represents the system state, u(t) is the control input, f() represents the system dynamic equation, and the nonlinear dynamic part of the system describes the behavior of the system when there is no control input u(t); It represents the first-order derivative of the system state, t is the time variable, g() is the control gain of the system, and describes how the control input u(t) affects the dynamics of the system; Sliding surface design: Where λ is the synovial parameter, s(x, t) represents the synovial surface function, is the derivative of the synovial surface function, is the second-order derivative of the system state; Design of synovial control law: Where u(t) is the control input, K is the control gain, and sign() is the sign function; Genetic algorithm to optimize sliding mode control parameters: Genetic algorithm is used to optimize the sliding mode control parameters λ and control gain K to minimize the control error; Initialize the population: determine the population size N; randomly generate the initial population, each individual i contains the synovial surface parameter λ and the control gain K; Fitness function: Where J is the value of the optimization objective function, e(t) is the tracking error, defined as the difference between the actual system state and the desired state; ρ is a positive weight factor, and t represents time; Selection: Use the roulette wheel selection algorithm to select excellent individuals based on fitness values; Calculate the selection probability for each individual: Where P i represents the probability of each individual's choice; J i The fitness function value corresponding to the i-th individual, N is the population size; Crossover method: Use single-point crossover to perform genetic recombination on selected individuals; Mutation method: Use uniform mutation to perform genetic mutation on newly generated individuals; the formula is as follows: l new =λ old +mutation_rate·(rand()-0.5) K new =K old +mutation_rate·(rand()-0.5) Among them, λ new represents the new synovial surface parameter after the mutation operation, λ old is the synovial surface parameter before mutation operation, K new is the new control gain after mutation operation, K old It is the new control increment before the mutation operation, mutation_rate represents the mutation rate parameter, and rand() is the random function; Repeat the above steps iteratively until the fitness function value converges to the preset number of iterations; Step D: adjusting parameters in real time according to the output of the control algorithm: using the output of the control algorithm obtained in step C, i.e., the control input, to adjust the heating temperature and stirring speed in real time to ensure the optimization of the reaction conditions; Control algorithm output processing: according to the synovial control parameters λ and K optimized in step C, the control input u(t) is calculated using the control algorithm; According to the value range of the control input, the corresponding adjustment method is: Heating temperature adjustment: When u(t)>0, increase the heating temperature; the adjustment method is When u(t)<0, reduce the heating temperature; the adjustment method is Where T represents temperature, T min Represents the minimum temperature, T max represents the maximum temperature, u max Represents the maximum value of the control input; Stirring speed adjustment: When u(t)>0, increase the stirring speed. The adjustment method is When u(t)<0, reduce the stirring speed and the adjustment method is Among them, S(t) represents the speed, S min Indicates the minimum value of the speed, S max Indicates the maximum speed.

2. The method for preparing a drawing film composite packaging material according to claim 1, characterized in that: The polymer monomers are vinyl acetate, polyvinyl chloride, methyl methacrylate and acetyl methyl methacrylate.

3. The method for preparing the drawn film composite packaging material according to claim 1, characterized in that: The polymer is dried by vacuum freezing and spray drying.

4. The method for preparing a drawing film composite packaging material according to claim 1, characterized in that: The film forming process is performed by casting and calendering.

Images