Production control method and system for net film composite product

Abstract

The invention belongs to the technical field of net film application. The invention provides a production control method and system for a net film composite product. The method comprises the steps that first type information of a hot melt adhesive net film is obtained, a high-resolution image of the hot melt adhesive net film is shot, net film structure information of the hot melt adhesive net film is obtained through extraction, and net film information is formed; acquiring second type information of the base material, the thickness of the base material and a composite mode of the base material to form base material information; analyzing and processing the net film information and the base material information by using a trained composite process parameter optimization model to obtain a first composite process parameter predicted by the composite process parameter optimization model; and the composite roller is controlled to conduct composite treatment on the hot melt adhesive net film and the base material according to the first composite technological parameters. The composite process parameters of the composite roller are obtained through comprehensive analysis according to the information of the hot melt adhesive net film, the information of the base material and other factors, the composite product with the firmness reaching the standard can be obtained, and the power consumption of the composite roller is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of omentum application, and in particular to a production control method and system for omentum composite products. Background Art

[0002] Omentum composite products are formed by combining omentum with one or more other materials through physical or chemical methods to form composite materials with new properties and functions. Omentum composite products can be divided into many types according to the type of omentum used and the type of composite material. Hot melt adhesive omentum composite products are a common omentum composite product.

[0003] The production process of hot melt adhesive mesh composite products mainly includes hot melt adhesive mesh preparation, composite processing, and post-processing. The hot melt adhesive mesh preparation and composite processing procedures are specifically introduced as follows: Hot melt adhesive mesh preparation: 1) Hot melt adhesive melting: Pour the purchased hot melt adhesive particles into the barrel of the hot melt adhesive machine and heat them to a specific temperature to convert the solid adhesive particles into a uniform liquid. This temperature is usually between 120℃ and 200℃, and there are differences for different types of adhesives. For example, EVA hot melt adhesive has a lower melting point, while PA and PES hot melt adhesives have higher melting points. At the same time, turn on the stirring device to avoid local overheating or burning, so that the glue is heated evenly.

[0004] 2) Membrane forming: Liquid hot melt adhesive is accurately delivered to the mesh forming mold with the help of a metering pump, extruded through the tiny pores of the mold, and rapidly cooled and solidified under the action of a cooling roller or cooling air to form a hot melt adhesive mesh with continuous pores and uniform thickness, usually between 0.01mm and 0.2mm.

[0005] Composite processing: 1) Material unwinding: Place the substrates to be composited, such as cloth, leather, plastic film, etc., on the unwinding rack, pass through the guide rollers, and ensure that they are transported to the composite area flat and wrinkle-free; the unwinding tension should be stable to avoid stretching and deformation of the substrate.

[0006] 2) Gluing and laminating: After unwinding, the substrate moves to the gluing station, and the hot melt adhesive web will adhere to the surface of one of the substrates. Then, the two (or more) layers of substrate enter the laminating roller together. Under the action of heat and pressure, the hot melt adhesive web melts and tightly adheres to the upper and lower substrates. The temperature and pressure of the laminating roller are set according to the material and product requirements. The general temperature is 50℃-150℃ and the pressure is 2-10kg / cm².

[0007] Cooling and shaping: The composite product immediately enters the cooling area and is cooled by air cooling and water cooling devices to quickly solidify the hot melt adhesive, stabilize the composite structure, and prevent subsequent displacement and debonding; after cooling, the product is wound into large rolls.

[0008] As introduced above, during the gluing and laminating stage, the temperature of the laminating roller is generally set at 50℃-150℃, and the pressure on the substrate is generally set at 2-10kg / cm². However, the above temperature range and pressure range are too wide, and the types of hot melt adhesive webs and substrates used to manufacture web composite products are different. The above wide temperature range and pressure range are not conducive to preparing web composite products with optimal performance, such as web composite products having sufficiently high viscosity to ensure that the substrates do not separate, and are not conducive to reducing the power consumption of the laminating roller.

[0009] Therefore, how to automatically determine the temperature of the composite roller and its pressure on the substrate in a more targeted manner during the manufacturing process of the web composite product to improve the bonding strength of the web composite product is a technical problem that needs to be solved at present. Summary of the invention

[0010] In this regard, the present invention provides a production control method, system, electronic equipment, computer storage medium and computer program product for a mesh composite product to solve the above technical problems.

[0011] The invention discloses a production control method for a mesh composite product, the method comprising the following steps: constructing a plurality of training data based on actual production record data of the mesh composite product, constructing the training data set, and using the training data set to train a composite process parameter optimization model; obtaining first type information of a hot melt adhesive mesh used in a coating and composite stage, and taking a high-resolution image of the hot melt adhesive mesh, extracting mesh structure information of the hot melt adhesive mesh according to the high-resolution image, wherein the first type information and the mesh structure information constitute mesh information; obtaining second type information, substrate thickness and substrate composite method of a substrate used in a coating and composite stage, wherein the second type information, substrate thickness and substrate composite method constitute substrate information; using the trained composite process parameter optimization model to analyze and process the mesh information and the substrate information, and obtaining first composite process parameters predicted by the composite process parameter optimization model, wherein the first composite process parameters include a composite roller temperature range and a composite roller pressure range; and controlling the composite roller to composite the hot melt adhesive mesh and the substrate according to the composite roller temperature range and the composite roller pressure range.

[0012] In some embodiments, the use of the training data set to train the composite process parameter optimization model includes: using the training data set to perform preliminary training on the composite process parameter optimization model, using the composite process parameter optimization model after preliminary training to conduct several trial productions, that is, using the composite process parameter optimization model to predict the corresponding second composite process parameters, and using the second composite process parameters to perform composite treatment on the trial-produced hot melt adhesive mesh and substrate; performing firmness testing and evaluation on the mesh composite product obtained from the trial production to obtain several firmness evaluation scores, calculating a comprehensive evaluation score based on each of the firmness evaluation scores, and obtaining a training intensity of the secondary training based on the matching of the comprehensive evaluation scores; wherein the training intensity includes the number of training data and the number of types of training data; and performing secondary training on the composite process parameter optimization model according to the training intensity.

[0013] In some embodiments, the mesh composite product obtained from the trial production is subjected to a firmness test and evaluation to obtain a number of firmness evaluation scores, and a comprehensive evaluation score is calculated based on each of the firmness evaluation scores, including: conducting a firmness test on a sample of the mesh composite product obtained from the trial production, taking the maximum tensile force value when the sample breaks as the first index value, and evaluating the second index value based on the failure morphology of the sample stratification position; calculating a first mean of each of the first index values, calculating a second mean of each of the second index values, and calculating the comprehensive evaluation score based on the first mean and its first weight, and the second mean and its second weight.

[0014] In some embodiments, the training intensity of the secondary training obtained by matching the comprehensive evaluation scores includes: obtaining a preliminary training intensity of the secondary training according to the comprehensive evaluation scores, the preliminary training intensity including a first quantity of training data and a first type of quantity of training data; a second type of quantity of omentum composite products involved in a statistical firmness test, obtaining an adjustment value according to matching the second type of quantity, and adjusting the preliminary training intensity to a target training intensity using the adjustment value, the target training intensity including a second quantity of training data and a third type of quantity of training data; wherein the size of the adjustment value is negatively correlated with the second type of quantity.

[0015] In some embodiments, the obtaining of an adjustment value based on the matching of the second type of quantity and using the adjustment value to adjust the preliminary training intensity to a target training intensity, wherein the target training intensity includes a second quantity of training data and a third type of quantity of training data, comprises: obtaining a first adjustment value and a second adjustment value based on the matching of the second type of quantity, using the first adjustment value to adjust the first quantity of training data in the preliminary training intensity to the second quantity, and using the second adjustment value to adjust the first type of quantity of training data in the preliminary training intensity to the third type of quantity.

[0016] The present invention also discloses a production control system for mesh composite products, the system comprising at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program, when executed by the processor, executes the following steps: constructing a plurality of training data based on actual production record data of mesh composite products, constructing the data into a training data set, and using the training data set to train a composite process parameter optimization model; obtaining first type information of the hot melt adhesive mesh used in the gluing and compounding stages, taking a high-resolution image of the hot melt adhesive mesh, and extracting the mesh structure of the hot melt adhesive mesh according to the high-resolution image; The invention relates to a method for preparing a composite film of a hot melt adhesive and a composite film, wherein the first type of information and the web structure information constitute the web information; obtaining the second type of information of the substrate used in the coating and composite stage, the substrate thickness and the substrate composite method, wherein the second type of information, the substrate thickness and the substrate composite method constitute the substrate information; using the trained composite process parameter optimization model to analyze and process the web information and the substrate information, and obtaining the first composite process parameter predicted by the composite process parameter optimization model, wherein the first composite process parameter includes a composite roller temperature range and a composite roller pressure range; controlling the composite roller to composite the hot melt adhesive web and the substrate according to the composite roller temperature range and the composite roller pressure range.

[0017] In some embodiments, the use of the training data set to train the composite process parameter optimization model includes: using the training data set to perform preliminary training on the composite process parameter optimization model, using the composite process parameter optimization model after preliminary training to conduct several trial productions, that is, using the composite process parameter optimization model to predict the corresponding second composite process parameters, and using the second composite process parameters to perform composite treatment on the trial-produced hot melt adhesive mesh and substrate; performing firmness testing and evaluation on the mesh composite product obtained from the trial production to obtain several firmness evaluation scores, calculating a comprehensive evaluation score based on each of the firmness evaluation scores, and obtaining a training intensity of the secondary training based on the matching of the comprehensive evaluation scores; wherein the training intensity includes the number of training data and the number of types of training data; and performing secondary training on the composite process parameter optimization model according to the training intensity.

[0018] In some embodiments, the mesh composite product obtained from the trial production is subjected to a firmness test and evaluation to obtain a number of firmness evaluation scores, and a comprehensive evaluation score is calculated based on each of the firmness evaluation scores, including: conducting a firmness test on a sample of the mesh composite product obtained from the trial production, taking the maximum tensile force value when the sample breaks as the first index value, and evaluating the second index value based on the failure morphology of the sample stratification position; calculating a first mean of each of the first index values, calculating a second mean of each of the second index values, and calculating the comprehensive evaluation score based on the first mean and its first weight, and the second mean and its second weight.

[0019] In some embodiments, the training intensity of the secondary training obtained by matching the comprehensive evaluation scores includes: obtaining a preliminary training intensity of the secondary training according to the comprehensive evaluation scores, the preliminary training intensity including a first quantity of training data and a first type of quantity of training data; a second type of quantity of omentum composite products involved in a statistical firmness test, obtaining an adjustment value according to matching the second type of quantity, and adjusting the preliminary training intensity to a target training intensity using the adjustment value, the target training intensity including a second quantity of training data and a third type of quantity of training data; wherein the size of the adjustment value is negatively correlated with the second type of quantity.

[0020] In some embodiments, the obtaining of an adjustment value based on the matching of the second type of quantity and using the adjustment value to adjust the preliminary training intensity to a target training intensity, wherein the target training intensity includes a second quantity of training data and a third type of quantity of training data, comprises: obtaining a first adjustment value and a second adjustment value based on the matching of the second type of quantity, using the first adjustment value to adjust the first quantity of training data in the preliminary training intensity to the second quantity, and using the second adjustment value to adjust the first type of quantity of training data in the preliminary training intensity to the third type of quantity.

[0021] The present invention also discloses an electronic device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program implements the method described in any of the preceding items when executed.

[0022] The present invention also discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed, the method described in any of the preceding items is implemented.

[0023] The present invention also discloses a computer program product. The computer program product is prepackaged with computer program codes. When the computer program codes are executed, the method described in any of the above items is implemented.

[0024] The beneficial effect of the present invention is that the scheme of the present invention can comprehensively analyze factors such as the information of the hot melt adhesive web used in the coating and compounding stages, the information of the substrate, and the substrate compounding method to obtain the temperature range and pressure range of the composite roller. The substrate and the hot melt adhesive web are heat-treated and compounded according to the temperature range and pressure range to obtain a composite product with a firmness that meets the standard. At the same time, it is also beneficial to reduce the power consumption of the composite roller to improve economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 It is a schematic diagram of the appearance of the hot melt adhesive mesh.

[0027] Figure 2 It is a flow chart of a production control method of a mesh composite product disclosed in an embodiment of the present invention.

[0028] Figure 3It is a structural schematic diagram of a production control system for a mesh composite product disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0030] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0031] like Figure 2 As shown, an embodiment of the present invention discloses a production control method for a mesh composite product, the method comprising the following steps: S1, constructing a plurality of training data based on actual production record data of the mesh composite product, constructing the data into a training data set, and using the training data set to train a composite process parameter optimization model.

[0032] In this step, the actual production record data of the web composite products in the past period of time is extracted from the production management system, preferably covering the manufacturing details of different batches and different specifications of products. These data include the type of hot melt adhesive web used, web structure, substrate type, substrate thickness and substrate composite method (such as substrate 1-hot melt adhesive web-substrate 1, substrate 1-hot melt adhesive web-substrate 2, substrate 1-hot melt adhesive web-substrate 1-hot melt adhesive web-substrate 2, etc.), the set composite process parameters (i.e., composite roller temperature range, composite roller pressure range), and the final product firmness quality inspection score results, a total of six key elements.

[0033] At the same time, it is also necessary to remove data rows with obvious errors or missing key values ​​in the actual production records. For example, if a record lacks the thickness information of the hot melt adhesive film and cannot be supplemented by related data, it will be eliminated. At the same time, it is also necessary to screen out abnormal production data generated during sudden equipment failures to ensure that the data can reflect the normal production status.

[0034] For the remaining valid actual production record data, the key elements including the above six aspects are extracted from each actual production record data to construct a training data, which includes label data, that is, the above-mentioned final product firmness quality inspection score result. The label data is used to evaluate the firmness score of the final product produced based on the other five key elements. Obviously, the higher the firmness of the final product, the higher the corresponding firmness score, and vice versa. Then, all the constructed training data are divided into a training set and a validation set according to a certain ratio (such as 70% for training and 30% for validation), which together constitute the training data set. A suitable machine learning algorithm is selected to build a composite process parameter optimization model, such as using a CNN algorithm or its improved body. The training set data is input into the composite process parameter optimization model to let the model learn the mapping relationship between different input features and expected outputs (i.e., the process parameters corresponding to high-quality products), and the validation set data is used to adjust the model's hyperparameters in a timely manner to prevent overfitting and improve the model's generalization ability. A composite process parameter optimization model that can achieve the prediction accuracy index can be obtained.

[0035] S2, obtaining the first type information of the hot melt adhesive web used in the gluing and compounding stage, and taking a high-resolution image of the hot melt adhesive web, extracting the web structure information of the hot melt adhesive web based on the high-resolution image, the first type information and the web structure information constitute the web information.

[0036] In this step, the first type information of the hot melt adhesive web used in this coating and compounding stage is read from the raw material management system. The first type information refers to the chemical type of the hot melt adhesive web, such as EVA, PA, PES, etc.

[0037] At the same time, high-resolution industrial cameras are also installed on the production line. When the hot melt adhesive web is unrolled and transported to a specific inspection station, the high-resolution industrial camera is triggered to shoot it and obtain a high-resolution image of the hot melt adhesive web. The camera lens uses a telephoto macro lens, combined with a ring light source, to ensure that the microscopic details of the web can be captured. The shooting resolution is set at more than 1000dpi to obtain clear images of the web surface and internal structure. Then, the image analysis software is used to process the captured high-resolution images, and the grayscale threshold segmentation algorithm is used to identify the web structure information such as the pore distribution (such as average area, porosity) and adhesive layer thickness of the web. Figure 1 shown.

[0038] Finally, the extracted retinal structure information and the first type of information are integrated into retinal information.

[0039] S3, obtaining the second type information, substrate thickness and substrate compounding method of the substrate used in the gluing and compounding stage, wherein the second type information, the substrate thickness and the substrate compounding method constitute substrate information.

[0040] In this step, the second type of information of the substrate used in the gluing and compounding stage is also queried with the help of the raw material management system, including textile fabric substrates, plastic film substrates, etc. Among them, the textile fabric substrate is further subdivided according to different fiber components (cotton, linen, polyester fiber, etc.) and fabric structure (plain, twill, satin); the plastic film substrate is also subdivided according to different polymer types, whether antistatic agents and other additives are added.

[0041] Before the substrate is unwound, its thickness is measured in real time using a high-precision laser thickness gauge. The laser thickness gauge emits a laser beam vertically to the surface of the substrate, and accurately calculates the thickness value based on the time difference of the reflected light. The average value is taken after multiple measurements to reduce the measurement error and ensure accurate substrate thickness data. The substrate composite method is extracted from the production process planning document to clarify whether it is double-layer lamination, multi-layer stacking composite, or sandwich composite. Each method also needs to include the specific substrate type and its location.

[0042] Finally, the obtained second type information, substrate thickness and substrate composite method are integrated into complete substrate information.

[0043] S4, using the trained composite process parameter optimization model to analyze and process the web information and the substrate information, to obtain the first composite process parameter predicted by the composite process parameter optimization model, wherein the first composite process parameter includes a composite roller temperature range and a composite roller pressure range.

[0044] In this step, the web information obtained in step S2 and the substrate information obtained in step S3 are integrated into a new set of data and input into the composite process parameter optimization model. The composite process parameter optimization model quickly calculates and outputs the predicted composite process parameters based on the previously learned rules. For the composite roller temperature range, the model will give the temperature range corresponding to the minimum temperature and the maximum temperature; for the composite roller pressure range, it will also accurately estimate the pressure range corresponding to the minimum pressure and the maximum pressure. The predicted composite process parameters are intended to ensure that the firmness of the produced composite products meets high quality standards.

[0045] S5, controlling the composite roller to composite the hot melt adhesive web and the substrate according to the composite roller temperature range and the composite roller pressure range.

[0046] The scheme of the present invention can obtain the temperature range and pressure range of the composite roller based on a comprehensive analysis of factors such as the information of the hot melt adhesive web used in the gluing and compounding stages, the information of the substrate, and the substrate compounding method. The substrate and the hot melt adhesive web are heat-treated and compounded according to the temperature range and pressure range to obtain a composite product with a firmness that meets the standard. At the same time, it is also beneficial to reduce the power consumption of the composite roller to improve economy.

[0047] In some embodiments, in step S1, the use of the training data set to train the composite process parameter optimization model includes: using the training data set to perform preliminary training on the composite process parameter optimization model, using the composite process parameter optimization model after preliminary training to perform several trial productions, that is, using the composite process parameter optimization model to predict the corresponding second composite process parameters, and using the second composite process parameters to perform composite treatment on the trial-produced hot melt adhesive mesh and substrate; performing firmness testing and evaluation on the mesh composite product obtained from the trial production to obtain several firmness evaluation scores, calculating a comprehensive evaluation score based on each of the firmness evaluation scores, and obtaining a training intensity of the secondary training based on the matching of the comprehensive evaluation scores; wherein the training intensity includes the number of training data and the number of types of training data; and performing secondary training on the composite process parameter optimization model according to the training intensity.

[0048] In an embodiment of the present invention, the composite process parameter optimization model is first trained once using a conventional training method. For example, 50% of the training data in the training data set obtained above can be used for one training, and 70% of the training data in one training can be used for training and 30% for verification until the training meets the standard, that is, the convergence condition is met. During the training process, the initial value of the learning rate can be set to 0.001. After every 50 epochs of training, if the loss value no longer decreases, the learning rate is appropriately reduced, and the training is continued until the loss function converges to a stable low value.

[0049] After the composite process parameter optimization model is trained to meet the standards, it is used for trial production. The composite process parameter optimization model analyzes and processes the web information of the hot melt adhesive web and the substrate information of the substrate actually used in the trial production to obtain the corresponding second composite process parameters, which also include the composite roller temperature range and the composite roller pressure range. The composite roller performs multiple trial productions based on each second composite process parameter. The multiple trial productions here include multiple trial productions of a certain product. A certain product refers to a web composite product produced using a specific type of hot melt adhesive web, a specific type and thickness of substrate, and a specific substrate composite method.

[0050] After the trial production is completed, the mesh composite products obtained from the trial production are tested and evaluated for firmness using a preset test method to obtain several firmness evaluation scores. For example, the tensile test method is used: the mesh composite products are cut into standard size specimens, fixed on the upper and lower clamps of the tensile testing machine, and pulled apart at a constant tensile speed (such as 100mm / min) until the specimens are delaminated or torn. The corresponding firmness evaluation score is obtained based on the test results, and then the various firmness evaluation scores are integrated, such as calculating the mean, weighted mean, etc., to obtain a comprehensive evaluation score. The higher the comprehensive evaluation score, the higher the firmness of the mesh composite products produced in the trial production, that is, the better the quality; the lower the comprehensive evaluation score, the lower the firmness of the mesh composite products produced in the trial production, that is, the worse the quality.

[0051] Then, the training intensity of the secondary training is obtained according to the comprehensive evaluation score obtained above, that is, the number of training data and the number of types of training data required in the secondary training process, so as to dynamically adjust the training intensity of the composite process parameter optimization model, so as to achieve a balance between training effect and training load.

[0052] In some embodiments, the mesh composite product obtained from the trial production is subjected to a firmness test and evaluation to obtain a number of firmness evaluation scores, and a comprehensive evaluation score is calculated based on each of the firmness evaluation scores, including: conducting a firmness test on a sample of the mesh composite product obtained from the trial production, taking the maximum tensile force value when the sample breaks as the first index value, and evaluating the second index value based on the failure morphology of the sample stratification position; calculating a first mean of each of the first index values, calculating a second mean of each of the second index values, and calculating the comprehensive evaluation score based on the first mean and its first weight, and the second mean and its second weight.

[0053] In an embodiment of the present invention, when the sample is delaminated or torn in the tensile test method, the maximum tensile force value at this time is recorded and used as the first index value. At the same time, the failure morphology of the delamination position is manually observed, and the corresponding second index value is evaluated based on the preset evaluation rules. For each trial production batch of composite products, 5-10 samples are randomly selected for testing to obtain multiple firmness evaluation scores (i.e., the first index value and the second index value). At the same time, reasonable weights are assigned to different first index values ​​and second index values, for example, the weight of the first index value is set to 0.7, and the weight of the second index value is set to 0.3. The first mean of all first index values ​​and the second mean of all second index values ​​are calculated, and then the first mean and the second mean are weighted and fused based on the corresponding weights to calculate the comprehensive evaluation score.

[0054] In some embodiments, the training intensity of the secondary training obtained by matching the comprehensive evaluation scores includes: obtaining a preliminary training intensity of the secondary training according to the comprehensive evaluation scores, the preliminary training intensity including a first quantity of training data and a first type of quantity of training data; a second type of quantity of omentum composite products involved in a statistical firmness test, obtaining an adjustment value according to matching the second type of quantity, and adjusting the preliminary training intensity to a target training intensity using the adjustment value, the target training intensity including a second quantity of training data and a third type of quantity of training data; wherein the size of the adjustment value is negatively correlated with the second type of quantity.

[0055] In an embodiment of the present invention, a control data set is pre-established, which contains a correspondence between multiple sets of comprehensive evaluation scores and training intensities of secondary training. The above-mentioned preliminary training intensity can be quickly determined by querying the control data set.

[0056] At the same time, the second type number of all mesh composite products involved in the firmness test is also counted, that is, how many types of mesh composite products are involved in the firmness test, such as the three types of double-layer lamination, multi-layer superposition composite, and sandwich composite. Similar to the above, another control data set is also established in advance, which contains a correspondence between multiple sets of second type numbers and adjustment values. By querying the other control data set, the size of the above adjustment value can be quickly determined. Among them, the adjustment value (for example, greater than 1) and the second type number are in a negative correlation relationship, that is, the more types of mesh composite products involved in the firmness test, the closer the corresponding comprehensive evaluation score is to the actual situation (that is, the accurate prediction ability of the composite process parameter optimization model). Correspondingly, the smaller the adjustment value is set at this time, the lower the secondary training intensity of the composite process parameter optimization model is, and the less number and less type of training data are used to train it. On the contrary, it indicates that the greater the deviation between the obtained comprehensive evaluation score and the actual situation, the larger the adjustment value is set, that is, the secondary training intensity of the composite process parameter optimization model is enhanced.

[0057] Therefore, the present invention dynamically determines the training intensity of the composite process parameter optimization model through the second type number of omentum composite products involved in the firmness test, thereby achieving a balance between training effect and training load.

[0058] In some embodiments, the obtaining of an adjustment value based on the matching of the second type of quantity and using the adjustment value to adjust the preliminary training intensity to a target training intensity, wherein the target training intensity includes a second quantity of training data and a third type of quantity of training data, comprises: obtaining a first adjustment value and a second adjustment value based on the matching of the second type of quantity, using the first adjustment value to adjust the first quantity of training data in the preliminary training intensity to the second quantity, and using the second adjustment value to adjust the first type of quantity of training data in the preliminary training intensity to the third type of quantity.

[0059] In an embodiment of the present invention, based on the aforementioned scheme, the present invention further sets an adjustment value including a first adjustment value and a second adjustment value, both of which are based on another different reference data set, and use the first adjustment value and the second adjustment value to adjust the first quantity of training data and the first type of training data in the preliminary training intensity respectively.

[0060] like Figure 3 As shown, the present invention also discloses a production control system for mesh composite products, the system includes at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, and when the computer program is executed by the processor, the following steps are executed: constructing a plurality of training data based on actual production record data of mesh composite products, constructing them into a training data set, and using the training data set to train a composite process parameter optimization model; obtaining first type information of the hot melt adhesive mesh used in the gluing and compounding stages, and taking a high-resolution image of the hot melt adhesive mesh, and extracting the mesh structure of the hot melt adhesive mesh according to the high-resolution image; The method comprises the following steps: obtaining the second type information of the substrate, the substrate thickness and the substrate composite method used in the gluing and composite stages, wherein the second type information, the substrate thickness and the substrate composite method constitute the substrate information; using the trained composite process parameter optimization model to analyze and process the web information and the substrate information, and obtaining the first composite process parameter predicted by the composite process parameter optimization model, wherein the first composite process parameter includes a composite roller temperature range and a composite roller pressure range; and controlling the composite roller to composite the hot melt adhesive web and the substrate according to the composite roller temperature range and the composite roller pressure range.

[0061] An embodiment of the present invention further discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program implements the method described in any of the preceding items when executed.

[0062] An embodiment of the present invention further discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed, the method described in any of the preceding items is implemented.

[0063] The embodiment of the present invention further discloses a computer program product. The computer program product is prepackaged with computer program code. When the computer program code is executed, the method described in any of the above items is implemented.

[0064] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more.

[0065] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0066] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0067] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0068] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0069] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0070] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0071] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A production control method for omentum composite products, characterized in that: The method comprises the following steps: constructing a plurality of training data based on actual production record data of mesh composite products, constructing the data into a training data set, and using the training data set to train a composite process parameter optimization model; obtaining first type information of the hot melt adhesive mesh used in the gluing and compounding stages, and taking a high-resolution image of the hot melt adhesive mesh, extracting mesh structure information of the hot melt adhesive mesh according to the high-resolution image, wherein the first type information and the mesh structure information constitute mesh information; obtaining second type information, substrate thickness and substrate compounding method of the substrate used in the gluing and compounding stages, wherein the second type information, the substrate thickness and the substrate compounding method constitute substrate information; The trained composite process parameter optimization model is used to analyze and process the web information and the substrate information to obtain the first composite process parameter predicted by the composite process parameter optimization model, wherein the first composite process parameter includes a composite roller temperature range and a composite roller pressure range; the composite roller is controlled to composite the hot melt adhesive web and the substrate according to the composite roller temperature range and the composite roller pressure range.

2. The production control method of a mesh composite product according to claim 1, characterized in that: The use of the training data set to train the composite process parameter optimization model includes: using the training data set to perform preliminary training on the composite process parameter optimization model, using the composite process parameter optimization model after preliminary training to conduct several trial productions, that is, using the composite process parameter optimization model to predict the corresponding second composite process parameters, and using the second composite process parameters to perform composite treatment on the trial-produced hot melt adhesive mesh and substrate; performing firmness testing and evaluation on the mesh composite product obtained from the trial production to obtain several firmness evaluation scores, calculating a comprehensive evaluation score based on each of the firmness evaluation scores, and obtaining a training intensity of the secondary training based on the matching of the comprehensive evaluation scores; wherein the training intensity includes the number of training data and the number of types of training data; and performing secondary training on the composite process parameter optimization model according to the training intensity.

3. The production control method of a mesh composite product according to claim 2, characterized in that: The mesh composite product obtained from the trial production is subjected to a firmness test and evaluation to obtain a number of firmness evaluation scores, and a comprehensive evaluation score is calculated based on each of the firmness evaluation scores, including: a firmness test is performed on a sample of the mesh composite product obtained from the trial production, and the maximum tensile force value when the sample breaks is used as a first index value, and a second index value is obtained based on the failure morphology evaluation of the delamination position of the sample; a first mean of each of the first index values ​​is calculated, a second mean of each of the second index values ​​is calculated, and the comprehensive evaluation score is calculated based on the first mean and its first weight, and the second mean and its second weight.

4. The production control method of a mesh composite product according to claim 3, characterized in that: The training intensity of the secondary training obtained by matching the comprehensive evaluation score includes: obtaining a preliminary training intensity of the secondary training according to the comprehensive evaluation score matching, wherein the preliminary training intensity includes a first quantity of training data and a first type of quantity of training data; a second type of quantity of omentum composite products involved in the statistical firmness test, obtaining an adjustment value according to matching the second type of quantity, and adjusting the preliminary training intensity to a target training intensity using the adjustment value, wherein the target training intensity includes a second quantity of training data and a third type of quantity of training data; wherein the size of the adjustment value is negatively correlated with the second type of quantity.

5. The production control method of a mesh composite product according to claim 4, characterized in that: The method of obtaining an adjustment value based on the matching of the second type of quantity and using the adjustment value to adjust the preliminary training intensity to a target training intensity, wherein the target training intensity includes a second quantity of training data and a third type of quantity of training data, includes: obtaining a first adjustment value and a second adjustment value based on the matching of the second type of quantity, using the first adjustment value to adjust the first quantity of training data in the preliminary training intensity to the second quantity, and using the second adjustment value to adjust the first type of quantity of training data in the preliminary training intensity to the third type of quantity.

6. A production control system for omentum composite products, characterized in that: The system includes at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. When the computer program is executed by the processor, the following steps are performed: constructing a plurality of training data based on actual production record data of mesh composite products, constructing the data into a training data set, and using the training data set to train a composite process parameter optimization model; obtaining first type information of a hot melt adhesive mesh used in the gluing and compounding stages, taking a high-resolution image of the hot melt adhesive mesh, extracting mesh structure information of the hot melt adhesive mesh based on the high-resolution image, wherein the first type information and the mesh structure information constitute mesh information; obtaining second type information, substrate thickness, and substrate compounding method of a substrate used in the gluing and compounding stages, wherein the second type information, the substrate thickness, and the substrate compounding method constitute substrate information; The trained composite process parameter optimization model is used to analyze and process the web information and the substrate information to obtain the first composite process parameter predicted by the composite process parameter optimization model, wherein the first composite process parameter includes a composite roller temperature range and a composite roller pressure range; the composite roller is controlled to composite the hot melt adhesive web and the substrate according to the composite roller temperature range and the composite roller pressure range.

7. The production control system of a mesh composite product according to claim 6, characterized in that: The use of the training data set to train the composite process parameter optimization model includes: using the training data set to perform preliminary training on the composite process parameter optimization model, using the composite process parameter optimization model after preliminary training to conduct several trial productions, that is, using the composite process parameter optimization model to predict the corresponding second composite process parameters, and using the second composite process parameters to perform composite treatment on the trial-produced hot melt adhesive mesh and substrate; performing firmness testing and evaluation on the mesh composite product obtained from the trial production to obtain several firmness evaluation scores, calculating a comprehensive evaluation score based on each of the firmness evaluation scores, and obtaining a training intensity of the secondary training based on the matching of the comprehensive evaluation scores; wherein the training intensity includes the number of training data and the number of types of training data; and performing secondary training on the composite process parameter optimization model according to the training intensity.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program implements the method according to any one of claims 1 to 5 when executed.

9. A computer storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method according to any one of claims 1 to 5 is implemented.

10. A computer program product, characterized in that: The computer program product is prepackaged with computer program codes, and when the computer program codes are executed, the method according to any one of claims 1 to 5 is implemented.

Images