Intelligent optimization method and system for BOPP (Biaxially-oriented Polypropylene) film making process parameters

Abstract

The invention relates to an intelligent optimization method and system for BOPP film making process parameters, and the method comprises the following steps: obtaining historical data of BOPP film making, including process parameters and corresponding product performance data; constructing a product performance control model by taking the process parameters as input and the product performance data as output, formulating a control strategy according to requirements, and solving to obtain a first process parameter through a multi-objective optimization algorithm; preparing a BOPP film by using the first process parameter, and obtaining equipment operation data; constructing an equipment operation control model by taking the first process parameter as input and the equipment operation data as output, and solving to obtain a second process parameter; and producing the BOPP film by using the second process parameters. According to the intelligent optimization method and system for the BOPP film manufacturing process parameters, the process parameters of two dimensions of product performance and equipment operation are optimized, so that the optimal operation state of the equipment is kept to the greatest extent while high product quality is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of BOPP film-making processes, and in particular to an intelligent optimization method and system for BOPP film-making process parameters. Background Art

[0002] In traditional BOPP (biaxially oriented polypropylene) film-making production practices, determining process parameters has always faced many difficulties. In the past, it mainly relied on operators to grope for appropriate parameters based on past experience or repeated tests, with the following significant deficiencies:

[0003] Obvious efficiency restriction: Relying on manual experience to adjust parameters, each trial-and-error process is extremely time-consuming, making the entire production preparation stage long, seriously slowing down the production rhythm, and making it difficult to improve the production efficiency of BOPP film-making.

[0004] Difficult quality control: The experience of different operators varies, and the judgment of the same operator also fluctuates at different times. This makes the quality of BOPP film products produced unstable and difficult to maintain stably at a high quality level. Key quality indicators such as the thickness uniformity and tensile strength of the film are prone to deviation.

[0005] Lack of comprehensive consideration: When determining process parameters in the past, most were one-sidedly concerned with a certain angle, such as simply emphasizing a certain aspect like the yield rate or quality. It was impossible to comprehensively weigh from key dimensions such as the optimal speed, optimal quality, and optimal yield rate at the same time, and it was difficult to accurately select the optimal process parameter combination that truly meets the production requirements of BOPP film-making.

[0006] With the development of data analysis technology, some enterprises have begun to try to apply data analysis to the optimization of BOPP film-making process parameters, but these applications are often relatively limited. For example, only simple correlation analysis is performed on some process parameters and product performance data to find parameters that have a significant impact on product quality or production efficiency. However, this analysis lacks systematicness and depth, fails to fully explore the internal relationships between parameters, and does not consider more dimensions such as energy consumption and equipment life, and cannot achieve comprehensive parameter optimization and multi-objective comprehensive decision-making.

[0007] This preliminary application is often not deep and comprehensive enough, unable to dynamically recommend the optimal process parameter combination according to different product types, order requirements, and production stages, and it is difficult to meet the enterprise's demand for further improving production efficiency. Summary of the Invention

[0008] To this end, the technical problem to be solved by the present invention is to overcome the deficiencies existing in the BOPP film-making process in the prior art, and to provide an intelligent optimization method and system for BOPP film-making process parameters, which optimize the process parameters from the dimensions of product performance data and equipment operation data in sequence, and can ensure that the equipment is in the best working state as much as possible on the premise of ensuring a high product production level.

[0009] To solve the above technical problem, the present invention provides an intelligent optimization method for BOPP film-making process parameters, including the following steps:

[0010] Obtain the historical data of BOPP film-making, including: various process parameters involved in the film-making process and the product performance data corresponding to the process parameters;

[0011] Taking various process parameters involved in the historical data as input parameters and taking the product performance data as output, construct a product performance control model, wherein: the product performance data includes multiple product performance evaluation indicators;

[0012] Among multiple product performance evaluation indicators, formulate multiple control strategies according to requirements, and solve the process parameters in each control strategy based on a multi-objective optimization algorithm to obtain multiple first process parameters;

[0013] Prepare BOPP films with multiple first process parameters, and obtain the equipment operation data during the film-making process;

[0014] Taking multiple first process parameters as input parameters and taking the equipment operation data as output, construct an equipment operation control model, wherein: the equipment operation data includes multiple equipment operation evaluation indicators;

[0015] Among multiple equipment operation evaluation indicators, taking one of the optimal equipment operation indicators as the target, solve the process parameters of the equipment operation control model to obtain the second process parameter;

[0016] Produce BOPP films with the second process parameter.

[0017] In an embodiment of the present invention, the product performance data includes multiple product performance evaluation indicators, and the product performance evaluation indicators at least include: product quality, product production speed, and product yield.

[0018] In an embodiment of the present invention, among multiple product performance evaluation indicators, formulating multiple control strategies according to requirements, and solving the process parameters in each control strategy based on a multi-objective optimization algorithm to obtain multiple first process parameters includes:

[0019] Aiming at the optimal product quality, variable optimization is carried out on the product performance control model through a neural network algorithm to obtain a process parameter, including:

[0020] Traverse various process parameters and corresponding product quality data in historical data;

[0021] Design a neural network model to simulate the relationship between process parameters and product quality data, and train the neural network model with process parameters as input parameters and product quality data as output parameters;

[0022] Define an objective function to maximize a certain quality index of a specific product, and use a global search algorithm to find the optimal solution based on the trained model.

[0023] In an embodiment of the present invention, among multiple product performance evaluation indicators, multiple control strategies are formulated according to requirements, and the process parameters in each control strategy are solved based on a multi-objective optimization algorithm to obtain multiple first process parameters, including:

[0024] Aiming at the optimal product production speed, variable optimization is carried out on the product performance control model through a neural network algorithm to obtain a process parameter, including:

[0025] According to the time-related data and corresponding process parameters in the BOPP film production records, calculate the production speed under different parameter settings;

[0026] Design a neural network model to simulate the relationship between process parameters and product production speed, and train the neural network model with process parameters as input parameters and product production speed as output parameters;

[0027] Use the neural network model to screen out the process parameter combination that can make the production process turn around the fastest on the premise of ensuring the quality of BOPP film products;

[0028] Apply a multi-objective optimization algorithm, combined with the trained neural network model, set the finished product rate of the product as a constraint condition, and find the optimal solution based on the trained model.

[0029] In an embodiment of the present invention, among multiple product performance evaluation indicators, multiple control strategies are formulated according to requirements, and the process parameters in each control strategy are solved based on a multi-objective optimization algorithm to obtain multiple first process parameters, including:

[0030] Aiming at the optimal product finished product rate, variable optimization is carried out on the product performance control model through a neural network algorithm to obtain a process parameter, including:

[0031] Traverse various process parameters and corresponding product finished product rate data in historical data;

[0032] Design a neural network model to simulate the relationship between process parameters and product yield data. Use the process parameters as input parameters and the product yield data as output parameters to train the neural network model.

[0033] Utilize the neural network model to set up a multi-channel calculation model according to the process steps involved in the preparation process, and analyze the combination of process parameters that can maximize the production yield in each process step.

[0034] Apply a multi-objective optimization algorithm, combined with the trained neural network model, set the production speed of the product as a constraint condition, and find the optimal solution based on the trained model.

[0035] In an embodiment of the present invention, the device operation data includes multiple device operation evaluation indicators, and the device operation evaluation indicators at least include: device energy consumption and device life.

[0036] In an embodiment of the present invention, among the multiple device operation evaluation indicators, with the optimal device energy consumption as the goal, solve the process parameters of the device operation control model to obtain the second process parameters, including:

[0037] Traverse the first process parameters and the corresponding device energy consumption data in the historical data;

[0038] Design a neural network model to simulate the relationship between the first process parameters and the device energy consumption. Use the first process parameters as input parameters and the device energy consumption as output parameters to train the neural network model.

[0039] Define an objective function to minimize the overall device energy consumption. Combine the trained neural network model, set the device life as a constraint condition, and find the optimal solution based on the trained model.

[0040] In an embodiment of the present invention, among the multiple device operation evaluation indicators, with the optimal device life as the goal, solve the process parameters of the device operation control model to obtain the second process parameters, including:

[0041] Traverse the first process parameters and the corresponding device life data in the historical data;

[0042] Design a neural network model to simulate the relationship between the first process parameters and the device life. Use the first process parameters as input parameters and the device life as output parameters to train the neural network model.

[0043] Utilize the neural network model to set up a multi-channel calculation model according to the process steps involved in the preparation process, and analyze the combination of the first process parameters that can maximize the device life in each process step.

[0044] Apply a multi-objective optimization algorithm, combined with the trained neural network model, set the device energy consumption as a constraint condition, and find the optimal solution based on the trained model.

[0045] In an embodiment of the present invention, after the production of BOPP films with the second process parameters, it further includes real-time collecting product performance data and device operation data, and updating the second process parameters according to the collection results.

[0046] The above technical solution of the present invention has the following advantages compared with the prior art:

[0047] For the intelligent optimization method of BOPP film-making process parameters of the present invention, first, make full use of the process parameters and corresponding product performance result data in the historical process of BOPP film-making. By means of machine learning, accurately recommend the optimal process parameter combination from the dimension of product performance data, thereby effectively improving the production efficiency of BOPP film-making, stabilizing the product quality, and ensuring that the finished product rate reaches a better level; while ensuring a relatively high product production level, then use the obtained process parameters and corresponding device operation data, and by means of machine learning, further optimize the process parameters from the dimension of device operation data to ensure that the device can also be in the best working state of long-term stability.

[0048] Compared with the prior art, it abandons the inefficient way of traditional manual trial and error, quickly locks the optimal parameters through intelligent recommendation, greatly shortens the production preparation time of BOPP film-making, speeds up the entire BOPP film-making production, and improves the output efficiency of BOPP films per unit time; recommends parameters based on scientific data analysis, avoids the instability and subjectivity interference of human experience, enables the product quality of the produced BOPP films to be stably maintained at a relatively high level, and reduces quality problems such as uneven film thickness and insufficient strength caused by parameter fluctuations; moreover, it not only controls the parameters from the product dimension, but also optimizes the parameters from the device dimension, overcomes the limitations of single-dimension decision-making in the past, makes the selection of BOPP film-making process parameters more scientific and reasonable, and comprehensively helps the BOPP film-making production reach the best state. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to make the content of the present invention easier to be clearly understood, the following further details the present invention according to the specific embodiments of the present invention and in combination with the accompanying drawings, wherein:

[0050] Figure 1 is the step flow chart of the intelligent optimization method of BOPP film-making process parameters of the present invention;

[0051] Figure 2 is the step flow chart of obtaining multiple first process parameters by using the method of the present invention;

[0052] Figure 3 It is a flowchart of the steps to obtain multiple second process parameters by using the method of the present invention;

[0053] Figure 4 It is a structural framework diagram of the intelligent optimization system for BOPP film-making process parameters of the present invention. Specific embodiments

[0054] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it, but the specific embodiments cited are not intended to limit the present invention.

[0055] Embodiment 1

[0056] Referring to Figure 1 as shown, the present invention discloses an intelligent optimization method for BOPP film-making process parameters, including the following steps:

[0057] S10. Obtain historical data of BOPP film-making, including: various process parameters involved in the film-making process and product performance data corresponding to the process parameters;

[0058] Specifically, the film-making process of BOPP film is a complex and precise process, mainly including the following key process steps: extrusion, cooling, longitudinal stretching, transverse stretching, heat setting treatment, cooling and winding. In this process, each process step has corresponding process parameters to guide the operation. For example: extrusion temperature, stretching ratio, traction speed, cooling temperature. And, the characteristics of different raw materials of polypropylene are also different. For example: parameters such as molecular weight distribution, isotacticity, etc.; corresponding to different process parameters, the quality of the prepared BOPP film is also different. Corresponding to the product performance result data of the produced BOPP film collected, such as indicators such as thickness uniformity, tensile strength, elongation at break, haze, gloss, and yield of the BOPP film.

[0059] Specifically, in order to comprehensively evaluate the product performance, it is necessary to collect the parameters from the raw materials to each process step as described above, enrich the data dimension as much as possible, and ensure that there is sufficient and high-quality data as the basis to construct an accurate and reliable prediction model. The quality of the data directly affects the effectiveness of all subsequent steps.

[0060] S20. Take various process parameters involved in the historical data as input parameters, and construct a product performance control model with the product performance data as the output, where: the product performance data includes multiple product performance evaluation indicators;

[0061] Specifically, by constructing a model to determine the correlation between process parameters and product performance, and selecting an appropriate machine learning or deep learning model (such as neural network, support vector machine, etc.) according to specific problems, historical data can be divided into a training set, a test set, and a validation set. The model is trained using historical data, and the performance of the model is evaluated through techniques such as cross-validation to ensure its generalization ability. Establishing a model that can accurately predict product quality under different process conditions is the core of the entire optimization process, providing a theoretical basis and technical means for subsequent process parameter optimization.

[0062] S30. Among multiple product performance evaluation indicators, formulate multiple control strategies according to requirements, solve the process parameters in each control strategy based on a multi-objective optimization algorithm, and obtain multiple first process parameters.

[0063] Specifically, in the actual production process, various BOPP films of different qualities are generated based on different requirements. For example, some customers need high-quality BOPP films, so emphasis should be placed on the quality of film production, and every process parameter from raw materials to control needs to ensure the optimal solution. Some customers need BOPP films urgently, so emphasis should be placed on the speed of film production. On the premise of ensuring qualified quality, the speed of each production process should be increased as soon as possible. In this way, when training the model, multiple control strategies need to be formulated. Based on enterprise requirements (such as emphasizing quality, speed, or yield rate), weights are assigned to each product performance evaluation indicator to form a comprehensive objective function. Methods such as genetic algorithms and particle swarm optimization are used to find the combination of process parameters that optimizes the objective function under given constraints, and multiple first process parameters can be obtained corresponding to different strategies.

[0064] S40. Prepare BOPP films with multiple first process parameters and obtain the equipment operation data during the film production process.

[0065] Specifically, the BOPP films prepared with the first process parameters are products that can meet the requirements and can ensure the quality of the products. On this basis, sensors are set to monitor the operation parameters of each device during the production process, such as the energy consumption of the device, the operating temperature of the device, and the wear of the device. Recording and analyzing these data can understand the impact of the first process parameters on the device and provide a basis for subsequent modeling.

[0066] S50. Construct an equipment operation control model with multiple first process parameters as input parameters and equipment operation data as output, where: the equipment operation data includes multiple equipment operation evaluation indicators.

[0067] Specifically, by constructing a model, determine the correlation between process parameters and equipment operation, and train the model using the collected data so that it can predict the device performance under specific process settings.

[0068] S60. Among the operation evaluation indicators of multiple devices, taking one of the optimal device operation indicators as the target, solve the process parameters of the device operation control model to obtain the second process parameters;

[0069] Specifically, in the actual production process, the usage degree of the device varies based on different requirements. For example, the impact of different extrusion temperatures and cooling temperatures on the overall energy consumption is different, and the impact of different draw ratios and traction speeds on the wear of device components such as the extruder screw and stretching roller is different. Select one or more key device operation evaluation indicators (such as the lowest energy consumption, the longest service life) as the optimization target, and use an optimization algorithm to find the process parameter configuration that makes the selected device operation indicator reach the optimal under the premise of meeting the product quality requirements.

[0070] S70. Produce BOPP films with the second process parameters;

[0071] Specifically, after producing BOPP films with the second process parameters, it also includes real-time collection of product performance data and device operation data, and updating the second process parameters according to the collection results to ensure that it is always in the best state.

[0072] Specifically, the product performance data includes multiple product performance evaluation indicators. In the embodiment, according to the actual requirements, set the product performance evaluation indicators to at least include: product quality, product production speed, and product yield. If there are other product performance evaluation indicators, they can also be expanded according to the actual situation. In this embodiment, the above three product performance evaluation indicators are used for illustration to obtain different first process parameters.

[0073] Refer to Figure 2 As shown, among one of the product performance evaluation indicators, taking the optimal product quality as the target, perform variable optimization on the product performance control model through the neural network algorithm to obtain a process parameter, including:

[0074] Traverse the various process parameters and corresponding product quality data in the historical data; in this embodiment, only collect the process parameters that determine the product quality, involve screening of the process parameters, clean the collected data, remove outliers or incomplete records. Here, mainly remove the process parameters that have little impact on the product quality to ensure the quality and integrity of the data, and perform normalization processing on the process parameters of different magnitudes for subsequent model training.

[0075] A neural network model is designed to simulate the relationship between process parameters and product quality data, and the neural network model is trained with process parameters as input parameters and product quality data as output parameters. A suitable neural network architecture is selected according to the complexity of the problem and the data characteristics. In this embodiment, due to the existence of multiple parameters of multiple processes, a multi-layer perceptron is more appropriate. The preprocessed historical data is divided into a training set, a validation set and a test set, usually in a ratio of 70% training set, 15% validation set and 15% test set.

[0076] Define an objective function that maximizes a certain quality indicator of a specific product, and use a global search algorithm to find the optimal solution based on the trained model. For example, if the focus is on the transparency of the product, the objective function can be set to maximize the transparency score, and other relevant product quality indicators can be used as constraints to ensure that other important attributes are not affected while pursuing the main goal; run a global search algorithm, including: genetic algorithm (GA), particle swarm optimization (PSO) or Bayesian optimization, until the best process parameter combination that meets the termination conditions (such as reaching a predetermined number of iterations or the improvement is lower than a threshold) is found.

[0077] Reference Figure 2 As shown, in one of the product performance evaluation indicators, taking the optimal product production speed as the goal, the product performance control model is optimized by the neural network algorithm to obtain a process parameter, including:

[0078] Calculate the production speed under different parameter settings based on the time-related data and corresponding process parameters in the BOPP film production records;

[0079] A neural network model is designed to simulate the relationship between process parameters and product production speed. The neural network model is trained with process parameters as input parameters and product production speed as output parameters.

[0080] The neural network model is used to screen out the process parameter combination that can make the production process turnover the fastest under the premise of ensuring that the quality of the BOPP film products meets the standards. In this embodiment, it is necessary to ensure that the quality of the BOPP film products meets the standards and screen out the process parameter combination that can achieve the fastest production speed. Combined with the neural network model of product quality, check whether each predicted process parameter combination meets the pre-set quality standards (such as transparency, strength, etc.), and retain those parameter combinations that meet the quality requirements and provide a higher production speed.

[0081] It should be noted that in this embodiment, a multi-objective optimization algorithm is applied. Combining with the trained neural network model, the product yield is set as a constraint condition, and the optimal solution is searched based on the trained model. Specifically, when training the neural network model, the objective function of multi-objective optimization is defined, where the main objective is to minimize the production time (i.e., maximize the production speed), and at the same time, the product yield is set as a constraint condition.

[0082] Referring to Figure 2 As shown, in one of the product performance evaluation indicators, aiming at the optimal product yield, variable optimization is performed on the product performance control model through the neural network algorithm, and a process parameter is obtained, including:

[0083] Traverse all kinds of process parameters and corresponding product yield data in the historical data.

[0084] Design a neural network model to simulate the relationship between process parameters and product yield data. Use the process parameters as input parameters and the product yield data as output parameters to train the neural network model.

[0085] Using the neural network model, set up a multi-channel calculation model according to the process steps involved in the preparation process. Analyze the combination of process parameters that can achieve the highest production yield in each process step. In the actual production process, there is a yield detection for each process step. In order to deeply analyze the influence of process parameters in each specific process step on the yield, the entire production process is decomposed into multiple independent but interrelated sub-processes (i.e., "channels"), which can more accurately identify and optimize the key process parameters at each stage, thereby improving the overall yield. The specific process is as follows:

[0086] Decompose the production process steps, and regard each production process step as an independent "channel". For each process step, design an independent calculation model, which aims to simulate the relationship between process parameters and yield under a specific process step. For each channel, use historical data to train the corresponding model. Based on the trained model, analyze the best combination of process parameters for each process step, analyze the influence weight of each channel on the yield, comprehensively consider the results of all channels, and configure the globally optimal process parameters according to the influence weight.

[0087] It should be noted that in this embodiment, a multi-objective optimization algorithm is applied. Combining with the trained neural network model, the production speed of the product is set as a constraint condition, and the optimal solution is searched based on the trained model.

[0088] Specifically, the device operation data includes multiple device operation evaluation indicators. In the embodiment, according to actual requirements, the device operation evaluation indicators are set to at least include: device energy consumption and device life. If there are other device operation evaluation indicators, they can also be expanded according to the actual situation. In this embodiment, the above two device operation evaluation indicators are used for illustration to obtain different second process parameters.

[0089] Refer to Figure 3 As shown, in one embodiment, with the goal of the optimal energy consumption of the device, the process parameters of the device operation control model are solved to obtain the second process parameters, including:

[0090] Traverse the first process parameters and the corresponding device energy consumption data in the historical data.

[0091] Design a neural network model to simulate the relationship between the first process parameters and the device energy consumption. Use the first process parameters as input parameters and the device energy consumption as output parameters to train the neural network model; in this embodiment, the products prepared by the first process parameters can meet the performance requirements of the products. On this basis, collect the process parameters that affect the device energy consumption in the first process parameters and conduct a secondary screening in the first process parameters.

[0092] Define the objective function of minimizing the overall device energy consumption. Combine the trained neural network model, set the device life as the constraint condition, and find the optimal solution based on the trained model.

[0093] Refer to Figure 3 As shown, in another embodiment, with the goal of the optimal life of the device, the process parameters of the device operation control model are solved to obtain the second process parameters, including:

[0094] Traverse the first process parameters and the corresponding device life data in the historical data;

[0095] Design a neural network model to simulate the relationship between the first process parameters and the device life. Use the first process parameters as input parameters and the device life as output parameters to train the neural network model;

[0096] Use the neural network model to set up a multi-channel calculation model according to the process steps involved in the preparation process, and analyze the combination of the first process parameters that can make the device life the longest in each process step;

[0097] Apply the multi-objective optimization algorithm. Combine the trained neural network model, set the device energy consumption as the constraint condition, and find the optimal solution based on the trained model.

[0098] The present invention provides an intelligent optimization method for BOPP film-making process parameters. By analyzing historical data and constructing a product performance control model, it can accurately identify the key process parameters affecting product performance, provides at least three product performance evaluation indicators, and uses a multi-objective optimization algorithm to find the best compromise solution to ensure optimal product performance while meeting all quality standards. On this basis, by constructing an equipment operation control model, it can predict and optimize the performance of the equipment under different process conditions. The optimized process parameters not only focus on product quality and production efficiency but also pay attention to reducing energy consumption, which is of great significance for energy conservation and emission reduction. At the same time, preventive maintenance measures are taken to avoid excessive wear and extend the service life of the equipment.

[0099] The intelligent optimization method for process parameters of the present invention allows for dynamically adjusting the optimization strategy according to changes in market or customer demands. For example, when the market demands high-end products, it focuses on quality optimization, while under normal demands, it pays more attention to cost-effectiveness. Due to its high flexibility and adaptability, this method can support diverse production demands and meet the requirements of different customer groups.

[0100] Example Two

[0101] Based on the above Example One, the present invention also discloses an intelligent optimization system for BOPP film-making process parameters, which can execute the above method. As shown in Figure 4 it includes:

[0102] A data acquisition unit for acquiring historical data of BOPP film-making, where the historical data at least includes various process parameters involved in the film-making process and product performance data corresponding to the process parameters;

[0103] A product performance control model construction unit connected to the data acquisition unit for constructing a product performance control model with various process parameters in the historical data as input parameters and product performance data as output parameters, where the product performance data includes multiple product performance evaluation indicators;

[0104] A first process parameter solving unit connected to the product performance control model construction unit for solving the process parameters in the product performance control model with the optimal product performance evaluation indicators as the objectives respectively among multiple product performance evaluation indicators to obtain multiple first process parameters;

[0105] An equipment operation data acquisition unit for preparing BOPP films according to the multiple first process parameters and acquiring equipment operation data during the film-making process;

[0106] The device operation control model construction unit is connected to the device operation data acquisition unit and is used to construct a device operation control model with the multiple first process parameters as input parameters and the device operation data as output parameters, where the device operation data includes multiple device operation evaluation indicators;

[0107] The second process parameter solving unit is connected to the device operation control model construction unit and is used to select one of the optimal device operation indicators from the multiple device operation evaluation indicators as the target, solve the process parameters of the device operation control model, and obtain the second process parameter;

[0108] The production execution unit is used to produce BOPP films according to the second process parameter.

[0109] Specifically, for the intelligent optimization system of BOPP film-making process parameters of the present invention, the specific process of implementing the above method refers to Embodiment 1 and will not be repeated here.

[0110] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of completely hardware embodiments, completely software embodiments, or embodiments combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0111] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, so that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0112] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide for implementing the steps in a process Figure 1 one process or multiple processes and / or blocks Figure 1 or steps of functions specified in multiple blocks.

[0114] Obviously, the above embodiments are only examples for clear illustration and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all the implementation manners here. And the obvious changes or modifications derived therefrom still fall within the protection scope of the present invention.

Claims

1. A method for intelligent optimization of BOPP film making process parameters, characterized in that: The following steps are involved: Obtain historical data of BOPP film production, including: various process parameters involved in the film production process and product performance data corresponding to the process parameters; A product performance control model is constructed using various process parameters involved in historical data as input parameters and product performance data as output, wherein: the product performance data includes multiple product performance evaluation indicators; Among multiple product performance evaluation indicators, multiple control strategies are formulated according to needs, and the process parameters in each control strategy are solved based on a multi-objective optimization algorithm to obtain multiple first process parameters; A BOPP film is prepared using a plurality of first process parameters, and equipment operation data in the film-making process is obtained; A device operation control model is constructed with a plurality of first process parameters as input parameters and device operation data as output, wherein: the device operation data includes a plurality of device operation evaluation indicators; Among the multiple equipment operation evaluation indicators, taking one of the optimal equipment operation indicators as a target, solving the process parameters of the equipment operation control model to obtain a second process parameter; The BOPP film is produced with the second process parameters.

2. The method for intelligent optimization of BOPP film making process parameters according to claim 1, characterized in that: The product performance data includes a plurality of product performance evaluation indicators, and the product performance evaluation indicators include at least: product quality, product production speed and product yield rate.

3. The method for intelligent optimization of BOPP film making process parameters according to claim 2, characterized in that: Among multiple product performance evaluation indicators, multiple control strategies are formulated according to demand, and the process parameters in each control strategy are solved based on the multi-objective optimization algorithm to obtain multiple first process parameters, including: With the goal of optimal product quality, the product performance control model is optimized by using a neural network algorithm to obtain a process parameter, including: Go through the historical data, various process parameters and corresponding product quality data; Design a neural network model to simulate the relationship between process parameters and product quality data, and train the neural network model with process parameters as input parameters and product quality data as output parameters; Define the objective function of maximizing a certain quality indicator of a specific product, and use a global search algorithm to find the optimal solution based on the trained model.

4. The method for intelligent optimization of BOPP film making process parameters according to claim 2, characterized in that: Among multiple product performance evaluation indicators, multiple control strategies are formulated according to demand, and the process parameters in each control strategy are solved based on the multi-objective optimization algorithm to obtain multiple first process parameters, including: With the optimal product production speed as the goal, the product performance control model is optimized by using a neural network algorithm to obtain a process parameter, including: Calculate the production speed under different parameter settings based on the time-related data and corresponding process parameters in the BOPP film production records; Design a neural network model to simulate the relationship between process parameters and product production speed, and train the neural network model with process parameters as input parameters and product production speed as output parameters; Use the neural network model to screen out the process parameter combination that can make the production process turn around the fastest while ensuring that the BOPP film product quality meets the standards; Apply the multi-objective optimization algorithm, combine it with the trained neural network model, set the product yield as a constraint, and find the optimal solution based on the trained model.

5. The method for intelligent optimization of BOPP film making process parameters according to claim 2, characterized in that: Among multiple product performance evaluation indicators, multiple control strategies are formulated according to demand, and the process parameters in each control strategy are solved based on the multi-objective optimization algorithm to obtain multiple first process parameters, including: With the goal of optimal product yield, the product performance control model is optimized by using a neural network algorithm to obtain a process parameter, including: Go through the historical data, various process parameters and corresponding product yield data; Design a neural network model to simulate the relationship between process parameters and product yield data, and train the neural network model with process parameters as input parameters and product yield data as output parameters; Using the neural network model, a multi-channel calculation model is set up according to the process steps involved in the preparation process, and the process parameter combination that can achieve the highest production yield in each process step is analyzed; Apply the multi-objective optimization algorithm, combine it with the trained neural network model, set the product production speed as the constraint, and find the optimal solution based on the trained model.

6. The method for intelligent optimization of BOPP film making process parameters according to claim 1, characterized in that: The equipment operation data includes a plurality of equipment operation evaluation indicators, and the equipment operation evaluation indicators at least include: equipment energy consumption and equipment life.

7. The method for intelligent optimization of BOPP film making process parameters according to claim 6, characterized in that: Among multiple equipment operation evaluation indicators, taking the optimal energy consumption of the equipment as the goal, solving the process parameters of the equipment operation control model to obtain the second process parameters, including: Traverse the historical data, the first process parameter and the corresponding equipment energy consumption data; Designing a neural network model to simulate the relationship between the first process parameter and the equipment energy consumption, and training the neural network model with the first process parameter as an input parameter and the equipment energy consumption as an output parameter; Define the objective function of minimizing the overall equipment energy consumption, combine it with the trained neural network model, set the equipment life as a constraint, and find the optimal solution based on the trained model.

8. The method for intelligent optimization of BOPP film making process parameters according to claim 6, characterized in that: Among the multiple equipment operation evaluation indicators, taking the optimal life of the equipment as the goal, solving the process parameters of the equipment operation control model to obtain the second process parameters, including: Traverse the historical data, the first process parameter and the corresponding equipment life data; Designing a neural network model to simulate the relationship between the first process parameter and the equipment life, and training the neural network model with the first process parameter as an input parameter and the equipment life as an output parameter; Using the neural network model, a multi-channel calculation model is set up according to the process steps involved in the preparation process to analyze the first process parameter combination that can make the equipment life longest in each process step; Apply the multi-objective optimization algorithm, combine it with the trained neural network model, set the equipment energy consumption as the constraint condition, and find the optimal solution based on the trained model.

9. The method for intelligent optimization of BOPP film making process parameters according to claim 1, characterized in that: After the BOPP film is produced with the second process parameters, the method also includes real-time collection of product performance data and equipment operation data, and updating the second process parameters according to the collection results.

10. A BOPP film making process parameter intelligent optimization system, characterized in that: include: A data acquisition unit, used to acquire historical data of BOPP film making, wherein the historical data at least includes various process parameters involved in the film making process and product performance data corresponding to the process parameters; A product performance control model building unit, connected to the data acquisition unit, for building a product performance control model using various process parameters in the historical data as input parameters and product performance data as output parameters, wherein the product performance data includes a plurality of product performance evaluation indicators; A first process parameter solving unit, connected to the product performance control model building unit, is used to solve the process parameters in the product performance control model with the optimal product performance evaluation index as the target among multiple product performance evaluation indexes, so as to obtain multiple first process parameters; An equipment operation data acquisition unit, used for preparing the BOPP film according to the plurality of first process parameters and acquiring equipment operation data during the film making process; an equipment operation control model building unit, connected to the equipment operation data acquisition unit, for building an equipment operation control model using the plurality of first process parameters as input parameters and the equipment operation data as output parameters, wherein the equipment operation data includes a plurality of equipment operation evaluation indicators; A second process parameter solving unit, connected to the equipment operation control model building unit, is used to select one of the optimal equipment operation indicators from multiple equipment operation evaluation indicators as a target, solve the process parameters of the equipment operation control model, and obtain a second process parameter; A production execution unit is used to produce the BOPP film according to the second process parameters.

Images