An intelligent monitoring outdoor HAA-cured polyester resin production quality management system and method

Abstract

The present application discloses an intelligently monitored outdoor HAA cured polyester resin production quality management system and method, which relates to the technical field of cured polyester resin production. The system collects performance data and monomer ratio data, curing reaction data, and quality data of polyester resin under different application conditions, establishes a quality relationship mapping model and a mathematical model of a performance evaluation function, obtains a standard performance range under each application condition, calculates a comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, designs an objective function and constraints, and uses the objective function and constraints to calculate an available solution set of monomer ratio data and curing reaction data. The available solution set is used as input data of the quality relationship mapping model, predicts the quality data of the available solution set, and screens out the optimal quality data. The corresponding monomer ratio data and curing reaction data are used as the optimal production data, thereby improving the accuracy of quality control of cured polyester resin production.

Description

Technical Field

[0001] The present application relates to the technical field of cured polyester resin production, and in particular to an intelligently monitored outdoor HAA cured polyester resin production quality management system and method. Background Art

[0002] Outdoor high activity curing (HAA) polyester resin is a special polymer material with the characteristics of fast curing, strong weather resistance, excellent mechanical properties, etc. It is widely used in coatings, boats, automobiles and construction. In order to meet the needs of different application conditions, the formula and preparation process of HAA polyester resin need to be adjusted and optimized.

[0003] The key steps in producing HAA polyester resin include monomer ratio, prepolymerization reaction, curing reaction, etc. Each step has an important impact on the performance of the final product, so strict quality control measures must be implemented. The monomer ratio is the starting point of the entire production process and determines the basic skeleton and performance indicators of the polyester resin. The correct monomer ratio is not only related to the basic properties of the polymer such as molecular weight, molecular weight distribution, and cross-linking density, but also directly affects the curing speed, weather resistance and mechanical properties of the product. The choice of curing reaction temperature and time not only affects the curing speed, but also affects the cross-linking density and the formation of the network structure, which in turn affects the mechanical properties and weather resistance of the polyester resin.

[0004] Traditional quality management methods mainly rely on laboratory testing and manual monitoring. Product samples are taken regularly, and quality is evaluated through physical and chemical tests and performance tests, and process parameters are adjusted accordingly. However, this approach has some obvious defects: manual monitoring and data analysis are inefficient, it is difficult to detect potential quality risk factors in a timely manner, it relies heavily on manual experience and judgment, and lacks scientific process control and quality prediction models.

[0005] In order to solve the above problems, HAA polyester resin manufacturers urgently need to adopt advanced digital quality management systems. Summary of the invention

[0006] The present application aims to solve at least one of the technical problems in the related art to a certain extent. To this end, one purpose of the present application is to propose an intelligent monitoring outdoor HAA cured polyester resin production quality management system and method to improve the accuracy of production quality control of cured polyester resin.

[0007] One aspect of the present application provides an intelligently monitored outdoor HAA-cured polyester resin production quality management method, comprising:

[0008] Step S100: collecting performance data of polyester resin under different application conditions and corresponding monomer ratio data, curing reaction data, and quality data;

[0009] The performance data of the polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data refer to that due to different application conditions, the monomer ratio data and curing reaction data of the polyester resin produced are different, and the corresponding performance data and quality data are different. The performance data include: weather resistance value and mechanical strength value of the polyester resin. The monomer ratio data include component concentration and catalyst type. The curing reaction data include production temperature, production pressure, and reaction time. The monomer ratio data is type-based production data, and the curing reaction data is continuous production data.

[0010] The specific method for collecting the performance data of the polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data is as follows:

[0011] Step S110: installing a temperature sensor, a pressure sensor and a timer on the production line to collect curing reaction data in the curing reaction stage, including production temperature, production pressure and reaction time;

[0012] Step S120: using infrared spectroscopy to analyze the components of the raw materials of the curing polyester resin, obtaining the type of the catalyst, and obtaining its monomer ratio data, including component concentration and catalyst type;

[0013] Step S130: performing quality inspection on the produced cured polyester resin according to its application conditions to obtain corresponding quality data;

[0014] Step S140: obtaining performance data under different application conditions according to the corresponding performance standard of the cured polyester resin required under each application condition;

[0015] Step S200: establishing a quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins;

[0016] The specific method of establishing the quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins is:

[0017] Step S210: Standardize the collected monomer ratio data Z, curing reaction data X and quality data Y, perform one-hot encoding on the monomer ratio data, and convert them into vector form, where Z = (z1, z2, ..., z n ), X=(x1,x2,...,x n ), Y={(y 11 ,y 12 ,...,y 1k ),(y 21 ,y 22 ,...,y2k ),...,(y n1 ,y n2 ,...,y nk )}, wherein there are n groups of monomer ratio data Z, curing reaction data X, and quality data Y, and each group of quality data contains k quality indicators;

[0018] Step S220: constructing a hybrid input deep neural network model as an initial network model of the quality relationship mapping model, and inputting the continuous production data features and the categorized production data features as input data from the input layer;

[0019] Step S230: For continuous production data features, use the fully connected layer as input, and for categorical production data features, use the embedding layer as input;

[0020] Step S240: The hidden layer of the mixed input deep neural network model is connected in series with two convolutional layers to automatically extract the features of the monomer ratio data to obtain the classified production data features, and the long short-term memory network layer is used to capture the features of the curing reaction data to obtain the continuous production data features. The features captured by the convolutional layer and the long short-term memory network layer are fused and input into the fully connected layer for integration, and the predicted quality data is used as the output data, and outputted by the output layer;

[0021] Step S250: using the root mean square error as the loss function and adopting the gradient optimization algorithm to optimize the model parameters, when the error between the predicted output quality data and the actual quality data is the smallest, the model training stops, and a trained quality relationship mapping model is obtained;

[0022] The calculation formula of the loss function is: Among them, n is the number of samples of input data, y ij represents the true value of the jth quality indicator of the quality data in the i-th input data, represents the predicted value of the jth quality indicator of the quality data in the i-th input data;

[0023] Step S300: establishing a mathematical model of a performance evaluation function using quality data, performance data, monomer ratio data and curing reaction data of the polyester resin under different application conditions;

[0024] The specific method of establishing a mathematical model of a performance evaluation function using the quality data, performance data, monomer ratio data and curing reaction data of the polyester resin under different application conditions is:

[0025] Step S310: Define performance index calculation function The functional expression of the performance indicator calculation function is: Among them, aξ represents the ξth application condition, ψ(a ξ ) is the application condition correction function of the ξth application condition, α ξ is the coefficient of the monomer ratio data item under the ξth application condition, γ is the coefficient of the curing reaction data, β j is the coefficient of the jth quality index;

[0026] Step S320: Design application condition correction function ψ(a ξ ), used to adjust the calculation of performance indicators according to different application conditions;

[0027] The application condition correction function ψ(a ξ ) is: Among them, η ξ represents the coefficient of the ξ-th application condition, μ ξ is the phase parameter of the sin function;

[0028] Step S330: Constructing a performance evaluation function f based on the performance indicator calculation function and the application condition correction function c (a ξ ,Z,X,Y), the functional expression of the performance evaluation function is: Among them, λ c is the weight coefficient of the cth performance indicator in the performance data, is the cth performance indicator;

[0029] Step S400: Obtain the standard performance range under each application condition, calculate the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and design the objective function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data;

[0030] The specific method of obtaining the standard performance range under each application condition, calculating the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and designing the objective function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data is as follows:

[0031] Step S410: According to m application conditions A=(a1, a2, ..., a m ), obtain the standard performance range corresponding to the ξth application condition Integrate the standard performance ranges under all application conditions to obtain the standard performance range set

[0032] Step S420: Design an objective function J(Z,X) using the standard performance range, monomer ratio data, curing reaction data and quality data. The functional expression of the objective function is:

[0033] , where F(a ξ , Z, X, Y) is a comprehensive performance evaluation value, which is the sum of the performance evaluation function values ​​of all performance indicators;

[0034] Step S430: setting constraints for the objective function, wherein the constraints include monomer ratio constraints, curing reaction constraints and quality constraints;

[0035] The constraint expression of the monomer ratio constraint is: g1(Z)≤0, indicating that the monomer ratio data Z is within a feasible range less than or equal to 0;

[0036] The constraint expression of the curing reaction constraint is: L X ≤X≤U X , indicating that the curing reaction data X is within the given constraint range, L X is the lower limit of the curing reaction data, U X It is the upper limit of the curing reaction data;

[0037] The constraint expression of the quality constraint is: Indicates that the mass relationship mapping model h(Z,X,Y) is used to ensure that the mass data Y matches the monomer ratio data Z and the curing reaction data, where δ j represents the standard deviation of the jth quality indicator of the current quality data, y j represents the true value of the jth quality indicator of the current quality data, Represents the predicted value of the jth quality indicator of the current quality data;

[0038] The predicted value of the jth quality indicator of the current quality data It is predicted by the quality relationship mapping model;

[0039] Step S500: Calculate the available solution set of monomer ratio data and curing reaction data using the objective function and constraint conditions, use the available solution set as input data of the quality relationship mapping model, predict the quality data of the available solution set and select the optimal quality data from it, and use the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data;

[0040] The specific method of using the objective function and the constraint conditions to calculate the available solution set of monomer ratio data and curing reaction data, using the available solution set as the input data of the quality relationship mapping model, predicting the quality data of the available solution set and screening out the optimal quality data, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data is as follows:

[0041] Step S510: randomly generate N particles, each of which contains curing reaction data, and use the particle swarm optimization algorithm to search for the current optimal curing reaction data X* on the N particles. For the given optimal curing reaction data, use the branch and bound method to obtain the search space of the monomer ratio data [Z min , Z max ], search for the current optimal monomer ratio data Z* in the search space, combine X* and Z*, and obtain an available solution (X*, Z*);

[0042] Step S520: repeatedly searching for the best curing reaction data among the N particles and the best monomer ratio data in the search space until the maximum number of iterations W is reached and the search is stopped to obtain a usable solution set;

[0043] The calculation formula for the maximum number of iterations is: Among them, R x is the dimension of the curing reaction data, R z is the dimension of monomer ratio data, R J is the total number of constraints, s1 is the adjustment coefficient of monomer ratio data and curing reaction data, s2 is the adjustment coefficient of constraint terms, and ε is the expected maximum relative error;

[0044] Step S530: using the available solutions in the available solution set as input data, inputting into the quality relationship mapping model, and predicting and obtaining quality data corresponding to each available solution;

[0045] Step S540: Calculate the comprehensive quality score Q of the quality data corresponding to each available solution;

[0046] The calculation formula of the comprehensive quality score is: Among them, q j is the scoring weight coefficient of the j-th quality indicator, is the jth quality index of the quality data predicted by the available solution;

[0047] Step S550: selecting the quality data with the largest comprehensive quality score as the optimal quality data, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data for producing polyester resin.

[0048] One aspect of the present application provides an intelligently monitored outdoor HAA-cured polyester resin production quality management system, comprising:

[0049] Data collection module, used to collect performance data of polyester resin under different application conditions and corresponding monomer ratio data, curing reaction data, and quality data;

[0050] A quality mapping module is used to establish a quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins;

[0051] A performance evaluation module is used to establish a mathematical model of a performance evaluation function using quality data, performance data, monomer ratio data and curing reaction data of polyester resin under different application conditions;

[0052] The target design module is used to obtain the standard performance range under each application condition, calculate the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and design the target function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data;

[0053] The optimal search module is used to calculate the available solution set of monomer ratio data and curing reaction data using the objective function and constraint conditions, use the available solution set as the input data of the quality relationship mapping model, predict the quality data of the available solution set and screen out the optimal quality data, and use the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data.

[0054] One aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a method for intelligently monitoring the production quality management of outdoor HAA-cured polyester resin are implemented.

[0055] One aspect of the present application provides a readable storage medium storing a computer program, wherein the computer program is suitable for loading into a processor to execute steps in a method for intelligently monitoring outdoor HAA-cured polyester resin production quality management.

[0056] Compared with the prior art, the intelligent monitoring outdoor HAA-cured polyester resin production quality management method proposed in this application has the following advantages:

[0057] The deep learning model learns the complex relationship between monomer ratio data, curing reaction data and quality data, which improves the accuracy of quality index prediction. The automatic feature extraction of the model reduces the dependence on professional knowledge and makes the model more adaptable.

[0058] The performance is evaluated by comprehensively considering application conditions, production parameters and quality data, which ensures the comprehensiveness and accuracy of the evaluation results. The design of the application condition correction function takes into account the different impacts of different application scenarios on performance indicators, which improves the scope of application and flexibility of the model.

[0059] The optimization process ensures that the selection of production parameters meets both quality requirements and performance standards, achieving the optimal configuration of production. The setting of constraint conditions ensures the feasibility and practicality of the solution, ensuring the stability and reliability of the production process.

[0060] By constructing a comprehensive quality scoring function, the quality of each available solution can be comprehensively evaluated based on the performance of all key quality indicators, thereby ensuring that the selected optimal solution meets the quality requirements as a whole;

[0061] This application organically integrates data analysis, mathematical modeling and optimization algorithms to achieve more efficient and accurate production practice guidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A method flow chart of an intelligently monitored outdoor HAA-cured polyester resin production quality management method provided in this application;

[0063] Figure 2 A functional module diagram of an intelligently monitored outdoor HAA-cured polyester resin production quality management system provided in this application;

[0064] Figure 3 A schematic diagram of the structure of an electronic device provided in this application;

[0065] Figure 4 It is a schematic diagram of the structure of a readable storage medium provided by this application. DETAILED DESCRIPTION

[0066] In order to better understand the present application, various aspects of the present application will be described in more detail with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present application and do not limit the scope of the present application in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0067] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in this application, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0068] It should also be understood that expressions such as "including", "comprising", "having", "containing" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present application, "may" is used to mean "one or more embodiments of the present application". And, the term "exemplary" is intended to refer to an example or illustration.

[0069] Unless otherwise specified, all words (including engineering terms and scientific and technological terms) used in this article have the same meaning as those commonly understood by ordinary technicians in the field to which this application belongs. It should also be understood that, unless clearly stated in this application, words defined in common dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0070] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0071] Example 1

[0072] like Figure 1 As shown, a smart monitoring outdoor HAA-cured polyester resin production quality management method provided by the present application includes:

[0073] Step S100: collecting performance data of polyester resin under different application conditions and corresponding monomer ratio data, curing reaction data, and quality data;

[0074] The performance data of the polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data refer to that due to different application conditions, the monomer ratio data and curing reaction data of the polyester resin produced are different, and the corresponding performance data and quality data are different. The performance data include: weather resistance value and mechanical strength value of the polyester resin. The monomer ratio data include component concentration and catalyst type. The curing reaction data include production temperature, production pressure, and reaction time. The monomer ratio data is type-based production data, and the curing reaction data is continuous production data.

[0075] The classified production data refers to data that distinguishes different states, attributes or categories by symbols or names. In the production process of curing polyester resin, the classified production data refers to the composition or type of raw materials or additives;

[0076] The continuous production data refers to the continuous data related to the production conditions. In the production process of curing polyester resin, the continuous production data refers to the production temperature, production pressure and reaction time;

[0077] The performance data refers to different performance indicators of different cured polyester resins;

[0078] The monomer ratio data refers to the component concentration of each monomer type and the type of catalyst required for the catalytic reaction;

[0079] The curing reaction data refers to the temperature, pressure and reaction time set in the curing reaction stage of producing polyester resin;

[0080] The quality data refers to the data obtained by quality testing of the cured polyester resin of the finished product;

[0081] The specific method for collecting the performance data of the polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data is as follows:

[0082] Step S110: installing a temperature sensor, a pressure sensor and a timer on the production line to collect curing reaction data in the curing reaction stage, including production temperature, production pressure and reaction time;

[0083] Step S120: using infrared spectroscopy to analyze the components of the raw materials of the curing polyester resin, obtaining the type of the catalyst, and obtaining its monomer ratio data, including component concentration and catalyst type;

[0084] Step S130: performing quality inspection on the produced cured polyester resin according to its application conditions to obtain corresponding quality data;

[0085] Step S140: obtaining performance data under different application conditions according to the corresponding performance standard of the cured polyester resin required under each application condition;

[0086] This step provides a data basis for subsequent modeling and optimization by collecting data samples under various application conditions from actual production;

[0087] Step S200: establishing a quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins;

[0088] The mass relationship mapping model learns the mapping relationship between monomer ratio data, curing reaction data and quality data from the existing monomer ratio data, curing reaction data and quality data;

[0089] The specific method of establishing the quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins is:

[0090] Step S210: Standardize the collected monomer ratio data Z, curing reaction data X and quality data Y, perform one-hot encoding on the monomer ratio data, and convert them into vector form, where Z = (z1, z2, ..., z n ), X=(x1,x2,...,x n ), Y={(y 11 ,y 12 ,...,y 1k ),(y 21 ,y 22 ,...,y 2k ),...,(y n1 ,y n2 ,...,y nk )}, wherein there are n groups of monomer ratio data Z, curing reaction data X, and quality data Y, and each group of quality data contains k quality indicators;

[0091] Step S220: constructing a hybrid input deep neural network model as an initial network model of the quality relationship mapping model, and inputting the continuous production data features and the categorized production data features as input data from the input layer;

[0092] Step S230: For continuous production data features, use the fully connected layer as input, and for categorical production data features, use the embedding layer as input;

[0093] Step S240: The hidden layer of the mixed input deep neural network model is connected in series with two convolutional layers to automatically extract the features of the monomer ratio data to obtain the classified production data features, and the long short-term memory network layer is used to capture the features of the curing reaction data to obtain the continuous production data features. The features captured by the convolutional layer and the long short-term memory network layer are fused and input into the fully connected layer for integration, and the predicted quality data is used as the output data, and outputted by the output layer;

[0094] Step S250: using the root mean square error as the loss function and adopting the gradient optimization algorithm to optimize the model parameters, when the error between the predicted output quality data and the actual quality data is the smallest, the model training stops, and a trained quality relationship mapping model is obtained;

[0095] The calculation formula of the loss function is: Among them, n is the number of samples of input data, y ij represents the true value of the jth quality indicator of the quality data in the i-th input data, represents the predicted value of the jth quality indicator of the quality data in the i-th input data;

[0096] This step can predict the quality indicators of products when given production parameters by establishing a quality relationship mapping model, providing a basis for setting quality constraints;

[0097] Step S300: establishing a mathematical model of a performance evaluation function using quality data, performance data, monomer ratio data and curing reaction data of the polyester resin under different application conditions;

[0098] The specific method of establishing a mathematical model of a performance evaluation function using the quality data, performance data, monomer ratio data and curing reaction data of the polyester resin under different application conditions is:

[0099] Step S310: Define performance index calculation function The functional expression of the performance indicator calculation function is: Among them, a ξ represents the ξth application condition, ψ(a ξ ) is the application condition correction function of the ξth application condition, α ξ is the coefficient of the monomer ratio data item under the ξth application condition, γ is the coefficient of the curing reaction data, β j is the coefficient of the jth quality index;

[0100] The coefficients of the monomer ratio data items, the coefficients of the curing reaction data, and the coefficients of the quality index are set according to actual production requirements;

[0101] Step S320: Design application condition correction function ψ(a ξ ), used to adjust the calculation of performance indicators according to different application conditions;

[0102] The application condition correction function ψ(a ξ ) is: Among them, η ξ represents the coefficient of the ξ-th application condition, μ ξ is the phase parameter of the sin function;

[0103] The coefficient of the application condition is set according to the actual production requirements;

[0104] Under different application conditions, the same monomer ratio data and curing reaction data will produce different performance of polyester resin. Therefore, the application condition correction function is designed to take into account the impact of different application conditions on the calculation of performance indicators.

[0105] Step S330: Constructing a performance evaluation function f based on the performance indicator calculation function and the application condition correction functionc (a ξ ,Z,X,Y), the functional expression of the performance evaluation function is: Among them, λ c is the weight coefficient of the cth performance indicator in the performance data, is the cth performance indicator;

[0106] The weight coefficient λ c The setting is performed according to the experience and experiments of those skilled in the art;

[0107] In this step, a performance evaluation function model that comprehensively considers multiple performance indicators is constructed to comprehensively evaluate the performance level of polyester resin products, providing a basis for the subsequent establishment of the objective function;

[0108] Step S400: Obtain the standard performance range under each application condition, calculate the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and design the objective function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data;

[0109] The specific method of obtaining the standard performance range under each application condition, calculating the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and designing the objective function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data is as follows:

[0110] Step S410: According to m application conditions A=(a1, a2, ..., a m ), obtain the standard performance range corresponding to the ξth application condition Integrate the standard performance ranges under all application conditions to obtain the standard performance range set

[0111] The standard performance range refers to the ideal range of performance data that meets production requirements under each application condition;

[0112] Step S420: Design an objective function J(Z,X) using the standard performance range, monomer ratio data, curing reaction data and quality data. The functional expression of the objective function is:

[0113] , where F(a ξ , Z, X, Y) is a comprehensive performance evaluation value, which is the sum of the performance evaluation function values ​​of all performance indicators;

[0114] Step S430: setting constraints for the objective function, wherein the constraints include monomer ratio constraints, curing reaction constraints and quality constraints;

[0115] The constraint expression of the monomer ratio constraint is: g1(Z)≤0, indicating that the monomer ratio data Z is within a feasible range less than or equal to 0;

[0116] The constraint expression of the curing reaction constraint is: L X ≤X≤U X , indicating that the curing reaction data X is within the given constraint range, L X is the lower limit of the curing reaction data, U X It is the upper limit of the curing reaction data;

[0117] The constraint expression of the quality constraint is: Indicates that the mass relationship mapping model h(Z,X,Y) is used to ensure that the mass data Y matches the monomer ratio data Z and the curing reaction data, where δ j represents the standard deviation of the jth quality indicator of the current quality data, y j represents the true value of the jth quality indicator of the current quality data, Represents the predicted value of the jth quality indicator of the current quality data;

[0118] The predicted value of the jth quality indicator of the current quality data It is predicted by the quality relationship mapping model;

[0119] This step integrates multiple conditions such as performance indicators, quality constraints, and production parameters into mathematical objective functions and constraints, providing a formulaic description for the optimization of production parameters;

[0120] Step S500: Calculate the available solution set of monomer ratio data and curing reaction data using the objective function and constraint conditions, use the available solution set as input data of the quality relationship mapping model, predict the quality data of the available solution set and select the optimal quality data from it, and use the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data;

[0121] The available solution set refers to monomer ratio data and curing reaction data that meet the objective function and constraint conditions. Each available solution contains the optimal monomer ratio data and the optimal curing reaction data currently being searched for. The set consisting of all available solutions is the available solution set.

[0122] The specific method of using the objective function and the constraint conditions to calculate the available solution set of monomer ratio data and curing reaction data, using the available solution set as the input data of the quality relationship mapping model, predicting the quality data of the available solution set and screening out the optimal quality data, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data is as follows:

[0123] Step S510: randomly generate N particles, each of which contains curing reaction data, and use the particle swarm optimization algorithm to search for the current optimal curing reaction data X* on the N particles. For the given optimal curing reaction data, use the branch and bound method to obtain the search space of the monomer ratio data [Z min , Z max ], search for the current optimal monomer ratio data Z* in the search space, combine X* and Z*, and obtain an available solution (X*, Z*);

[0124] Step S520: repeatedly searching for the best curing reaction data among the N particles and the best monomer ratio data in the search space until the maximum number of iterations W is reached and the search is stopped to obtain a usable solution set;

[0125] The calculation formula for the maximum number of iterations is: Among them, R x is the dimension of the curing reaction data, R z is the dimension of monomer ratio data, R J is the total number of constraints, s1 is the adjustment coefficient of monomer ratio data and curing reaction data, s2 is the adjustment coefficient of constraint terms, and ε is the expected maximum relative error;

[0126] Step S530: using the available solutions in the available solution set as input data, inputting into the quality relationship mapping model, and predicting and obtaining quality data corresponding to each available solution;

[0127] Step S540: Calculate the comprehensive quality score Q of the quality data corresponding to each available solution;

[0128] The calculation formula of the comprehensive quality score is: Among them, q j is the scoring weight coefficient of the j-th quality indicator, is the jth quality index of the quality data predicted by the available solution;

[0129] Step S550: selecting the quality data with the largest comprehensive quality score as the optimal quality data, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data for producing polyester resin;

[0130] This step combines the particle swarm optimization algorithm and the branch and bound method to search for feasible solutions in parallel and efficiently, thereby improving the search speed. By calculating the comprehensive quality score, the optimal solution that meets the quality and performance constraints is given.

[0131] Example 2

[0132] like Figure 2As shown, an intelligent monitoring outdoor HAA curing polyester resin production quality management system provided by the present application includes:

[0133] Data collection module, used to collect performance data of polyester resin under different application conditions and corresponding monomer ratio data, curing reaction data, and quality data;

[0134] A quality mapping module is used to establish a quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins;

[0135] A performance evaluation module is used to establish a mathematical model of a performance evaluation function using quality data, performance data, monomer ratio data and curing reaction data of polyester resin under different application conditions;

[0136] The target design module is used to obtain the standard performance range under each application condition, calculate the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and design the target function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data;

[0137] The optimal search module is used to calculate the available solution set of monomer ratio data and curing reaction data using the objective function and constraint conditions, use the available solution set as the input data of the quality relationship mapping model, predict the quality data of the available solution set and screen out the optimal quality data, and use the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data.

[0138] Example 3

[0139] Figure 3 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3 As shown, according to another aspect of the present application, an electronic device is also provided. The electronic device may include one or more processors and one or more memories. The memory stores a computer readable code, and when the computer readable code is executed by one or more processors, it can execute the above-mentioned intelligent monitoring outdoor HAA curing polyester resin production quality management method.

[0140] The method or system according to the embodiment of the present application can also be used by Figure 3 The electronic device architecture shown in FIG. Figure 3As shown, the electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, may store an intelligently monitored outdoor HAA-cured polyester resin production quality management method provided in the present application. An intelligently monitored outdoor HAA-cured polyester resin production quality management method may, for example, include: collecting performance data of polyester resins under different application conditions and corresponding monomer ratio data, curing reaction data, and quality data; establishing a quality relationship mapping model using monomer ratio data, curing reaction data, and quality data corresponding to different cured polyester resins; establishing a mathematical model of a performance evaluation function using the quality data, performance data, monomer ratio data, and curing reaction data of polyester resins under different application conditions; obtaining a standard performance range under each application condition, calculating a comprehensive performance evaluation value based on the mathematical model of the performance evaluation function, and using the standard performance range, monomer ratio data, curing reaction data, and quality data to design an objective function and constraints; using the objective function and constraints to calculate an available solution set of monomer ratio data and curing reaction data, using the available solution set as input data for the quality relationship mapping model, predicting the quality data of the available solution set and screening out the optimal quality data, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data. Furthermore, the electronic device may also include a user interface. Of course, Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.

[0141] Example 4

[0142] Figure 4 Schematic diagram of a readable storage medium structure provided by an embodiment of the present application. Figure 4 As shown, it is a readable storage medium according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, an intelligently monitored outdoor HAA-cured polyester resin production quality management method according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0143] In addition, according to an embodiment of the present application, the process described above with reference to the flowchart may be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: collecting performance data and corresponding monomer ratio data, curing reaction data, and quality data of polyester resins under different application conditions; establishing a quality relationship mapping model using monomer ratio data, curing reaction data, and quality data corresponding to different cured polyester resins; establishing a mathematical model of a performance evaluation function using the quality data, performance data, monomer ratio data, and curing reaction data of polyester resins under different application conditions; obtaining a standard performance range under each application condition, calculating a comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and designing an objective function and constraints using the standard performance range, monomer ratio data, curing reaction data, and quality data; calculating an available solution set of monomer ratio data and curing reaction data using the objective function and constraints, using the available solution set as input data of the quality relationship mapping model, predicting the quality data of the available solution set and screening out the optimal quality data therefrom, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are performed.

[0144] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0145] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0146] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent monitoring method for outdoor HAA-cured polyester resin production quality management, characterized in that: include: Collect the performance data of polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data; A quality relationship mapping model is established using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins; A hybrid input deep neural network model is constructed as the initial network model of the quality relationship mapping model, and continuous production data features and categorized production data features are input from the input layer as input data; The hidden layer of the mixed input deep neural network model is connected in series with two convolutional layers to automatically extract the features of the monomer ratio data to obtain the features of the classified production data. The long short-term memory network layer is used to capture the features of the curing reaction data to obtain the features of the continuous production data. The features captured by the convolutional layer and the long short-term memory network layer are fused and input into the fully connected layer for integration. The predicted quality data is used as the output data and output by the output layer. The mathematical model of the performance evaluation function is established using the quality data, performance data, monomer ratio data and curing reaction data of polyester resin under different application conditions; Obtain the standard performance range under each application condition, calculate the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and use the standard performance range, monomer ratio data, curing reaction data and quality data to design the objective function and constraint conditions; The objective function and constraint conditions are used to calculate the available solution set of monomer ratio data and curing reaction data, and the available solution set is used as the input data of the quality relationship mapping model. The quality data of the available solution set is predicted and the optimal quality data is screened out, and the monomer ratio data and curing reaction data corresponding to the optimal quality data are used as the optimal production data. The performance data of the polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data refer to that due to different application conditions, the monomer ratio data and curing reaction data of the polyester resin produced are different, and the corresponding performance data and quality data are different. The performance data include: weather resistance value and mechanical strength value of the polyester resin. The monomer ratio data include component concentration and catalyst type. The curing reaction data include production temperature, production pressure, and reaction time. The monomer ratio data is type-based production data, and the curing reaction data is continuous production data. The specific method of establishing a mathematical model of a performance evaluation function using the quality data, performance data, monomer ratio data and curing reaction data of the polyester resin under different application conditions is: Define performance indicator calculation function The functional expression of the performance indicator calculation function is: Among them, a ξ represents the ξth application condition, ψ(a ξ ) is the application condition correction function of the ξth application condition, α ξ is the coefficient of the monomer ratio data item under the ξth application condition, γ is the coefficient of the curing reaction data, β j is the coefficient of the jth quality index; Design application condition correction function ψ(a ξ ), used to adjust the calculation of performance indicators according to different application conditions; The application condition correction function ψ(a ξ ) is: Among them, η ξ represents the coefficient of the ξ-th application condition, μ ξ is the phase parameter of the sin function; Construct a performance evaluation function f based on the performance indicator calculation function and the application condition correction function c (a ξ ,Z,X,Y), the functional expression of the performance evaluation function is: Among them, λ c is the weight coefficient of the cth performance indicator in the performance data, is the cth performance indicator; The specific method of obtaining the standard performance range under each application condition, calculating the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and designing the objective function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data is as follows: According to m application conditions A=(a1,a2,...,a m ), obtain the standard performance range corresponding to the ξth application condition Integrate the standard performance ranges under all application conditions to obtain the standard performance range set The objective function J(Z,X) is designed using the standard performance range, monomer ratio data, curing reaction data and quality data. The functional expression of the objective function is: Where F(a ξ , Z, X, Y) is a comprehensive performance evaluation value, which is the sum of the performance evaluation function values ​​of all performance indicators; Constraints are set for the objective function, wherein the constraints include monomer ratio constraints, curing reaction constraints, and quality constraints.

2. The intelligent monitoring outdoor HAA-cured polyester resin production quality management method according to claim 1, characterized in that: The performance data of the polyester resin under different application conditions and the corresponding monomer ratio data, curing reaction data, and quality data refer to that due to different application conditions, the monomer ratio data and curing reaction data for producing the polyester resin are different, and the corresponding performance data and quality data are different. The performance data include: weather resistance value and mechanical strength value of the polyester resin, the monomer ratio data include component concentration and catalyst type, the curing reaction data include production temperature, production pressure, and reaction time, the monomer ratio data are classified production data, and the curing reaction data are continuous production data.

3. The intelligent monitoring outdoor HAA-cured polyester resin production quality management method according to claim 2, characterized in that: The specific method of establishing the quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins is: The collected monomer ratio data Z, curing reaction data X and quality data Y are standardized, and the monomer ratio data are one-hot encoded and converted into vector form, where Z = (z1, z2, ..., z n ), X=(x1,x2,...,x n ), Y={(y 11 ,y 12 ,...,y 1k ),(y 21 ,y 22 ,...,y 2k ),...,(y n1 ,y n2 ,...,y nk )}, wherein there are n groups of monomer ratio data Z, curing reaction data X, and quality data Y, and each group of quality data contains k quality indicators; A hybrid input deep neural network model is constructed as the initial network model of the quality relationship mapping model, and continuous production data features and categorized production data features are input from the input layer as input data; For continuous production data features, a fully connected layer is used as input, and for categorical production data features, an embedding layer is used as input; The hidden layer of the mixed input deep neural network model is connected in series with two convolutional layers to automatically extract the features of the monomer ratio data to obtain the features of the classified production data. The long short-term memory network layer is used to capture the features of the curing reaction data to obtain the features of the continuous production data. The features captured by the convolutional layer and the long short-term memory network layer are fused and input into the fully connected layer for integration. The predicted quality data is used as the output data and output by the output layer. The root mean square error is used as the loss function, and the gradient optimization algorithm is used to optimize the model parameters. When the error between the predicted output quality data and the actual quality data is minimized, the model training stops, and the trained quality relationship mapping model is obtained; The calculation formula of the loss function is: Among them, n is the number of samples of input data, y ij represents the true value of the jth quality indicator of the quality data in the i-th input data, Represents the predicted value of the jth quality indicator of the quality data in the i-th input data.

4. The intelligent monitoring outdoor HAA-cured polyester resin production quality management method according to claim 3, characterized in that: The specific method of setting constraints for the objective function is: The constraint expression of the monomer ratio constraint is: g1(Z)≤0, indicating that the monomer ratio data Z is within a feasible range less than or equal to 0; The constraint expression of the curing reaction constraint is: L X ≤X≤U X , indicating that the curing reaction data X is within the given constraint range, L X is the lower limit of the curing reaction data, U X It is the upper limit of the curing reaction data; The constraint expression of the quality constraint is: Indicates that the mass relationship mapping model h(Z,X,Y) is used to ensure that the mass data Y matches the monomer ratio data Z and the curing reaction data, where δ j represents the standard deviation of the jth quality indicator of the current quality data, y j represents the true value of the jth quality indicator of the current quality data, Represents the predicted value of the j-th quality indicator of the current quality data.

5. The intelligent monitoring outdoor HAA-cured polyester resin production quality management method according to claim 4, characterized in that: The specific method of using the objective function and the constraint conditions to calculate the available solution set of monomer ratio data and curing reaction data, using the available solution set as the input data of the quality relationship mapping model, predicting the quality data of the available solution set and screening out the optimal quality data, and using the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data is as follows: N particles are randomly generated, each of which contains curing reaction data. The particle swarm optimization algorithm is used to search for the current optimal curing reaction data X* on the N particles. For the given optimal curing reaction data, the branch and bound method is used to obtain the search space of monomer ratio data [Z min , Z max ], search for the current optimal monomer ratio data Z* in the search space, combine X* and Z*, and obtain an available solution (X*, Z*); Repeatedly search for the optimal curing reaction data among N particles and the optimal monomer ratio data in the search space until the maximum number of iterations W is reached and the search is stopped to obtain an available solution set; The calculation formula for the maximum number of iterations is: Among them, R x is the dimension of the curing reaction data, R z is the dimension of monomer ratio data, R J is the total number of constraints, s1 is the adjustment coefficient of monomer ratio data and curing reaction data, s2 is the adjustment coefficient of constraint terms, and ε is the expected maximum relative error; The available solutions in the available solution set are used as input data and input into the quality relationship mapping model to predict the quality data corresponding to each available solution; Calculate the comprehensive quality score Q of the quality data corresponding to each available solution; The calculation formula of the comprehensive quality score is: Among them, q j is the scoring weight coefficient of the j-th quality indicator, is the jth quality index of the quality data predicted by the available solution; The quality data with the largest comprehensive quality score is selected as the optimal quality data, and the monomer ratio data and curing reaction data corresponding to the optimal quality data are used as the optimal production data for producing polyester resin.

6. An intelligently monitored outdoor HAA-cured polyester resin production quality management system, which is implemented based on an intelligently monitored outdoor HAA-cured polyester resin production quality management method as described in any one of claims 1 to 5, characterized in that: include: Data collection module, used to collect performance data of polyester resin under different application conditions and corresponding monomer ratio data, curing reaction data, and quality data; A quality mapping module is used to establish a quality relationship mapping model using monomer ratio data, curing reaction data and quality data corresponding to different curing polyester resins; A performance evaluation module is used to establish a mathematical model of a performance evaluation function using quality data, performance data, monomer ratio data and curing reaction data of polyester resin under different application conditions; The target design module is used to obtain the standard performance range under each application condition, calculate the comprehensive performance evaluation value according to the mathematical model of the performance evaluation function, and design the target function and constraint conditions using the standard performance range, monomer ratio data, curing reaction data and quality data; The optimal search module is used to calculate the available solution set of monomer ratio data and curing reaction data using the objective function and constraint conditions, use the available solution set as the input data of the quality relationship mapping model, predict the quality data of the available solution set and screen out the optimal quality data, and use the monomer ratio data and curing reaction data corresponding to the optimal quality data as the optimal production data.

7. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the intelligently monitored outdoor HAA-cured polyester resin production quality management method as described in any one of claims 1 to 5 are implemented.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which is suitable for loading by a processor to execute the steps in the intelligently monitored outdoor HAA-cured polyester resin production quality management method as described in any one of claims 1-5.

Images