Control Method for Centralized Feeding System of a PVC Fully Automatic Blending Line

Abstract

This application relates to the technical field of plastic production, and specifically relates to a control method for a centralized feeding system of a PVC full-automatic mixing line. The method includes: obtaining the melt plasticization values of each sub-domain and each sub-domain at each acquisition moment; respectively obtaining the ideal melt plasticization value of each sub-domain and the simulated melt plasticization value of each sub-domain at the next acquisition moment; obtaining the fitness of each parameter vector; in the process of obtaining the optimal parameter vector at the current acquisition moment, the convergence factor at each iteration is updated by the convergence factor at its adjacent iteration, the degree of proximity of the fitness of all parameter vectors at its adjacent iteration, and combining the difference between the parameter vectors between each iteration and its adjacent iteration, and comparing the simulated temperature at the next acquisition moment with the preset ideal temperature; based on the obtained optimal parameter vector, controlling the pressure of the extruder die head. The purpose of this application is to improve the control accuracy of the pressure of the extruder die head.

Description

Technical Field

[0001] This application relates to the technical field of plastic production, and particularly to a control method for a centralized feeding system of a PVC full-automatic mixing line. Background Art

[0002] The centralized feeding system of the PVC full-automatic mixing line consists of an auxiliary material pouring layer, a metering layer, a mixing layer, etc. The particle sizes of the PVC mixed materials in the centralized feeding system of the PVC full-automatic mixing line are different, and the melting states of the PVC mixed materials with different particle sizes in the extruder are different, resulting in fluctuations in the pressure of the extruder die head and affecting the forming of PVC pellets.

[0003] To ensure the production quality of PVC pellets, currently, the PID (Proportional Integral Derivative) algorithm is adopted, and the die head pressure is used as a feedback signal to control the screw rotation speed in the cylinder to ensure a constant die head pressure. However, the centralized feeding system of the PVC full-automatic mixing line needs to adapt to different production conditions and material characteristics, and the fixed PID algorithm parameters are difficult to accurately control the die head pressure, resulting in a decrease in the stability of the extrusion forming quality of PVC pellets. Summary of the Invention

[0004] In view of the above, it is necessary to provide a control method for a centralized feeding system of a PVC full-automatic mixing line, which improves the control accuracy of the pressure of the extruder die head compared with the traditional control method for the centralized feeding system of the PVC full-automatic mixing line.

[0005] A control method for a centralized feeding system of a PVC full-automatic mixing line in this application adopts the following technical solutions:

[0006] An embodiment of this application provides a control method for a centralized feeding system of a PVC full-automatic mixing line, and this method includes the following steps:

[0007] Divide the melting section and the melt conveying section of the extruder barrel into each sub-domain, obtain the actual temperature of each sub-domain at each acquisition moment, and obtain the ideal pressure of the extruder die head;

[0008] Based on the actual temperature, conduct an isothermal flow field modeling for each sub-domain, and by analyzing the pressure distribution and flow velocity distribution in the flow field and combining the actual temperature of each sub-domain, obtain the melt plasticization value of each sub-domain at each acquisition moment;

[0009] Through the transfer function of the frequency converter in the feeding system, simulate the simulated temperature and simulated pressure of each sub-domain and the extrusion die head at the next acquisition moment under the control of any parameter vector including proportional parameter, integral parameter and differential parameter; based on the preset ideal temperature of each sub-domain and the simulated temperature, adopt the same calculation method as the melt plasticization value to respectively obtain the ideal melt plasticization value and simulated melt plasticization value of each sub-domain;

[0010] Based on the difference between the simulated pressure and the ideal pressure, the difference between the simulated melt plasticization value and the ideal melt plasticization value of each sub-domain, and by comparing the distribution of the simulated melt plasticization value and the ideal melt plasticization value of all sub-domains, obtain the fitness of the any parameter vector;

[0011] In the process of obtaining the optimal parameter vector at the current acquisition moment by using the whale optimization algorithm, the convergence factor at each iteration is updated by the convergence factor at its adjacent iteration, the closeness of the fitness of all parameter vectors at its adjacent iteration, and by combining the difference between the parameter vectors between each iteration and its adjacent iteration, and by comparing the simulated temperature at the next acquisition moment with the preset ideal temperature;

[0012] Based on the optimal parameter vector obtained by the whale optimization algorithm, control the pressure of the extrusion die head.

[0013] In one embodiment, the process of obtaining the melt plasticization value is as follows:

[0014] Perform grid meshing on the modeled flow field region to obtain the velocity tensor and pressure tensor of each grid cell at each acquisition moment; through the velocity tensor and the pressure tensor, respectively obtain the flow rate data and pressure data of each grid cell at each acquisition moment;

[0015] Based on the pressure distribution and velocity distribution of all grid cells in each sub-domain at each acquisition moment, and by combining the actual temperature of each sub-domain, obtain the melt plasticization value of each sub-domain at each acquisition moment.

[0016] In one embodiment, the further process of obtaining the melt plasticization value is as follows:

[0017] Calculate the dispersion degree of the velocity data of all grid cells in each sub-domain at each acquisition moment;

[0018] Denote the mean value of the velocity data and the mean value of the pressure data of all grid cells in each sub-domain at each acquisition moment as the velocity mean value and the pressure mean value respectively;

[0019] Analyze the distribution difference of the pressure data of all grid cells in each sub-domain at each acquisition moment;

[0020] The melt plasticization values are positively correlated with the average flow rate, the actual temperature, and the degree of dispersion, and negatively correlated with the average pressure and the distribution difference, respectively.

[0021] In one embodiment, the specific analysis method for the distribution difference is as follows:

[0022] Take the minimum and maximum values of the pressure data of all grid cells in each subdomain at each acquisition moment as the endpoints of the pressure interval of each subdomain at each acquisition moment;

[0023] Divide each pressure interval into multiple quantization intervals, and assign all pressure data within any quantization interval to the minimum value of the pressure data within the any quantization interval;

[0024] Form each sub-interval by combining every preset number of consecutive quantization intervals;

[0025] Take all the assigned pressure data within each sub-interval as the input of the maximum between-class variance algorithm, and output the maximum between-class variance of each sub-interval;

[0026] The distribution difference is reflected by the sum of all the maximum between-class variances corresponding to each subdomain.

[0027] In one embodiment, the process of obtaining the fitness is as follows:

[0028] Form the simulated plasticization state vector and the ideal plasticization state vector of each subdomain by combining the simulated melt plasticization value and the ideal melt plasticization value of each subdomain and its adjacent subdomains, and calculate the similarity between the simulated plasticization state vector and the ideal plasticization state vector of each subdomain;

[0029] Record the difference between the simulated pressure and the ideal pressure as the pressure difference;

[0030] Record the difference between the simulated melt plasticization value and the ideal melt plasticization value of each subdomain as the plasticization difference;

[0031] The fitness is positively correlated with the pressure difference, the plasticization difference, and the serial number of each subdomain, and negatively correlated with the similarity; among them, each subdomain is numbered in order from far to near from the extrusion die head.

[0032] In one embodiment, the further process of obtaining the fitness is: calculate the product of the plasticization difference of each subdomain and its serial number; the fitness is positively correlated with the pressure difference and the product, and negatively correlated with the similarity.

[0033] In one embodiment, the expression of the fitness is:

[0034] ; where F represents the fitness of any one of the parameter vectors; P represents the pressure difference of the extruder die head; L represents the number of sub-domains of the extruder barrel; i represents the serial number of the sub-domain; represents the plasticization difference of the i-th sub-domain; represents the similarity of the i-th sub-domain; exp( ) represents the exponential function with the natural constant as the base.

[0035] In one of the embodiments, the update method of the convergence factor at each iteration is as follows:

[0036] Calculate the distance between the parameter vector with the minimum fitness at each iteration and the parameter vector with the minimum fitness at the adjacent iteration, denoted as the vector distance;

[0037] Calculate the cumulative value of the difference between the simulated temperature of all sub-domains and the preset ideal temperature;

[0038] Statistically calculate the minimum value among the fitnesses of all parameter vectors at each iteration, denoted as the minimum fitness;

[0039] Calculate the difference between the fitness of each parameter vector at the adjacent iteration of each iteration and the minimum fitness;

[0040] The convergence factor at each iteration is positively correlated with the convergence factor at its adjacent iteration, the vector distance, the cumulative value, and the difference.

[0041] In one of the embodiments, the further acquisition process of the convergence factor at each iteration is as follows:

[0042] Calculate the normalized value of the fusion result of the vector distance, the cumulative value, and the difference;

[0043] The convergence factor at each iteration is the product of the convergence factor at its adjacent iteration and the normalized value.

[0044] In one of the embodiments, controlling the pressure of the extruder die head with the optimal parameter vector obtained based on the whale optimization algorithm includes:

[0045] Take the components in the optimal parameter vector as the control parameters of the PID algorithm, calculate the difference between the pressure of the extruder die head at the current moment and the ideal pressure, and combine the PID control algorithm to control the pressure of the extruder die head.

[0046] This application has at least the following beneficial effects:

[0047] In this application, isothermal flow field modeling is carried out for different sub-domains to obtain the flow velocity and pressure distribution of PVC mixture in the cylinder. The beneficial effect is that by analyzing the changes of rheological properties and thermal performance parameters of the mixture with temperature in the sub-domain, the accuracy of establishing the flow field model is improved, and then the accuracy of obtaining the plasticization effect of the mixture melt in the follow-up is enhanced;

[0048] Furthermore, according to the flow velocity and pressure distribution of PVC mixture in the extruder cylinder, the melt plasticization value of the sub-domain is calculated. The beneficial effect is that by analyzing the physical properties of the PVC mixture melt and its extrusion characteristics in the cylinder, and using the degree of difference in the pressure distribution in the cylinder sub-domain, the plasticization state of the melt in each sub-domain can be accurately obtained;

[0049] Furthermore, by simulating the degree to which the plasticization state of the melt deviates from the ideal plasticization state under the control of each parameter vector, the fitness of each parameter vector is obtained, which is used to evaluate the possibility of each parameter vector as the optimal parameter vector, facilitating the subsequent selection of the most suitable parameter vector to precisely control the pressure of the extruder die head;

[0050] Furthermore, a convergence factor is calculated based on the fitness of the parameter vector, which can improve the accuracy of optimizing the parameter vector. The whale optimization algorithm is used to obtain the optimal parameter vector, and each component in the optimal parameter vector is used as the control parameter of the PID algorithm, which can optimize the rotational speed control of the extruder screw, and then achieve more precise control of the pressure of the extruder die head. Description of the Drawings

[0051] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0052] Figure 1 It is a step flow chart of a control method for a centralized feeding system of a PVC full-automatic mixing line provided by the present application;

[0053] Figure 2 It is a schematic diagram of the sensor installation position and sub-domain division;

[0054] Figure 3 It is a schematic diagram of the acquisition process of the melt plasticization value. Detailed Embodiments

[0055] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example", etc. are used to indicate examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary", "or", "for example" is intended to present relevant concepts in a specific manner.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. It should be understood that unless otherwise stated in this application, " / " means "or".

[0057] In addition, it should be noted that the terms "first" and "second" in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0058] The following specifically describes the specific solution of a control method for a centralized feeding system of a PVC full-automatic mixing line provided by this application with reference to the accompanying drawings.

[0059] A control method for a centralized feeding system of a PVC full-automatic mixing line provided by an embodiment of this application. Specifically, the following control method for a centralized feeding system of a PVC full-automatic mixing line is provided. Please refer to Figure 1 , the method includes the following steps:

[0060] Step S1: Divide the melting section and the melt conveying section of the extruder barrel into each sub-domain, and obtain the actual temperature of each sub-domain at each acquisition moment.

[0061] The barrel of the extruder is divided into three functional sections: a solid conveying section, a melting section, and a melt conveying section. Since in the solid conveying section of the barrel, the PVC mixture is in the initial stage of the melting process, the subsequent flow field simulation error of the two-phase flow in this functional section is relatively large, and the state of the PVC mixture in the solid conveying section has a relatively small impact on the quality of the extruded pellets. Therefore, the flow field of the PVC mixture in the solid conveying section is not simulated.

[0062] Further, U temperature sensors are evenly installed on the melting section and the melt conveying section of the barrel. The area between every two adjacent temperature sensors is recorded as a sub-domain of the barrel. A total of L sub-domains are obtained, and the sub-domains are numbered according to the distance from the extruder die. The sub-domain farthest from the extruder die is the first sub-domain, and the sub-domain closest to the extruder die is the Lth sub-domain. A pressure sensor is installed at the die of the extruder.

[0063] In this embodiment, the acquisition frequencies of the temperature sensor and the pressure sensor are both once per second. The values of L and U are 9 and 10 respectively. The acquisition frequency, as well as the values of L and U, are preset manually and can be set by the implementer himself / herself without special restrictions in this application.

[0064] The schematic diagram of the sensor installation position and sub-domain division is as Figure 2 shown, Figure 2 where 1 represents the feed inlet, 2 represents the motor, 3 represents the cylinder, 4 represents the die head, 5 represents the solid conveying section, 6 represents the melting section and the melt conveying section, 7 represents the temperature sensor, and 8 represents the pressure sensor.

[0065] POLYFLOW is a CFD software using the finite element method, which is dedicated to the flow simulation of viscoelastic materials. It integrates a variety of rheological models, thermodynamic models, and chemical reaction models, and has the ability to handle complex boundary conditions, and can simulate the extrusion molding process of PVC blends in an extruder well.

[0066] Since there are temperature differences between different sub-domains, therefore, the simulation model is simplified such that the PVC blends in each sub-domain have the same temperature, and the average value of the temperature data measured at both ends of each sub-domain is used as the actual temperature of each sub-domain, and isothermal flow field modeling is performed on each sub-domain respectively through POLYFLOW software.

[0067] Since the temperatures of different sub-domains are different, there are differences in their flow field modeling. Before establishing the finite element model of the flow field of PVC blends in an extruder using POLYFLOW software and performing finite element mesh division, it is necessary to first obtain the changes of the rheological properties and thermal properties of PVC blends with temperature. In this application, the Cross model and the Arrhenius model are used to represent the changes of the rheological properties and thermal properties of PVC blends in any sub-domain with temperature. Taking the j-th acquisition moment as an example.

[0068] The constitutive equation of the PVC blend flow field is: ; where τ and D represent the pressure tensor and the velocity tensor of the grid element respectively, which are output by the simulation software during the simulation process; η represents the viscosity of the PVC blend.

[0069] The relational expression of the viscosity of the PVC blend changing with the shear rate and temperature is: ; where η represents the viscosity of the PVC blend; and represent the Cross model and the Arrhenius model respectively, and the expressions are as follows:

[0070] ; ; where represents the zero shear rate viscosity; λ represents the relaxation time; γ represents the shear rate of the grid cell, which is output by the simulation software during the simulation process; m represents the Cross model exponent; exp[ ] represents the exponential function with the natural constant as the base; b represents the temperature sensitivity coefficient; represents the reference temperature; T represents the actual temperature of any one of the sub-domains at the j-th acquisition moment.

[0071] In this embodiment, the value of the zero shear rate viscosity is 8000 Pa·s, the value of the relaxation time is 0.12 s; the value of the Cross model exponent is 0.75, and the value of the temperature sensitivity coefficient is 0.0025 K -1 , the value of the reference temperature is 573 K. The implementer can set the values of the zero shear rate viscosity, relaxation time, Cross model exponent, temperature sensitivity coefficient and reference temperature according to the actual situation, and this application does not make special restrictions.

[0072] Based on the structural parameters of the conveying elements such as the screw in the extruder, as well as the above rheological properties and thermal performance parameters of the PVC mixture, this application uses POLYFLOW software to establish a finite element model of the flow field of the PVC mixture in the extruder. After finite element mesh division and boundary condition preprocessing, the flow velocity and pressure distribution of the PVC mixture melt in the extruder barrel are obtained.

[0073] The flow boundary conditions of the PVC mixture flow field are set such that the pressures at the flow inlet and the flow outlet are zero, and there is no slip on the inner surface of the barrel. The finite element mesh uses tetrahedral meshes, and the minimum mesh size is set to 2 mm; the RNG k-epsilon model is used for the PVC mixture flow field simulation model. The establishment process of the PVC mixture flow field simulation model is well known to those skilled in the art, and this application will not elaborate.

[0074] Take the 2-norm of the velocity tensor of each grid cell at the j-th acquisition moment as the velocity data of each grid cell at the j-th acquisition moment, and take the 2-norm of the pressure tensor of each grid cell at the j-th acquisition moment as the pressure data of each grid cell at the j-th acquisition moment.

[0075] Step S2, perform isothermal flow field modeling on each sub-domain based on the actual temperature. By analyzing the pressure distribution and velocity distribution in the flow field, and combining with the actual temperature of each sub-domain, obtain the melt plasticization value of each sub-domain at each acquisition moment.

[0076] Affected by the differences in the particle sizes of the PVC mixed materials in the centralized feeding system of the PVC full-automatic mixing line, there are differences in the melting process of the PVC mixed materials in the barrel of the extruder. The smaller the mixed material particles, the more heat generated by heat conduction and the interaction between the mixed materials, the higher the temperature in the barrel, the higher the melting state of the PVC mixed materials, and the better the plasticizing effect of the melt; on the contrary, the larger the mixed material particles, the lower the temperature in the barrel, the worse the plasticizing effect of the PVC mixed material melt, and at the same time, it causes large fluctuations in the pressure of the extruder die head, affecting the extrusion molding quality of the PVC pellets.

[0077] Considering the physical properties of the PVC mixed material melt, the higher its temperature, the better its fluidity. Although the pressure distribution at different positions in the barrel fluctuates, the fluctuation range is small; while the lower the temperature, the worse the fluidity, the high pressure is generated on the advancing surface of the screw thread of the PVC mixed material melt, and the pressure on the back of the thread is lower, resulting in obvious differences in the pressure distribution at different positions in the barrel. Still taking the jth acquisition moment as an example.

[0078] To obtain the degree of difference in the pressure distribution in each sub-domain of the barrel, the minimum and maximum values of the pressure data of all grid cells in each sub-domain at the jth acquisition moment are statistically calculated, and the minimum value and the maximum value are used as the interval endpoints to form the pressure interval of each sub-domain. Each pressure interval is equally divided into a preset number of quantization intervals, and all pressure data within any quantization interval are assigned the minimum value of the pressure data within the any quantization interval.

[0079] Furthermore, a plurality of consecutive quantization intervals in each pressure interval are combined into a sub-interval, and all the pressure data after assignment within each sub-interval are used as the input of the maximum between-class variance algorithm, and the maximum between-class variance of each sub-interval is output. The larger the maximum between-class variance, the more obvious the distribution difference of the pressure data within each sub-interval, the greater the pressure difference at different extrusion positions in the barrel, the worse the fluidity of the PVC mixed material, and the worse the plasticizing effect of the mixed material melt. Among them, the maximum between-class variance algorithm is a well-known technology and will not be elaborated in this application.

[0080] In this embodiment, the value of the preset number is 256, and the value of the preset number is preset manually, and the implementer can set it by himself / herself, and this application does not make special restrictions.

[0081] In this embodiment, 32 consecutive quantization intervals are combined into a sub-interval, and there are 8 sub-intervals in each pressure interval. The number of consecutive quantization intervals that make up the sub-interval can be set by the implementer himself / herself, and this application does not make special restrictions.

[0082] Furthermore, analyze the pressure distribution and flow velocity distribution of all grid cells within each sub-domain, and combine the actual temperature of each sub-domain to obtain the melt plasticization value of each sub-domain, which is used to reflect the plasticization effect when the PVC mixture forms a melt within each sub-domain. The larger the melt plasticization value, the better the plasticization effect. However, when it is too large, the risk of PVC mixture decomposition is greater, which is not conducive to subsequent extrusion molding. The expression is:

[0083] ; In the formula, represents the melt plasticization value of the i-th sub-domain; and respectively represent the average value of the flow velocity data and the average value of the pressure data of all grid cells within the i-th sub-domain; represents the actual temperature of the i-th sub-domain; represents the degree of dispersion of the flow velocity data of all grid cells within the i-th sub-domain; represents the sum of the maximum between-class variances of all sub-intervals within the pressure interval of the i-th sub-domain.

[0084] In this embodiment, the degree of dispersion is the standard deviation. As other implementation manners, on the basis of being able to measure the uneven degree of the distribution of the flow velocity data, the implementer can use other existing technologies for measurement, such as variance, coefficient of variation, etc. This application does not make special restrictions.

[0085] It should be noted that temperature, flow velocity, and pressure are important indicators reflecting the melt plasticization state of the PVC mixture. The higher the actual temperature and the average flow velocity of the melt, the better the melt plasticization effect. At the same time, the more uniform the flow velocity of the melt, the smaller the calculated degree of dispersion, indicating that the change in the flow velocity of the melt in different pressure regions during screw transmission is smaller, and the flow performance of the melt is worse, and the melt plasticization value is smaller. In addition, the smaller the pressure of the melt, it means that the stress of the mixture during screw transmission is smaller, the flow performance of the melt is better, and the calculated melt plasticization value is larger; reflects the degree of difference in the pressure distribution at different positions in the barrel. The greater the degree of difference in the pressure distribution, the greater the sum of the maximum between-class variances, the worse the flow performance of the PVC mixture in the barrel, the worse the melt plasticization effect, and the smaller the calculated melt plasticization value. The schematic flow chart of obtaining the melt plasticization value is as Figure 3 shown.

[0086] Adopt the same acquisition method as the melt plasticization value of the i-th sub-domain at the j-th acquisition moment to obtain the melt plasticization values of each sub-domain at each acquisition moment.

[0087] Step S3: Obtain the ideal pressure of the extruder die head. Through the transfer function of the frequency converter in the feeding system, simulate the simulated temperature and simulated pressure of each sub-domain and the extruder die head at the next acquisition moment at the current acquisition moment under the control of any parameter vector including proportional parameter, integral parameter, and derivative parameter. Based on the preset ideal temperature of each sub-domain and the simulated temperature, respectively obtain the ideal melt plasticization value and simulated melt plasticization value of each sub-domain by using the same calculation method as the melt plasticization value.

[0088] To ensure that the pressure of the extruder die head remains constant and avoid defects such as depressions and drags in the extruded PVC pellets, it is necessary to analyze whether there are deviations in the melting state of the PVC mixture in each sub-domain according to the melt plasticization value of each sub-domain of the extruder barrel, and adjust the screw speed in a timely manner when deviations occur to ensure that the die head pressure remains stable during melt transfer.

[0089] Since the heating power of each sub-domain is constant, each sub-domain has a preset ideal temperature. However, due to differences in the particle size of the pellets and the complexity of the mixing and stirring process, there are differences between the actual temperature of the melt in each sub-domain and the preset ideal temperature. Therefore, taking the parameters of the pellets in the extruder at the preset ideal temperature as the evaluation standard, adjust the screw speed to improve the control accuracy of the die head pressure.

[0090] In this embodiment, the preset ideal temperatures of the first sub-domain to the L-th sub-domain are 126°C, 130°C, 132°C, 134°C, 136°C, 138°C, 140°C, 142°C, 142°C respectively, and the preset ideal temperature of each sub-domain can be set by the implementer according to the actual situation.

[0091] Take the pressure data collected by the pressure sensor when the actual temperature of each sub-domain is the preset ideal temperature as the ideal pressure of the die head.

[0092] For the frequency converter used to control the screw speed in the centralized feeding system of the PVC full-automatic compounding line, use the PID algorithm to control the frequency converter, so as to achieve the purpose of keeping the die head pressure stable. Select three random numbers in the open interval (0, 100) as the proportional parameter, integral parameter, and derivative parameter of the PID algorithm respectively, and form a parameter vector. Take the difference between the die head pressure and the ideal pressure as the input of the PID algorithm, output a control instruction, and use the frequency converter to control the screw speed of the extruder according to the control instruction to keep the pressure of the extruder die head stable.

[0093] Through the transfer function of the frequency converter used to control the screw speed, simulate the simulated temperature and simulated pressure of each sub-domain and the extruder die head at the next acquisition moment under the control of any parameter vector. Among them, the calculation of the transfer function and the calculation of the simulated temperature and simulated pressure through the transfer function are all well-known technologies, which will not be elaborated in this application.

[0094] Based on the preset ideal temperature of each sub-domain, using the same acquisition method as the melt plasticization value of the i-th sub-domain at the j-th acquisition moment, replace the actual temperature with the preset ideal temperature to obtain the melt plasticization value of each sub-domain, denoted as the ideal melt plasticization value; based on the simulated temperature of each sub-domain at the next acquisition moment of the current acquisition moment, using the same acquisition method as the melt plasticization value of the i-th sub-domain at the j-th acquisition moment, replace the actual temperature with the simulated temperature to obtain the melt plasticization value of each sub-domain at the next acquisition moment, denoted as the simulated melt plasticization value.

[0095] When the melt plasticization value is less than the ideal melt plasticization value, the plasticization effect of the melt in the sub-domain is poor; when the melt plasticization value exceeds the ideal melt plasticization value, the risk of decomposition of the PVC mixture in the sub-domain is relatively large. Both situations will affect the quality of the subsequent extrusion molding of PVC pellets.

[0096] Step S4, based on the difference between the simulated pressure and the ideal pressure, the difference between the simulated melt plasticization value and the ideal melt plasticization value of each sub-domain, and comparing the distribution of the simulated melt plasticization value and the ideal melt plasticization value of all sub-domains, obtain the fitness of the any parameter vector.

[0097] To select a suitable parameter vector, obtain W parameter vectors as the initial population of the whale optimization algorithm, and use the whale optimization algorithm to perform iterative processing on multiple parameter vectors. Finally, output the optimal parameter vector, and use each component in the optimal parameter vector as the control parameter of the PID algorithm. During the iteration process, it is necessary to measure the quality of each parameter vector. The specific process is as follows:

[0098] The simulated melt plasticization values and the ideal melt plasticization values of each sub-domain and its adjacent sub-domains are respectively composed of the simulated plasticization state vector and the ideal plasticization state vector of each sub-domain.

[0099] In this embodiment, the value of W is 50, and the value of W is preset manually. The implementer can set it by himself, and this application does not make special restrictions.

[0100] In this embodiment, the simulated melt plasticization value and the ideal melt plasticization value of each sub-domain and its adjacent subsequent sub-domain are respectively used to form the simulated plasticization state vector and the ideal plasticization state vector of each sub-domain. For the L-th sub-domain, both components of the simulated plasticization state vector of the L-th sub-domain are the simulated melt plasticization value of the L-th sub-domain, and both components of the ideal plasticization state vector of the L-th sub-domain are the ideal melt plasticization value of the L-th sub-domain. As another implementation, the simulated melt plasticization value and the ideal melt plasticization value of each sub-domain and its adjacent previous sub-domain can also be respectively used to form the simulated plasticization state vector and the ideal plasticization state vector of each sub-domain. For the first sub-domain, both components of the simulated plasticization state vector of the first sub-domain are the simulated melt plasticization value of the first sub-domain, and both components of the ideal plasticization state vector of the first sub-domain are the ideal melt plasticization value of the first sub-domain.

[0101] Analyze the difference between the simulated pressure and the ideal pressure of the extruder die head, and the difference between the simulated melt plasticization value and the ideal melt plasticization value of each sub-domain. Combine the similarity between the simulated plasticization state vector and the ideal plasticization state vector of each sub-domain to obtain the fitness of any one of the parameter vectors, which is used to reflect the extrusion molding quality of PVC pellets when controlling the rotation speed of the screw in the cylinder by the parameter vector. The smaller the fitness, the better the extrusion molding quality of the pellets. The specific process is as follows:

[0102] Calculate the similarity between the simulated plasticization state vector and the ideal plasticization state vector of each sub-domain. Denote the difference between the simulated pressure and the ideal pressure as the pressure difference; denote the difference between the simulated melt plasticization value and the ideal melt plasticization value of each sub-domain as the plasticization difference. The fitness of any one of the parameter vectors is positively correlated with the pressure difference, the plasticization difference, and the serial number of each sub-domain, and negatively correlated with the similarity.

[0103] The expression for the fitness of any one of the parameter vectors is:

[0104] ; where F represents the fitness of any one of the parameter vectors; P represents the pressure difference of the extruder die head; L represents the number of sub-domains of the extruder barrel; i represents the serial number of the sub-domain; represents the plasticization difference of the i-th sub-domain; represents the similarity of the i-th sub-domain; exp( ) represents the exponential function with the natural constant as the base, and the purpose is to map to a positive value.

[0105] In this embodiment, during the process of calculating the fitness, all the involved differences are the squares of the differences. As other implementation manners, on the basis of being able to measure the difference between the simulated pressure and the ideal pressure, and the difference between the simulated melt plasticization value and the ideal melt plasticization value, implementers can adopt other calculation methods, such as the absolute value of the difference, the ratio, etc. This application does not make special restrictions.

[0106] In this embodiment, the similarity between the simulated plasticization state vector and the ideal plasticization state vector is the cosine similarity. As other implementation manners, on the basis of being able to measure the similarity between the simulated plasticization state vector and the ideal plasticization state vector, implementers can adopt other existing technologies for measurement, such as the reciprocal of the Euclidean distance, the reciprocal of the DTW distance between the sequences formed by arranging vectors in time series, etc. This application does not make special restrictions.

[0107] It should be noted that: on the one hand, the die head pressure directly affects the extrusion molding quality of PVC pellets. The greater the difference between the simulated pressure and the ideal pressure, the greater the calculated fitness, indicating that the extrusion molding quality of the pellets is worse; on the other hand, the ideal plasticization states of the mixed materials in different sub-domains are different, and their ideal melt plasticization values change according to a certain gradient; the greater the similarity between the simulated plasticization state vector of the sub-domain and the ideal plasticization state vector, the closer the change trends of their plasticization states, and the smaller the calculated fitness; at the same time, the greater the difference between the simulated melt plasticization value and the ideal melt plasticization value of the sub-domain, the greater the fitness, and the larger the serial number of the sub-domain, the closer its distance to the die head, and the greater the impact on the extrusion molding of PVC pellets. Therefore, a larger calculation weight is set; the parameter vector with a smaller fitness is more suitable as the optimal parameter vector.

[0108] Step S5, during the process of obtaining the optimal parameter vector at the current acquisition moment by using the whale optimization algorithm, the convergence factor at each iteration is updated by the convergence factor at its adjacent iteration, the proximity of the fitness of all parameter vectors at its adjacent iteration, and in combination with the difference between the parameter vectors between each iteration and its adjacent iteration, and by comparing the simulated temperature at the next acquisition moment with the preset ideal temperature; based on the optimal parameter vector obtained by the whale optimization algorithm, control the pressure of the extrusion die head.

[0109] Considering the complexity of the PVC mixed material flow field in the barrel, the fitness distribution of different parameter vectors has great randomness. Therefore, the update method of the convergence factor at each iteration is as follows:

[0110] Calculate the distance between the parameter vector with the minimum fitness at each iteration and the parameter vector with the minimum fitness at the adjacent iteration, denoted as the vector distance; calculate the accumulated value of the difference between the simulated temperature of all sub-domains and the preset ideal temperature; count the minimum value among the fitness values of all parameter vectors at each iteration, denoted as the adaptation minimum; calculate the difference between the fitness of each parameter vector at the adjacent iteration of each iteration and the adaptation minimum, denoted as the adaptation difference; calculate the normalized value of the fusion result of the vector distance, the accumulated value and the difference; the convergence factor at each iteration is the product of the convergence factor at its adjacent iteration and the normalized value.

[0111] In this embodiment, the value of the convergence factor at the first iteration is 2. The value of the convergence factor at the first iteration can be set by the implementer according to the actual situation, and this application does not make special restrictions.

[0112] In this embodiment, the distance between parameter vectors is the Euclidean distance. As other implementation manners, on the basis of being able to measure the distance between parameter vectors, the implementer can use other existing technologies for measurement, such as Manhattan distance, cosine distance, etc., and this application does not make special restrictions.

[0113] In this embodiment, the Sigmoid function is used to normalize the fusion result. As other implementation manners, on the basis of being able to normalize the fusion result, the implementer can use other existing technologies, such as decimal scaling normalization method, hyperbolic tangent function, etc., and this application does not make special restrictions.

[0114] It should be noted that: Fusion refers to combining multiple independent variables in a way that enhances the overall effect, such as addition relationship, multiplication relationship, etc. The implementer can make limitations according to the actual situation, and this application does not make special restrictions.

[0115] In this embodiment, the expression of the fusion result is: ; where B represents the fusion result; d represents the distance between the parameter vector with the minimum fitness at the t-th iteration and the parameter vector with the minimum fitness at the (t - 1)-th iteration; represents the accumulated value; W represents the number of parameter vectors in the population of the whale optimization algorithm; represents the adaptation difference corresponding to the w-th parameter vector at the (t - 1)-th iteration.

[0116] In another embodiment, the expression of the fusion result is: ; where B represents the fusion result; d represents the distance between the parameter vector with the minimum fitness at the t-th iteration and the parameter vector with the minimum fitness at the (t - 1)-th iteration; represents the accumulated value; W represents the number of parameter vectors in the whale optimization algorithm population; represents the fitness difference corresponding to the w-th parameter vector at the (t - 1)-th iteration.

[0117] It should be noted that: on the one hand, if the distance d between the parameter vectors with the minimum fitness in two adjacent iterations is smaller, and the difference between the fitness values of each parameter vector in the population and the minimum fitness value is smaller, it indicates that the parameter vector with the current minimum fitness is closer to the optimum, the convergence factor decreases faster, and the accuracy of optimization is improved; on the other hand, the greater the difference between the simulated temperature of the subdomain and the preset ideal temperature, the greater the range of adjustment required for the screw speed, and thus the optimization range of the parameter vector needs to be expanded, and the convergence factor decreases more slowly to avoid falling into local optimum.

[0118] The process of obtaining the optimal parameter vector at the current moment using the whale optimization algorithm is as follows: taking the initial population as the input of the whale optimization algorithm, and based on the calculated fitness and convergence factor, outputting the optimal parameter vector, where the maximum number of iterations is M. Taking each component in the optimal parameter vector as the control parameter of the PID algorithm, taking the difference between the pressure of the extruder die head at the current moment and the ideal pressure as the input of the PID algorithm, outputting a control instruction, and using the frequency converter to control the screw speed of the extruder according to the control instruction to keep the pressure of the extruder die head stable and improve the quality of PVC pellet extrusion molding.

[0119] In this embodiment, the value of M is 50, and the value of M is preset manually. The implementer can set it by himself, and this application does not make special restrictions.

[0120] To sum up, this application separately conducts isothermal flow field modeling for different subdomains to obtain the flow velocity and pressure distribution of the PVC mixture in the cylinder. Its beneficial effect lies in improving the accuracy of the establishment of the flow field model by analyzing the changes of the rheological and thermal performance parameters of the mixture in the subdomain with temperature, and further improving the accuracy of obtaining the plasticization effect of the mixture melt in the subsequent process;

[0121] Furthermore, according to the flow velocity and pressure distribution of the PVC mixture in the extruder cylinder, calculate the melt plasticization value of the subdomain. Its beneficial effect lies in accurately obtaining the plasticization state of the melt in each subdomain by analyzing the physical properties of the PVC mixture melt and its extrusion characteristics in the cylinder, and using the difference degree of the pressure distribution in the cylinder subdomain.

[0122] Furthermore, by simulating the degree to which the plasticization state of the melt deviates from the ideal plasticization state in the future under the control of each parameter vector, obtain the fitness of each parameter vector, which is used to evaluate the possibility of each parameter vector as the optimal parameter vector, facilitating the subsequent selection of the most suitable parameter vector to precisely control the pressure of the extruder die head.

[0123] Furthermore, calculating the convergence factor based on the parameter vector can improve the accuracy of optimizing the parameter vector. The whale optimization algorithm is adopted to obtain the optimal parameter vector, and each component in the optimal parameter vector is used as the control parameter of the PID algorithm, which can optimize the rotational speed control of the extruder screw and further achieve more precise control of the pressure of the extruder die head.

[0124] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions noted in the blocks may occur in an order different from that noted in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in an order different from that disclosed in the description. Sometimes, there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. Each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0125] For those skilled in the art, it is obvious that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the basic features of the present application. Therefore, from any point of view, the above embodiments of the present application should be regarded as exemplary and non-limiting.

Claims

1. A control method for a centralized feeding system of a PVC fully automatic compounding line, characterized in that: The method comprises the following steps: The melting section and melt conveying section of the extruder barrel are divided into sub-areas, and the actual temperature of each sub-area at each sampling time is obtained; Based on the actual temperature, isothermal flow field modeling is performed on each subdomain, and the melt plasticization value of each subdomain at each sampling time is obtained by analyzing the pressure distribution and flow velocity distribution in the flow field and combining the actual temperature of each subdomain; The ideal temperature of each subdomain is preset, and when the actual temperature of each subdomain is the corresponding ideal temperature, the pressure data collected by the pressure sensor is used as the ideal pressure of the die head; through the transfer function of the frequency converter in the feeding system, the simulated temperature and simulated pressure of each subdomain and the extruder die head at the next acquisition time of the current acquisition time are simulated under the control of any parameter vector including proportional parameters, integral parameters and differential parameters; based on the simulated temperature of each subdomain, the same calculation method as the melt plasticization value is adopted to obtain the simulated melt plasticization value of each subdomain; based on the preset ideal temperature of each subdomain, the same acquisition method as the melt plasticization value is adopted, the actual temperature is replaced by the preset ideal temperature, and the melt plasticization value of each subdomain is obtained, which is recorded as the ideal melt plasticization value; Based on the difference between the simulated pressure and the ideal pressure, the difference between the simulated melt plasticization value and the ideal melt plasticization value of each subdomain, and comparing the distribution of the simulated melt plasticization value and the ideal melt plasticization value of all subdomains, the fitness of any parameter vector is obtained; In the process of using the whale optimization algorithm to obtain the optimal parameter vector at the current acquisition moment, the convergence factor at each iteration is updated by comparing the simulated temperature at the next acquisition moment with the preset ideal temperature through the convergence factor at the adjacent iteration and the fitness of all parameter vectors at the adjacent iterations, combined with the difference between the parameter vectors of each iteration and its adjacent iterations; The optimal parameter vector obtained based on the whale optimization algorithm is used to control the pressure of the extruder die head.

2. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 1, characterized in that: The process of obtaining the melt plasticization value is as follows: The flow field area obtained by modeling is meshed to obtain the velocity tensor and pressure tensor of each grid unit at each acquisition time; the flow data and pressure data of each grid unit at each acquisition time are respectively obtained through the velocity tensor and the pressure tensor; Based on the pressure distribution and velocity distribution of all grid cells in each subdomain at each acquisition time, and combined with the actual temperature of each subdomain, the melt plasticization value of each subdomain at each acquisition time is obtained.

3. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 2, characterized in that: The further acquisition process of the melt plasticization value is as follows: Calculate the discrete degree of velocity data of all grid cells in each subdomain at each acquisition time; The mean value of the velocity data and the mean value of the pressure data of all grid cells in each subdomain at each acquisition time are recorded as the velocity mean and the pressure mean, respectively; Analyze the distribution differences of pressure data of all grid cells in each subdomain at each acquisition time; The melt plasticization value is positively correlated with the mean flow rate, the actual temperature, and the discrete degree, and is negatively correlated with the mean pressure and the distribution difference.

4. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 3, characterized in that: The specific analysis method of the distribution difference is: The minimum and maximum values ​​of the pressure data of all grid cells in each subdomain at each acquisition time are taken as the endpoints of the pressure interval of each subdomain at each acquisition time; Each pressure interval is equally divided into a plurality of quantization intervals, and all pressure data in any quantization interval are assigned the minimum value of the pressure data in any quantization interval; Combining a preset number of continuous quantization intervals into each sub-interval; All the assigned pressure data in each sub-interval are used as the input of the maximum inter-class variance algorithm, and the maximum inter-class variance of each sub-interval is output; The distribution difference is reflected by the sum of all the maximum inter-class variances corresponding to each subdomain.

5. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 1, characterized in that: The process of obtaining the fitness is as follows: The simulated melt plasticization value and the ideal melt plasticization value of each subdomain and its adjacent subdomain are used to form a simulated plasticization state vector and an ideal plasticization state vector of each subdomain, respectively, and the similarity between the simulated plasticization state vector and the ideal plasticization state vector of each subdomain is calculated; The difference between the simulated pressure and the ideal pressure is recorded as a pressure difference; The difference between the simulated melt plasticization value and the ideal melt plasticization value of each subdomain is recorded as the plasticization difference; The fitness is positively correlated with the pressure difference, the plasticization difference, and the serial number of each subdomain, and negatively correlated with the similarity; wherein each subdomain is numbered in sequence from far to near according to the distance between the subdomain and the extruder die head.

6. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 5, characterized in that: The further acquisition process of the fitness is: calculating the product of the plasticization difference and the serial number of each subdomain; the fitness is positively correlated with the pressure difference and the product, and negatively correlated with the similarity.

7. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 5, characterized in that: The fitness expression is: ; Wherein, F represents the fitness of any parameter vector; P represents the pressure difference of the extruder die head; L represents the number of subdomains of the extruder barrel; i represents the sequence number of the subdomain; represents the plasticization difference of the i-th subdomain; represents the similarity of the ith subdomain; exp() represents an exponential function with a natural constant as the base.

8. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 1, characterized in that: The update method of the convergence factor at each iteration is: Calculate the distance between the parameter vector with the minimum fitness in each iteration and the parameter vector with the minimum fitness in the adjacent iteration, recorded as vector distance; Calculating the accumulated value of the difference between the simulated temperature and the preset ideal temperature of all sub-domains; The minimum value of the fitness of all parameter vectors at each iteration is counted and recorded as the minimum fitness value; Calculating the difference between the fitness of each parameter vector and the minimum fitness value in adjacent iterations of each iteration; The convergence factor in each iteration is positively correlated with the convergence factor in the adjacent iteration, the vector distance, the accumulated value, and the difference value.

9. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 8, characterized in that: The further acquisition process of the convergence factor at each iteration is as follows: Calculating a normalized value of a fusion result of the vector distance, the accumulated value, and the difference; The convergence factor at each iteration is the product of the convergence factor at the adjacent iteration and the normalized value.

10. The control method of a centralized feeding system for a PVC fully automatic compounding line according to claim 1, characterized in that: The optimal parameter vector obtained based on the whale optimization algorithm controls the pressure of the extruder die head, including: The components in the optimal parameter vector are used as control parameters of the PID algorithm, the difference between the pressure of the extruder die head at the current moment and the ideal pressure is calculated, and the pressure of the extruder die head is controlled in combination with the PID control algorithm.

Images