Temperature distribution control method and system for edge strengthening blow molding panel

Abstract

The invention discloses an edge strengthening blow molding panel temperature distribution control method and system, and relates to the technical field of plastic production, and the method comprises the steps: predicting a real mold temperature field presented by a mold in the future after blow molding of a panel blank according to a pre-trained first machine learning model; a pre-configured genetic algorithm is called, and an ideal temperature field which should be presented by the mold after blow molding is conducted on the panel blank is obtained; according to the real mold uniform temperature and the target mold uniform temperature, whether the panel blank meets the conversion condition for entering the next cooling link or not is judged; when the conversion condition is not met, obtaining at least one target temperature adjusting zone and the temperature adjusting quantity corresponding to the target temperature adjusting zone from all zones; after heating or cooling operation is carried out on the target temperature adjusting zone according to the temperature adjusting quantity, blow molding is carried out on the panel blank; according to the invention, the die can be accurately adjusted, so that the production quality and the yield of finished panels can be ensured.

Description

Technical Field

[0001] The invention relates to the technical field of plastic production, and in particular to a temperature distribution control method and system for an edge-reinforced blow-molded panel. Background Art

[0002] During the blow molding process of hollow plastic panels, the temperature distribution of the mold has a significant impact on the final shape, thickness distribution and structural strength of the product; in order to achieve local differentiated temperature control, the existing technology usually divides the mold into multiple heating / cooling zones, and configures temperature control units to adjust each zone.

[0003] However, existing mold temperature control methods mainly rely on uniform heating or regular temperature control strategies, and lack the ability to model and predict the actual evolution trend of the temperature field. In actual production, the edge area of ​​the mold is far away from the main heat source or the cooling path is complex, and often has problems such as low temperature and control lag, which can easily lead to defects such as insufficient thickness, warping, and stripes in the edge area of ​​the panel, seriously affecting the consistency and mechanical properties of the finished product.

[0004] In addition, there is an obvious thermal coupling effect between the temperature control zones of the mold, that is, the adjustment of the temperature of one zone often affects the temperature distribution of other zones through heat conduction or structural coupling; most of the existing temperature control strategies do not fully consider this factor, resulting in inaccurate temperature control effects, and even causing temperature deviations in other areas, affecting the overall temperature control stability.

[0005] At the same time, traditional temperature control solutions lack a systematic optimization mechanism. They are unable to select the most appropriate temperature control zone based on the actual temperature difference and thermal coupling relationship, and are unable to quantify the temperature adjustment amount. They often rely on manual experience settings, and the adjustment efficiency and accuracy are greatly limited. Summary of the invention

[0006] The object of the present invention is to provide a method and system for controlling the temperature distribution of an edge-reinforced blow-molded panel to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for controlling temperature distribution of an edge-reinforced blow-molded panel, comprising: In the blow molding process, the real mold temperature field of the mold after blow molding of the panel blank is predicted based on the pre-trained first machine learning model. The real mold temperature field includes the real mold average temperature of R partitions on the mold after blow molding. , R is a positive integer; Call the preconfigured genetic algorithm to obtain the ideal temperature field that the mold should present after blow molding the panel blank, the ideal temperature field includes the target mold average temperature of R partitions on the mold after blow molding. ; According to the actual mold temperature and target mold temperature , determine whether the panel blank meets the conversion conditions for entering the next cooling stage; the conversion conditions are the preset temperature difference tolerance interval ΔT and the area number interval Δ S ; When the conversion condition is not met, obtaining at least one target temperature adjustment zone and the temperature adjustment amount corresponding to the target temperature adjustment zone from all zones; After heating or cooling the target temperature adjustment zone according to the temperature adjustment amount, the panel blank is blow-molded to complete the blow-molding process.

[0008] Furthermore, the real mold temperature field is obtained as follows: Obtaining the initial average mold temperature of the R partitions, and obtaining second characteristic data, wherein the second characteristic data includes attribute data of the panel blank, basic data of the mold, and parameter data of the blow molding process; Among them, the attribute data includes the initial blank temperature, material type, blank viscosity, specific heat capacity and blank volume / mass of the panel blank; the basic data includes the thermal conductivity coefficient of the mold, the mold material, the number of partitions R and the partition distance of each partition; the parameter data includes the blowing pressure, blowing time and blowing flow rate; The second feature data and the initial mold average temperatures of the R partitions are input into the first machine learning model to obtain the real mold temperature field that the mold will present in the future after the panel blank is blow-molded; The training method of the first machine learning model is as follows: Acquire historical temperature detection training data, and divide the historical temperature detection training data into a temperature detection training set and a temperature detection test set, wherein the historical temperature detection training data includes the first feature data and its corresponding real mold temperature field, that is, the real mold average temperature of R partitions on the mold; Wherein, the first characteristic data includes the second characteristic data and the initial mold average temperature of R partitions; Constructing a recurrent neural network, taking the first feature data in the temperature detection training set as the input of the recurrent neural network, and taking the real mold temperature field as the output, training the recurrent neural network to obtain an initial recurrent neural network; The temperature detection test set is used to verify the model of the initial recurrent neural network, and the output of the initial recurrent neural network that is less than or equal to the preset test error threshold is used as the trained first machine learning model.

[0009] Furthermore, the method for obtaining the ideal temperature field is as follows: a1: Initialize the population: randomly generate an original population, which contains x individuals, each of which represents a test mold temperature field, including the average temperature of the test mold in R partitions, and x is an integer greater than zero; a2: Fitness evaluation: Under each individual, obtain the thickness distribution data of the panel blank in the mold; input the thickness distribution data into the pre-built fitness function to calculate the fitness of each individual, wherein the thickness distribution data is the thickness of the panel blank in each partition of the mold; a3: Selection: Use the roulette method to select two individuals with high fitness in the original population as the father and mother; a4: Crossover: Perform crossover operation on the father and mother to produce new individuals; a5: Mutation: Perform mutation operation on the new individuals to obtain Y new individuals, combine the Y new individuals into a new population, replace the original population with the new population, and return to step a2; a6: Repeat steps a2 to a5 above until the fitness of individuals in the original population or the new population is greater than or equal to the preset fitness threshold, or the number of iterations is greater than or equal to the preset maximum number of iterations threshold, and output the test mold temperature field represented by the corresponding individual as the ideal temperature field, that is, the target mold average temperature of the R partitions on the mold after blow molding is obtained.

[0010] Furthermore, the thickness distribution data of the panel blank in the mold includes: Obtaining the average temperature of the test mold of the R partitions, and obtaining the second characteristic data under the average temperature of the test mold; Input the average temperature of the test mold of R partitions and the second characteristic data into the pre-trained second machine learning model to obtain the thickness distribution data of the panel blank; The training method of the second machine learning model is as follows: Acquire historical thickness detection training data, and divide the historical thickness detection training data into a thickness detection training set and a thickness detection test set, wherein the historical thickness detection training data includes the third feature data and its corresponding thickness distribution data; Wherein, the third characteristic data includes the second characteristic data and the average temperature of the test mold in R partitions; Constructing a recurrent neural network, taking the third feature data in the thickness detection training set as the input of the recurrent neural network, and taking the thickness distribution data as the output, training the recurrent neural network to obtain an initial recurrent neural network; The initial recurrent neural network is model verified using the thickness detection test set, and an initial recurrent neural network with an output value less than or equal to a preset test error threshold is used as the trained second machine learning model.

[0011] Furthermore, the step of judging whether the panel blank meets the conversion conditions for entering the next cooling step includes: Extract the target mold average temperature of R partitions after blow molding based on the ideal temperature field ; For each zone, calculate the actual average mold temperature and target mold temperature The absolute difference between ; The temperature deviation Compare with the preset temperature difference tolerance interval ΔT; like , then mark the corresponding partition as a qualified partition; if , then the corresponding partition is marked as a non-compliant partition; Count the number of all partitions that do not meet the standard , and compare it with the preset area quantity interval ΔS; like , it is determined that the panel blank meets the conversion conditions, which means that the panel blank meets the requirements for entering the next cooling stage; like , it is determined that the panel blank does not meet the conversion conditions, which means that the panel blank does not meet the requirements for entering the next cooling stage.

[0012] Further, the obtaining of at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone includes: b1: For each partition, calculate all The deviation vector , call the established thermal influence coefficient matrix , It indicates the temperature change of the ith partition due to heat conduction or structural coupling when the temperature of the jth partition rises by 1°C; b2: Based on temperature deviation and the thermal influence coefficient matrix Calculate the thermal control contribution of each partition. The specific calculation formula is: ; b3: For all partitions The values ​​are sorted from large to small to form a ranking list Q of thermal control contribution; b4: Generate the target temperature control partition set S, initialize it to an empty set, and set the error tolerance threshold , initialize the partition and select index r as 1; b5: Iteratively select a target temperature adjustment partition and fit the solution of the temperature adjustment amount to obtain at least one target temperature adjustment partition and the temperature adjustment amount of the target temperature adjustment partition.

[0013] Furthermore, the iterative selection of the target temperature adjustment partition and the fitting of the solution of the temperature adjustment amount include: b51: Get the rth partition number from the sorted list Q , and add it to the target temperature control partition set S to obtain the updated target temperature control partition set S: , r is a positive integer; b52: According to the updated target temperature control partition set S, from the heat influence matrix H Extract the columns corresponding to S to form a submatrix ; b53: Based on the bias vector E and submatrices , and solve the temperature adjustment amount of the target temperature adjustment partition by the least squares method , the calculation formula is: ; Where: , represents the temperature adjustment amount of each target temperature adjustment partition in S, for The transposed matrix of , that is, the partitions added to the updated target temperature control partition set; b54: According to the obtained , calculate the deviation vector of the fit ; b55: Calculate the fitting residuals, ; b56: Determine the fitting residual Whether the error tolerance threshold is met ,like , then let r=r+1, and return to step b41 to continue selecting the next partition number from Q; if , the iteration is terminated, and the updated target temperature adjustment partition set S and its corresponding temperature adjustment amount are output .

[0014] In a second aspect, the present invention provides an edge-reinforced blow-molded panel temperature distribution control system, which is implemented based on the above-mentioned edge-reinforced blow-molded panel temperature distribution control method, and includes: The first acquisition module is used to predict the real mold temperature field of the mold after blow molding the panel blank according to the pre-trained first machine learning model in the blow molding process, and the real mold temperature field includes the real mold average temperature of R partitions on the mold after blow molding. , R is a positive integer; The second acquisition module is used to call the preconfigured genetic algorithm to obtain the ideal temperature field that the mold should present after blow molding the panel blank. The ideal temperature field includes the target mold average temperature of R partitions on the mold after blow molding. ; Judgment module, used to determine the actual mold temperature and target mold temperature , determine whether the panel blank meets the conversion conditions for entering the next cooling stage; the conversion conditions are the preset temperature difference tolerance interval ΔT and the area number interval Δ S ; An analysis module, used for obtaining at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone from all zones when the conversion condition is not met; The control module is used to blow-mold the panel blank after performing a heating or cooling operation on the target temperature adjustment zone according to the temperature adjustment amount to complete the blow-molding process.

[0015] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, any of the above-mentioned methods for controlling the temperature distribution of edge-reinforced blow-molded panels is implemented.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the method for controlling the temperature distribution of an edge-reinforced blow-molded panel described in any one of the above items is implemented.

[0017] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention constructs a first machine learning model and, based on multi-source data such as the physical properties of the panel blank, mold structure characteristics, and blow molding process conditions, achieves accurate prediction of the mold temperature field after blow molding, overcomes the problem of delayed temperature control response caused by the lack of thermal field evolution trend modeling in the prior art, and significantly improves the foresight and timeliness of the temperature control strategy.

[0018] The heat influence coefficient matrix was further introduced, and the deviation vector was combined with the thermal control contribution evaluation to establish a temperature control zone selection mechanism. The temperature control amount was solved using the least squares method, which fully considered the thermal coupling relationship between the mold partitions and solved the problems of the temperature control process relying on experience, ignoring thermal interference, and low adjustment accuracy in the existing technology.

[0019] By jointly optimizing the ideal temperature field through genetic algorithm and thickness prediction model, and performing differentiated temperature adjustment operations based on deviation judgment, the present invention can strengthen the temperature control of the edge area of ​​the mold, effectively avoid molding defects caused by insufficient or excessive temperature, such as warping, bubbles, streaks, etc., and significantly improve the edge molding quality, thickness uniformity and overall yield of the blow-molded panel. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0021] Figure 1 This is a flow chart of a method for controlling temperature distribution of an edge-reinforced blow-molded panel according to the present invention; Figure 2 A framework diagram of a temperature distribution control system for an edge-strengthened blow-molded panel according to the present invention; Figure 3 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0022] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description disclosed herein will be more comprehensive and complete, and the concepts of the example embodiments will be fully conveyed to those skilled in the art. The accompanying drawings are only schematic illustrations of the disclosure of the present application and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0023] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments disclosed in the present application. However, those skilled in the art will appreciate that the technical solutions disclosed in the present application may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring the various aspects disclosed in the present application.

[0024] Example 1 like Figure 1 As shown, this embodiment discloses a method for controlling temperature distribution of an edge-reinforced blow-molded panel, comprising: S101: In the blow molding process, according to the pre-trained first machine learning model, the real mold temperature field of the mold after blow molding the panel blank is predicted, and the real mold temperature field includes the real mold average temperature of R partitions on the mold after blow molding. , R is a positive integer; In implementation, the actual mold temperature field is obtained as follows: Obtaining the initial average mold temperature of the R partitions, and obtaining second characteristic data, wherein the second characteristic data includes attribute data of the panel blank, basic data of the mold, and parameter data of the blow molding process; Among them, the initial mold average temperature represents the average temperature in the partition, which is collected by the temperature sensor. The attribute data includes but is not limited to the initial blank temperature of the panel blank (collected by the temperature sensor), material type (such as PET, HDPE, PP, etc.), blank viscosity (which can be measured by the rotation method), specific heat capacity and blank volume / mass (determined according to manually entered data); the basic data includes but is not limited to the thermal conductivity of the mold (determined according to the type of mold, for example, the thermal conductivity of steel is 43 W / m·K, mold material (such as aluminum, steel), the number of partitions R and the partition distance of each partition (refers to the spatial distance between each partition and the panel blank inlet, which is determined according to historical input data); the parameter data includes but is not limited to the blowing pressure, blowing time and blowing flow, which are all monitored by various sensors, including but not limited to pressure sensors, timers and flow sensors; The second feature data and the initial mold average temperatures of the R partitions are input into the first machine learning model to obtain the real mold temperature field that the mold will present in the future after the panel blank is blow-molded; The training method of the first machine learning model is as follows: Acquire historical temperature detection training data, and divide the historical temperature detection training data into a temperature detection training set and a temperature detection test set, wherein the historical temperature detection training data includes the first feature data and its corresponding real mold temperature field, that is, the real mold average temperature of R partitions on the mold; Wherein, the first characteristic data includes the second characteristic data and the initial mold average temperature of R partitions; It should be understood that the R partitions on the mold are manually divided and determined in advance by the technicians according to different mold specifications (sizes), and all partitions are of the same size; for example, for a 500×300 hollow panel mold, it is evenly divided into 3×3 according to the vertical and horizontal coordinates, a total of 9 equal range partitions; It is worth noting that a corresponding temperature control unit is provided on the outer surface of each partition for adjusting the temperature change of the corresponding partition, and the temperature control unit includes but is not limited to components composed of a resistive heating belt (or an infrared heating unit) and a water cooling channel (or a thermoelectric cooling sheet (TEC)); It should be noted that the first characteristic data and the real mold temperature field in the historical temperature detection training data are actually collected and recorded by technicians based on experiments or historical real situations; Constructing a recurrent neural network, taking the first feature data in the temperature detection training set as the input of the recurrent neural network, and taking the real mold temperature field as the output, training the recurrent neural network to obtain an initial recurrent neural network; The temperature detection test set is used to verify the model of the initial recurrent neural network, and the initial recurrent neural network with a value less than or equal to the preset test error threshold is output as the trained first machine learning model; By predicting the actual mold average temperatures of the R partitions on the mold in advance based on the second characteristic data and the initial mold average temperatures of the R partitions, it is helpful to avoid the temperature control lag problem caused by the lack of temperature evolution trend modeling in traditional temperature control strategies.

[0025] S102: Calling a preconfigured genetic algorithm to obtain an ideal temperature field that the mold should present after blow molding the panel blank, wherein the ideal temperature field includes the target mold average temperature of R partitions on the mold after blow molding. ; In implementation, the method for obtaining the ideal temperature field is as follows: a1: Initialize the population: randomly generate an original population, which contains x individuals, each of which represents a test mold temperature field, that is, a test mold average temperature set, including R partitioned test mold average temperatures, and x is an integer greater than zero; a2: Fitness evaluation: Under each individual (i.e., each test mold temperature field in the original population), obtain the thickness distribution data of the panel blank in the mold; input the thickness distribution data into the pre-built fitness function to calculate the fitness of each individual, and the thickness distribution data is the thickness of the panel blank in each partition of the mold; In genetic algorithms, fitness evaluation is one of the core steps, which is used to determine the fitness of each individual, that is, to determine how well each individual performs in a given problem; the design of the fitness function directly affects the efficiency of the algorithm and the quality of the final solution; In implementation, the thickness distribution data of the panel blank in the mold includes: Obtaining the average temperature of the test mold of the R partitions, and obtaining the second characteristic data under the average temperature of the test mold; Input the average temperature of the test mold of R partitions and the second characteristic data into the pre-trained second machine learning model to obtain the thickness distribution data of the panel blank; The training method of the second machine learning model is as follows: Acquire historical thickness detection training data, and divide the historical thickness detection training data into a thickness detection training set and a thickness detection test set, wherein the historical thickness detection training data includes the third feature data and its corresponding thickness distribution data; Wherein, the third characteristic data includes the second characteristic data and the average temperature of the test mold in R partitions; It should be noted that the third characteristic data and thickness distribution data in the historical thickness detection training data are actually collected and recorded by technicians based on experiments or historical real situations; Constructing a recurrent neural network, taking the third feature data in the thickness detection training set as the input of the recurrent neural network, and taking the thickness distribution data as the output, training the recurrent neural network to obtain an initial recurrent neural network; The initial recurrent neural network is model verified using the thickness detection test set, and the initial recurrent neural network with an output value less than or equal to a preset test error threshold is used as the trained second machine learning model; It should be understood that: the first (second) machine learning model is preferably a recurrent neural network (RNN) for fitting the nonlinear mapping relationship between features and prediction results. However, although RNN is used as an example, other machine learning models (such as multi-layer perceptron MLP, convolutional neural network CNN, ensemble learning model, etc.) can also be used to achieve similar effects, which can be flexibly selected according to the training effect and actual deployment requirements; The calculation formula of the pre-constructed fitness function is: ; Where: For fitness, is the thickness of the panel blank in the ith partition, is the number of partitions, is a (small arbitrary) positive real number used to avoid singularities; a3: Selection: Use the roulette method to select two individuals with high fitness in the original population as the father and mother; The roulette wheel method is a commonly used selection method, which is used in genetic algorithms to select individuals with higher fitness to enter the next generation. It simulates the process of roulette, and each individual obtains a corresponding "roulette wheel" area according to its fitness. The higher the fitness of the individual, the larger the corresponding area, and the higher the probability of being selected. a4: Crossover: Perform crossover operation on the father and mother to produce new individuals; It should be noted that the crossover operation on the paternal parent and the maternal parent is implemented based on a crossover operation, and the crossover operation includes but is not limited to one of a single-point crossover, a uniform crossover or a sequential crossover, etc.; a5: Mutation: Perform mutation operation on the new individuals to obtain Y new individuals, combine the Y new individuals into a new population, replace the original population with the new population, and return to step a2; In genetic algorithms, mutation operations are used to introduce genetic diversity and prevent the algorithm from falling into a local optimum. The mutation operation on new individuals is implemented by uniform mutation or Gaussian mutation. a6: Repeat the above steps a2 to a5 until the fitness of the individuals in the original population or the new population is greater than or equal to the preset fitness threshold, or the number of iterations is greater than or equal to the preset maximum number of iterations threshold, and output the test mold temperature field represented by the corresponding individual as the ideal temperature field, that is, the target mold average temperature of the R partitions on the mold after blow molding is obtained; For example, assuming that the maximum number of iterations is 100, the individual with the highest fitness in the current population and its fitness value are recorded after each iteration; if it is found that the fitness value has not changed significantly in a certain generation, it is considered that the convergence condition is met, the iteration is stopped, and the test mold temperature field represented by the corresponding individual is output as the ideal temperature field; Determining the ideal temperature field through preconfigured genetic algorithms is conducive to constructing the optimal target distribution of mold zone temperature control, achieving molding thickness control and defect suppression in the panel edge area; at the same time, it improves the system's intelligent optimization capabilities and temperature control accuracy, thereby improving finished product consistency and production yield.

[0026] S103: According to the actual mold temperature and target mold temperature , determine whether the panel blank meets the conversion conditions for entering the next cooling stage; the conversion conditions are the preset temperature difference tolerance interval ΔT and the area number interval Δ S ; In implementation, the step of judging whether the panel blank meets the conversion conditions for entering the next cooling step includes: Extract the target mold average temperature of R partitions after blow molding based on the ideal temperature field ; For each zone, calculate the actual average mold temperature and target mold temperature The absolute difference between ; The temperature deviation Compare with the preset temperature difference tolerance interval ΔT; in, , Tmin represents the minimum value of the temperature difference, and Tmax represents the maximum value of the temperature difference; like , then mark the corresponding partition as a qualified partition; if , then the corresponding partition is marked as a non-compliant partition; Count the number of all partitions that do not meet the standard , and compare it with the preset area number interval Δ S Make a comparison; in, , Smin represents the minimum number of regions, and Smax represents the maximum number of regions; like , it is determined that the panel blank meets the conversion conditions, which means that the panel blank meets the requirements for entering the next cooling stage; like , it is determined that the panel blank does not meet the conversion conditions, which means that the panel blank does not meet the requirements for entering the next cooling stage; It should be understood that the prior art usually adopts a uniform temperature strategy to control the temperature of the R partitions on the mold during the blow molding process, that is, the R partitions on the mold all use the same temperature. However, due to a series of reasons such as the inconsistent thermal conductivity of the mold material and the mold edge being farther away from the blow molding, this leads to a certain temperature difference between the edge area and the center area when the uniform temperature strategy is adopted, resulting in an uneven distribution of the actual mold temperature field. Therefore, if this situation is used in the cooling process, the thickness of the panel blanks in the edge area and the center area will be uneven, and defects such as warping, bubbles, and stripes will occur in the edge area, seriously affecting the product quality. Therefore, accurately controlling the temperature distribution of the panel during the blow molding stage is crucial to improving the pass rate of blow molded products.

[0027] S104: when the conversion condition is not met, obtaining at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone from all zones; In implementation, the step of obtaining at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone includes: b1: For each partition, calculate all The deviation vector , call the established thermal influence coefficient matrix , It indicates the temperature change (in degrees Celsius) of the ith partition due to heat conduction or structural coupling when the temperature of the jth partition rises by 1°C. The value of can be obtained through experimental data, finite element simulation or historical monitoring data in actual production process; It should be noted that: the thermal influence coefficient matrix It is obtained through heat conduction experiment method or finite element thermal simulation modeling method; for example, in the finite element thermal simulation modeling method, a three-dimensional heat conduction finite element simulation software (such as ANSYS, COMSOL Multiphysics) is used to establish a heat conduction model of the mold. The specific steps include: constructing a thermal simulation model containing parameters such as mold partition structure, material properties (such as thermal conductivity, specific heat capacity), boundary conditions (such as heat flow, cooling channel, air convection), etc.; applying a 1°C temperature disturbance to the jth partition in the simulation model; observing the steady-state or transient temperature change response ΔTi of other partitions; using ΔTi as Fill in the matrix with estimated values H , repeat the above operation for j = 1 to R to obtain the complete H , that is, the thermal influence coefficient matrix; For example, assume that a mold is divided into R = 4 temperature control zones of equal area (denoted as Z1, Z2, Z3, Z4), and the distribution is as follows: Due to structural coupling and heat conduction, when adjusting the temperature of a partition, the temperature of other adjacent or structurally connected partitions will also be affected. Assuming that the ANSYS heat conduction simulation platform is used, the specific impact is found as shown in Table 1: Table 1: Thermal impact data table

[0028] According to Table 1 above, the thermal influence coefficient matrix is ​​generated ; b2: Based on temperature deviation and the thermal influence coefficient matrix Calculate the thermal control contribution of each partition. The specific calculation formula is: ; It should be understood that: Represents the potential regulatory influence of the jth partition on the entire temperature deviation correction; its calculation method comprehensively considers the temperature deviation required for the target mold temperature fitting and the thermal coupling path strength , used to evaluate the extent to which adjusting the temperature of the jth partition will affect the temperature compensation of other partitions, thereby guiding the selection order of the priority temperature adjustment partitions; b3: For all partitions The values ​​are sorted from large to small to form a thermal control contribution ranking list Q, expressed as: Q = [q1, q2, ..., qR], where q1 represents the partition number with the highest thermal control contribution, q2 is the second, and so on; b4: Generate the target temperature control partition set S, initialize it to an empty set, and set the error tolerance threshold (Unit: degrees Celsius), used to control the acceptance range of the overall temperature difference fitting result, the initialization partition selects index r as 1; It should be understood that the error tolerance threshold It is used in the least squares fitting process to evaluate whether the predicted temperature adjustment can effectively fit the current temperature deviation vector E An overall fitting error tolerance of ; it is set empirically based on the statistical relationship between the historical fitting error and the actual product defect rate, or reversely set by the target temperature fitting accuracy; b5: iteratively select a target temperature control partition and fit the solution of the temperature control amount to obtain at least one target temperature control partition and the temperature control amount of the target temperature control partition; Specifically, the iterative selection of the target temperature adjustment partition and the fitting of the solution of the temperature adjustment amount include: b51: Get the rth partition number from the sorted list Q , and add it to the target temperature control partition set S to obtain the updated target temperature control partition set S: , r is a positive integer; b52: According to the updated target temperature control partition set S, from the heat influence matrix H Extract the columns corresponding to S to form a submatrix ; For example, continuing the assumption, the heat influence coefficient matrix ; The rows represent the temperature response partition numbers i=1,2,3,4; the columns represent the temperature rise in a certain partition j=1,2,3,4; and assuming that the updated target temperature adjustment partition set S={2,4}, that is, to adjust the temperature of the 2nd and 4th partitions, then extract the submatrices from the 2nd and 4th partitions ,get , indicating that from H Extract columns 2 and 4 (keep all rows) from The dimension of is R×|S|=4×2, and each column represents the thermal impact of adjusting the 2nd or 4th partition on each partition; b53: Based on the bias vector E and the submatrix , and solve the temperature adjustment amount of the target temperature adjustment partition by the least squares method , the calculation formula is: ; Where: , represents the temperature adjustment amount of each target temperature adjustment partition in S, for The transposed matrix of , that is, the partitions added to the updated target temperature control partition set; It can be understood that the least squares solution process is essentially to find a set of minimum temperature adjustment increments in the error space so that the thermal response after temperature adjustment is as close as possible to the expected temperature difference correction value, thereby minimizing the error; b54: According to the obtained , calculate the deviation vector of the fit ; b55: Calculate the fitting residuals, ; b56: Determine the fitting residual Whether the error tolerance threshold is met ,like , then let r=r+1, and return to step b41 to continue selecting the next partition number from Q; if , the iteration is terminated, and the updated target temperature adjustment partition set S and its corresponding temperature adjustment amount are output ; In an optional embodiment, when the panel blank meets the conversion conditions, the next cooling step is performed to cool the panel blank in the mold; It can be understood that when the panel blank meets the conversion conditions, it means that the actual mold average temperature of the panel blank in each partition of the mold is close to or equal to the target mold average temperature. Therefore, there is no need to adjust the temperature. After the panel blank under this temperature enters the cooling link for cooling treatment, the edge area of ​​the panel blank will not or will have a small amount of defects such as warping, bubbles, and stripes, so that the formed panel meets the factory standards.

[0029] S105: After performing a temperature increase or temperature decrease operation on the target temperature adjustment zone according to the temperature adjustment amount, blow molding is performed on the panel blank to complete the blow molding process; The temperature increase or decrease operation of the target temperature adjustment zone can be realized by controlling the temperature control unit corresponding to the zone. Specifically, if , then the resistance heating belt or infrared heating device is controlled to turn on; if , then open the water cooling channel or thermoelectric cooling sheet to cool down; It is understandable that temperature regulation involves heating or cooling operations. This is because the area near the inlet of the panel blank is close to the heat source, so the temperature is higher and usually requires cooling treatment, while the edge area of ​​the mold is far away from the heat source, so the temperature is lower and usually requires heating treatment. Compared with the temperature equalization strategy, the present invention takes into account the impact of various actual factors on the mold temperature uniformity, and realizes differentiated temperature control by adopting a zoned temperature control method, thereby achieving true mold temperature consistency, which is beneficial to ensuring the structural quality and strength of the edge area of ​​the panel after molding.

[0030] Example 2 like Figure 2 As shown, the part not described in detail in this embodiment is as shown in Example 1. This embodiment discloses a temperature distribution control system for an edge-reinforced blow-molded panel, including: The first acquisition module 201 is used to predict the real mold temperature field of the mold after blow molding the panel blank according to the pre-trained first machine learning model in the blow molding process, and the real mold temperature field includes the real mold average temperature of R partitions on the mold after blow molding. , R is a positive integer; The second acquisition module 202 is used to call the preconfigured genetic algorithm to obtain the ideal temperature field that the mold should present after blow molding the panel blank, and the ideal temperature field includes the target mold average temperature of R partitions on the mold after blow molding. ; The judgment module 203 is used to determine the actual mold temperature. and target mold temperature , determine whether the panel blank meets the conversion conditions for entering the next cooling stage; the conversion conditions are the preset temperature difference tolerance interval ΔT and the area number interval Δ S ; An analysis module 204 is used to obtain at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone from all zones when the conversion condition is not met; The control module 205 is used to blow-mold the panel blank after performing a heating or cooling operation on the target temperature adjustment zone according to the temperature adjustment amount, so as to complete the blow-molding process.

[0031] Example 3 See also Figure 3 As shown, this embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, any one of the edge-reinforced blow-molded panel temperature distribution control methods provided by the above methods is implemented.

[0032] Since the electronic device introduced in this embodiment is an electronic device used to implement the edge-reinforced blow-molded panel temperature distribution control method in the embodiment of this application, based on the edge-reinforced blow-molded panel temperature distribution control method introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as the technical personnel of this field implement the electronic device used by the edge-reinforced blow-molded panel temperature distribution control method in the embodiment of this application, it belongs to the scope of protection of this application.

[0033] Example 4 This embodiment discloses a computer-readable storage medium, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, any one of the edge-reinforced blow-molded panel temperature distribution control methods provided by the above methods is implemented.

[0034] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0035] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired network or a wireless network. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD) or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0036] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for controlling the temperature distribution of an edge-strengthened blow-molded panel, characterized in that: include: In the blow molding process, the real mold temperature field of the mold after blow molding of the panel blank is predicted based on the pre-trained first machine learning model, and the real mold temperature field includes the real mold average temperature of R partitions on the mold after blow molding. , R is a positive integer; Call the preconfigured genetic algorithm to obtain the ideal temperature field that the mold should present after blow molding the panel blank, the ideal temperature field includes the target mold average temperature of R partitions on the mold after blow molding. ; According to the actual mold temperature and target mold temperature , judging whether the panel blank meets the conversion conditions for entering the next cooling link; the conversion conditions are the preset temperature difference tolerance interval ΔT and the area quantity interval ΔS; When the conversion condition is not met, obtaining at least one target temperature adjustment zone and the temperature adjustment amount corresponding to the target temperature adjustment zone from all zones; After heating or cooling the target temperature adjustment zone according to the temperature adjustment amount, the panel blank is blow-molded to complete the blow-molding process.

2. The edge-strengthened blow-molded panel temperature distribution control method according to claim 1, characterized in that: The method for obtaining the actual mold temperature field is as follows: Obtaining the initial average mold temperature of the R partitions, and obtaining second characteristic data, wherein the second characteristic data includes attribute data of the panel blank, basic data of the mold, and parameter data of the blow molding process; Among them, the attribute data includes the initial blank temperature, material type, blank viscosity, specific heat capacity and blank volume / mass of the panel blank; the basic data includes the thermal conductivity coefficient of the mold, the mold material, the number of partitions R and the partition distance of each partition; the parameter data includes the blowing pressure, blowing time and blowing flow rate; The second feature data and the initial mold average temperatures of the R partitions are input into the first machine learning model to obtain the real mold temperature field that the mold will present in the future after the panel blank is blow-molded; The training method of the first machine learning model is as follows: Acquire historical temperature detection training data, and divide the historical temperature detection training data into a temperature detection training set and a temperature detection test set, wherein the historical temperature detection training data includes the first feature data and its corresponding real mold temperature field, that is, the real mold average temperature of R partitions on the mold; Wherein, the first characteristic data includes the second characteristic data and the initial mold average temperature of R partitions; Constructing a recurrent neural network, taking the first feature data in the temperature detection training set as the input of the recurrent neural network, and taking the real mold temperature field as the output, training the recurrent neural network to obtain an initial recurrent neural network; The temperature detection test set is used to verify the model of the initial recurrent neural network, and the output of the initial recurrent neural network that is less than or equal to the preset test error threshold is used as the trained first machine learning model.

3. The edge-strengthened blow-molded panel temperature distribution control method according to claim 2, characterized in that: The method for obtaining the ideal temperature field is as follows: a1: Initialize the population: randomly generate an original population, which contains x individuals, each of which represents a test mold temperature field, including the average temperature of the test mold in R partitions, and x is an integer greater than zero; a2: Fitness evaluation: Under each individual, obtain the thickness distribution data of the panel blank in the mold; input the thickness distribution data into the pre-built fitness function to calculate the fitness of each individual, wherein the thickness distribution data is the thickness of the panel blank in each partition of the mold; a3: Selection: Use the roulette method to select two individuals with high fitness in the original population as the father and mother; a4: Crossover: Perform crossover operation on the father and mother to produce new individuals; a5: Mutation: Perform mutation operation on the new individuals to obtain Y new individuals, combine the Y new individuals into a new population, replace the original population with the new population, and return to step a2; a6: Repeat steps a2 to a5 above until the fitness of individuals in the original population or the new population is greater than or equal to the preset fitness threshold, or the number of iterations is greater than or equal to the preset maximum number of iterations threshold, and output the test mold temperature field represented by the corresponding individual as the ideal temperature field, that is, the target mold average temperature of the R partitions on the mold after blow molding is obtained.

4. The edge-strengthened blow-molded panel temperature distribution control method according to claim 3, characterized in that: The thickness distribution data of the panel blank in the mold includes: Obtaining the average temperature of the test mold of the R partitions, and obtaining the second characteristic data under the average temperature of the test mold; Input the average temperature of the test mold of R partitions and the second characteristic data into the pre-trained second machine learning model to obtain the thickness distribution data of the panel blank; The training method of the second machine learning model is as follows: Acquire historical thickness detection training data, and divide the historical thickness detection training data into a thickness detection training set and a thickness detection test set, wherein the historical thickness detection training data includes the third feature data and its corresponding thickness distribution data; Wherein, the third characteristic data includes the second characteristic data and the average temperature of the test mold in R partitions; Constructing a recurrent neural network, taking the third feature data in the thickness detection training set as the input of the recurrent neural network, and taking the thickness distribution data as the output, training the recurrent neural network to obtain an initial recurrent neural network; The initial recurrent neural network is model verified using the thickness detection test set, and an initial recurrent neural network with an output value less than or equal to a preset test error threshold is used as the trained second machine learning model.

5. The edge-strengthened blow-molded panel temperature distribution control method according to claim 1, characterized in that: The step of judging whether the panel blank meets the conversion conditions for entering the next cooling stage includes: Extract the target mold average temperature of R partitions after blow molding based on the ideal temperature field ; For each zone, calculate the actual average mold temperature and target mold temperature The absolute difference between ; The temperature deviation Compare with the preset temperature difference tolerance interval ΔT; like , then mark the corresponding partition as a qualified partition; if , then the corresponding partition is marked as a non-compliant partition; Count the number of all partitions that do not meet the standard , and compare it with the preset area quantity interval ΔS; like , it is determined that the panel blank meets the conversion conditions, which means that the panel blank meets the requirements for entering the next cooling stage; like , it is determined that the panel blank does not meet the conversion conditions, which means that the panel blank does not meet the requirements for entering the next cooling stage.

6. The edge-strengthened blow-molded panel temperature distribution control method according to claim 5, characterized in that: The obtaining of at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone includes: b1: For each partition, calculate all The deviation vector , call the established thermal influence coefficient matrix , It indicates the temperature change of the ith partition due to heat conduction or structural coupling when the temperature of the jth partition rises by 1°C; b2: Based on temperature deviation and the thermal influence coefficient matrix Calculate the thermal control contribution of each partition. The specific calculation formula is: ; b3: For all partitions The values ​​are sorted from large to small to form a ranking list Q of thermal control contribution; b4: Generate the target temperature control partition set S, initialize it to an empty set, and set the error tolerance threshold , initialize the partition to select index r as 1; b5: Iteratively select a target temperature adjustment partition and fit the solution of the temperature adjustment amount to obtain at least one target temperature adjustment partition and the temperature adjustment amount of the target temperature adjustment partition.

7. The edge-strengthened blow-molded panel temperature distribution control method according to claim 6, characterized in that: The iterative selection of the target temperature adjustment partition and the fitting of the solution of the temperature adjustment amount include: b51: Get the rth partition number from the sorted list Q , and add it to the target temperature control partition set S to obtain the updated target temperature control partition set S: , r is a positive integer; b52: According to the updated target temperature control partition set S, from the heat influence matrix H Extract the columns corresponding to S to form a submatrix ; b53: Based on the bias vector E and the submatrix , and solve the temperature adjustment amount of the target temperature adjustment partition by the least squares method , the calculation formula is: ; Where: , represents the temperature adjustment amount of each target temperature adjustment partition in S, for The transposed matrix of , that is, the partitions added to the updated target temperature control partition set; b54: According to the obtained , calculate the deviation vector of the fit ; b55: Calculate the fitting residuals, ; b56: Determine the fitting residual Whether the error tolerance threshold is met ,like , then let r=r+1, and return to step b41 to continue selecting the next partition number from Q; if , the iteration is terminated, and the updated target temperature adjustment partition set S and its corresponding temperature adjustment amount are output .

8. A temperature distribution control system for edge-strengthened blow-molded panels, implemented based on the temperature distribution control method for edge-strengthened blow-molded panels according to any one of claims 1 to 7, characterized in that: include: The first acquisition module is used to predict the real mold temperature field of the mold after blow molding the panel blank according to the pre-trained first machine learning model in the blow molding process, and the real mold temperature field includes the real mold average temperature of R partitions on the mold after blow molding. , R is a positive integer; The second acquisition module is used to call the preconfigured genetic algorithm to obtain the ideal temperature field that the mold should present after blow molding the panel blank. The ideal temperature field includes the target mold average temperature of R partitions on the mold after blow molding. ; Judgment module, used to determine the actual mold temperature and target mold temperature , determine whether the panel blank meets the conversion conditions for entering the next cooling stage; the conversion conditions are the preset temperature difference tolerance interval ΔT and the area number interval Δ S ; An analysis module, used for obtaining at least one target temperature adjustment zone and a temperature adjustment amount corresponding to the target temperature adjustment zone from all zones when the conversion condition is not met; The control module is used to blow-mold the panel blank after performing a heating or cooling operation on the target temperature adjustment zone according to the temperature adjustment amount to complete the blow-molding process.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the edge reinforcement blow molding panel temperature distribution control method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method for controlling the temperature distribution of an edge-reinforced blow-molded panel according to any one of claims 1 to 7 is implemented.

Images