A method and system for optimizing digital twin parameters in the pressing and molding of Pingli Gynostemma pentaphyllum Fu brick tea.

Abstract

This application relates to a digital twin parameter optimization method and system for pressing Pingli Gynostemma pentaphyllum Fu brick tea. The method includes: acquiring raw material characteristic parameters, press process parameters, and forming quality indicators during the brick tea pressing process; constructing an initial mapping relationship between the raw material characteristic parameters, press process parameters, and forming quality indicators; establishing a digital twin model of the brick tea pressing equipment; calibrating the digital twin model using the initial mapping relationship; acquiring the raw material characteristic parameters and target quality distribution of the current batch; and using the calibrated digital twin model, minimizing the spatial deviation between the pressing-related parameters of each spatial node and the target quality distribution as the optimization objective. This method, by minimizing the spatial deviation between the pressing-related parameters of each spatial node and the target quality distribution as the optimization objective, allows quality indicators at different locations to independently participate in the objective function, achieving uniform control of the overall quality of the tea brick.

Description

Technical Field

[0001] This application relates to the field of parameter optimization technology, and in particular to a digital twin parameter optimization method and system for pressing and molding Pingli Gynostemma pentaphyllum Fu brick tea. Background Technology

[0002] With the rapid development of digital twin and simulation technologies, digital twin models are increasingly widely used in the field of food processing technology optimization. The introduction of techniques such as discrete element method, finite element method, and moisture content field modeling has provided a technical foundation for the virtual simulation of brick tea pressing and forming process, making it possible to pre-evaluate process parameters. However, in actual production, there is a complex multi-physics coupling relationship between the flow characteristics of tea-Gynostemma pentaphyllum mixed particles and the deformation of the mold-pressing head structure and moisture migration. Most existing pressing and forming simulation methods only focus on a single physical field, with discrete element analysis and finite element structural analysis being independent of each other. They lack a complete technical framework for collaboratively modeling particle flow, structural deformation, and moisture content changes, resulting in simulation results that fail to accurately reflect the multi-field interaction effects during the pressing process and insufficient accuracy in predicting the spatial distribution of density and moisture content.

[0003] In related technologies, the impact of process parameters on product quality is usually evaluated using the overall mean, treating tea bricks as macroscopic homogeneous bodies. This lacks the ability to establish a spatial mapping relationship between pressing-related parameters and equipment control parameters using three-dimensional geometric model spatial nodes as anchor points. This makes it impossible to predict the differences in density and moisture content at different locations, and when raw material batch characteristics fluctuate, the uniformity of the whole brick's quality is difficult to guarantee. Furthermore, existing optimization methods mostly rely on trial and error or single-objective overall optimization, failing to minimize the spatial deviation between the pressing-related parameters of each spatial node and the target quality distribution as a point-by-point optimization objective. Quality indicators at different locations cannot independently participate in the objective function, and local quality defects in key areas cannot be effectively identified and corrected. Moreover, the allowable adjustment range of equipment control parameters is not considered as a process constraint during optimization, posing a risk of exceeding limits. The engineering feasibility of the optimal parameter combination is insufficient, making it difficult to directly implement. Summary of the Invention

[0004] Therefore, it is necessary to provide a digital twin parameter optimization method and system for pressing and molding Pingli Gynostemma pentaphyllum Fu brick tea to address the above-mentioned technical problems.

[0005] Firstly, this application provides a digital twin parameter optimization method for the pressing and molding of Pingli Gynostemma pentaphyllum Fu brick tea, including:

[0006] Obtain the raw material characteristic parameters, press process parameters, and molding quality indicators during the brick tea pressing process, and construct an initial mapping relationship between the raw material characteristic parameters, the press process parameters, and the molding quality indicators;

[0007] A three-dimensional geometric model of the tea brick blank is established, and the three-dimensional geometric model is divided into multiple spatial nodes. A digital twin model of the brick tea pressing and forming equipment is established. The digital twin model includes a discrete element model of tea material-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, a moisture content field model inside the tea brick, and a correlation model between the pressing correlation parameters of the tea brick blank and the equipment control parameters. The correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish a spatial mapping relationship between the pressing correlation parameters and the equipment control parameters at each spatial node.

[0008] The digital twin model is calibrated using the initial mapping relationship;

[0009] Obtain the characteristic parameters and target quality distribution of the current batch of raw materials. Using the calibrated digital twin model, with the optimization objective of minimizing the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution, and with the allowable adjustment range of the equipment control parameters as the constraint, solve for the optimal combination of equipment control parameters.

[0010] The optimal combination of equipment control parameters is sent to the compression molding equipment for execution.

[0011] In one embodiment, the initial mapping relationship between the raw material characteristic parameters, the press process parameters, and the molding quality indicators includes:

[0012] Obtain raw material characteristic parameters, press process parameters, and corresponding molding quality index data recorded in historical production batches. The raw material characteristic parameters include tea moisture content, Gynostemma pentaphyllum ratio, particle size distribution, and mixing uniformity. The press process parameters include pressing force, pressing speed, holding time, mold temperature, and demolding speed. The molding quality indexes include uniformity of tea brick density distribution and uniformity of moisture content.

[0013] The acquired raw material characteristic parameters, press process parameters, and molding quality index data were normalized, and abnormal samples were removed.

[0014] Using raw material characteristic parameters and press process parameters as input features, and molding quality indicators as output labels, a multiple linear regression model is used to train an initial mapping model to obtain the initial mapping relationship between the raw material characteristic parameters, the press process parameters, and the molding quality indicators.

[0015] In one embodiment, establishing the discrete element model of the tea-Gynostemma pentaphyllum mixture includes:

[0016] The particle size distribution, particle density, elastic modulus, friction coefficient, and cohesive force parameters of tea granules and Gynostemma pentaphyllum granules were obtained.

[0017] In the discrete element simulation software, a set of tea particles and a set of Gynostemma pentaphyllum particles are generated according to the particle size distribution, and then mixed in a virtual mold according to a preset ratio.

[0018] Contact models are set for tea granules and Gynostemma pentaphyllum granules respectively. The contact models include the normal contact stiffness, tangential contact stiffness, rolling friction coefficient and interparticle bonding strength.

[0019] The parameters of the established discrete element model were calibrated. By comparing the simulated and measured repose angles, the contact parameters were adjusted until they were consistent, thus obtaining the discrete element model of the tea-Gynostemma pentaphyllum mixed particles.

[0020] In one embodiment, establishing the mold-indenter finite element model includes:

[0021] Obtain the geometric dimensions, material properties, and assembly relationship of the mold and the indenter, wherein the material properties include elastic modulus, Poisson's ratio, and yield strength;

[0022] Finite element preprocessing software was used to mesh the mold and the pressure head, and contact elements were set on the contact surface between the mold and the pressure head.

[0023] Apply mold fixing constraints and indenter movement constraints, and apply pressing force load to the top of the indenter;

[0024] By setting material nonlinearity and geometric nonlinearity options, a finite element model of the mold-pressing head is established to calculate the stress distribution and deformation of the mold and pressing head during the pressing process.

[0025] In one embodiment, establishing the internal moisture content field model of the tea brick includes:

[0026] Treating the tea brick blank as a porous medium, a convection-diffusion equation is established to show the change of moisture content with time and space during the pressing process. The moisture migration rate is related to the local pressure gradient, temperature gradient and moisture content gradient.

[0027] The initial moisture content, saturated moisture content, and moisture diffusion coefficient of the tea-Gynostemma pentaphyllum mixture were obtained.

[0028] Set the mold wall as the water discharge boundary, and set the contact surface between the pressure head and the tea blank as the dynamic pressure boundary;

[0029] The convection-diffusion equation is solved discretically using the finite volume method to obtain the change in moisture content of each spatial node inside the tea brick over time during the pressing process, thus forming a moisture content field model inside the tea brick.

[0030] In one embodiment, establishing the correlation model between the pressing parameters of the tea brick blank and the equipment control parameters includes:

[0031] The three-dimensional geometric model of the tea brick blank is divided into multiple spatial nodes, and each spatial node corresponds to a local position.

[0032] Using the aforementioned discrete element model of tea-Gynostemma pentaphyllum mixed particles and finite element model of mold-pressing head, simulations were performed for multiple sets of different equipment control parameter combinations. Pressing correlation parameters at each spatial node were extracted, including local pressure, local density, and local moisture content.

[0033] Using equipment control parameters as input and suppression correlation parameters at each spatial node as output, a spatial node-level surrogate model is established. The surrogate model adopts a radial basis function neural network.

[0034] The proxy models of each spatial node are combined according to the location of the spatial node to form an association model that uses the spatial nodes of the three-dimensional geometric model as anchor points and represents the spatial mapping relationship between the suppression association parameters and the equipment control parameters at each spatial node.

[0035] In one embodiment, calibrating the digital twin model using the initial mapping relationship includes:

[0036] Obtain the characteristic parameters of the current batch of raw materials and the actual press process parameters, and use the initial mapping relationship to calculate the corresponding predicted values ​​of molding quality indicators;

[0037] The press process parameters are input into the digital twin model for simulation to obtain the spatial distribution of pressing-related parameters and the simulation results of moisture content field of tea brick blank;

[0038] The molding quality index predicted by the digital twin model is calculated based on the simulation results, and compared with the predicted value of the molding quality index calculated by the initial mapping relationship to calculate the deviation.

[0039] With the goal of minimizing the deviation, the particle contact parameters, moisture diffusion coefficient, and mold-indenter contact stiffness in the digital twin model are adjusted using the parameter inversion method until the output of the digital twin model is consistent with the output of the initial mapping relationship within the allowable deviation range, thus completing the calibration of the digital twin model.

[0040] In one embodiment, the step of using the calibrated digital twin model, with the optimization objective of minimizing the spatial deviation between the suppression correlation parameters of each spatial node and the target quality distribution, and with the allowable adjustment range of the equipment control parameters as constraints, to solve for the optimal combination of equipment control parameters includes:

[0041] Obtain the characteristic parameters and target mass distribution of the raw materials in the current batch, wherein the target mass distribution is the expected density value and expected moisture content value of each spatial node of the tea brick;

[0042] Input the characteristic parameters of the current batch of raw materials into the calibrated digital twin model, and establish an optimization objective function with the equipment control parameters as decision variables. The optimization objective function is the weighted norm of the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution.

[0043] Set the allowable adjustment range constraints for the equipment control parameters, including the upper and lower limits of pressing force, pressing speed, holding time, mold temperature, and demolding speed;

[0044] Within the allowable adjustment range, the device control parameters are iteratively searched. In each iteration, the digital twin model is called to calculate the suppression correlation parameters of each spatial node and evaluate the optimization objective function value until the optimization objective function value converges to the preset threshold, and the optimal combination of device control parameters is output.

[0045] In one embodiment, the step of sending the optimal combination of equipment control parameters to the compression molding equipment for execution includes:

[0046] The pressing force, pressing speed, holding time, mold temperature and demolding speed in the optimal equipment control parameter combination are respectively converted into control command values ​​that the pressing and molding equipment can execute;

[0047] The control command values ​​are sent to the programmable logic controller of the pressing and molding equipment through the equipment control interface. The programmable logic controller controls the press hydraulic system, mold heating system and demolding mechanism to operate according to the command values.

[0048] During the pressing and molding process, the actual operating parameters of the equipment are collected in real time and compared with the control command values ​​issued. If a deviation occurs, fine-tuning is performed through the feedback control loop.

[0049] Secondly, this application also provides a digital twin parameter optimization system for the pressing and molding of Pingli Gynostemma pentaphyllum Fu brick tea, including:

[0050] The parameter mapping module is used to obtain the raw material characteristic parameters, press process parameters and forming quality indicators during the brick tea pressing process, and to construct the initial mapping relationship between the raw material characteristic parameters, the press process parameters and the forming quality indicators.

[0051] The digital modeling module is used to establish a three-dimensional geometric model of the tea brick blank and divide the three-dimensional geometric model into multiple spatial nodes; it also establishes a digital twin model of the brick tea pressing and forming equipment, which includes a discrete element model of tea material-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, a moisture content field model inside the tea brick, and a correlation model between the pressing correlation parameters of the tea brick blank and the equipment control parameters; wherein, the correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish a spatial mapping relationship between the pressing correlation parameters and the equipment control parameters at each spatial node;

[0052] A model calibration module is used to calibrate the digital twin model using the initial mapping relationship;

[0053] The parameter optimization module is used to obtain the characteristic parameters and target quality distribution of the current batch of raw materials. Using the calibrated digital twin model, the optimization objective is to minimize the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution. The optimal combination of equipment control parameters is solved with the allowable adjustment range of equipment control parameters as the constraint.

[0054] The equipment control module is used to send the optimal combination of equipment control parameters to the compression molding equipment for execution.

[0055] The aforementioned digital twin parameter optimization method and system for pressing Pingli Gynostemma pentaphyllum Fu brick tea integrates a discrete element model of tea-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, and a moisture content field model within the tea brick. It also considers the interplay of particle flow, structural deformation, and moisture migration, enabling the digital twin model to possess multi-physics co-simulation capabilities. Furthermore, by using spatial nodes of the three-dimensional geometric model as anchor points, it establishes a spatial mapping relationship between pressing-related parameters and equipment control parameters at each node, allowing key indicators such as density and moisture content to have differentiated predictive capabilities at different locations within the tea brick, overcoming the bias caused by the assumption of macroscopic homogeneity. Subsequently, by minimizing the spatial deviation between the pressing-related parameters of each spatial node and the target quality distribution as the optimization objective, the quality indicators at different locations independently participate in the objective function, achieving uniform control of the overall quality of the tea brick. Moreover, through iterative search within the allowable adjustment range of the equipment control parameters, the digital twin model is called each time to evaluate the objective function value until convergence. The output optimal parameter combination is constrained by process limitations, ensuring direct execution without the risk of exceeding limits. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a flowchart illustrating a digital twin parameter optimization method for pressing Pingli Gynostemma pentaphyllum brick tea in one embodiment.

[0058] Figure 2 This is a structural block diagram of a digital twin parameter optimization system for pressing Pingli Gynostemma pentaphyllum brick tea, as shown in one embodiment. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0060] In one exemplary embodiment, such as Figure 1 As shown, a digital twin parameter optimization method for pressing Pingli Gynostemma pentaphyllum Fu brick tea is provided, including:

[0061] Step S102: Obtain the raw material characteristic parameters, press process parameters, and molding quality indicators during the brick tea pressing process, and construct the initial mapping relationship between the raw material characteristic parameters, press process parameters, and molding quality indicators.

[0062] For example, raw material characteristic parameters, press process parameters, and corresponding molding quality index data recorded in historical production batches are obtained. The raw material characteristic parameters include tea moisture content, Gynostemma pentaphyllum ratio, particle size distribution, and mixing uniformity. The press process parameters include pressing force, pressing speed, holding time, mold temperature, and demolding speed. The molding quality indexes include the uniformity of tea brick density distribution and moisture content. The obtained raw material characteristic parameters, press process parameters, and molding quality index data are normalized, and outlier samples are removed. Using the raw material characteristic parameters and press process parameters as input features and the molding quality indexes as output labels, a multiple linear regression model is used to train the initial mapping model to obtain the initial mapping relationship between the raw material characteristic parameters, press process parameters, and molding quality indexes.

[0063] In this embodiment, raw material inspection records, press operation logs, and quality inspection data are extracted from the production history database and associated by batch number to form a complete sample set containing raw material characteristics, process parameters, and quality indicators. Z-score standardization is applied to continuous features, transforming the data of each dimension to a distribution with a mean of zero and a standard deviation of one. Records with missing values ​​or exceeding physically reasonable ranges in the quality indicators are deleted. Subsequently, using raw material characteristic parameters and press process parameters as input feature matrices and molding quality indicators as output labels, the least squares method is used to estimate regression coefficients, training an initial linear mapping model. After model training, the goodness of fit and residual distribution are calculated on the validation set to confirm that the initial mapping relationship can basically reflect the changing trend between input and output. The model parameters are then saved to a digital twin parameter library.

[0064] Step S104: Establish a three-dimensional geometric model of the tea brick blank and divide the three-dimensional geometric model into multiple spatial nodes; establish a digital twin model of the brick tea pressing and forming equipment. The digital twin model includes a discrete element model of tea material-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, a moisture content field model inside the tea brick, and a correlation model between the pressing correlation parameters of the tea brick blank and the equipment control parameters; wherein, the correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish a spatial mapping relationship between the pressing correlation parameters and the equipment control parameters at each spatial node.

[0065] Among them, the discrete element model of tea-Gynostemma pentaphyllum mixed particles is used to simulate the flow, rearrangement and compression behavior of mixed particles during the pressing process; the finite element model of mold-pressing head is used to simulate the deformation and stress distribution of mold and pressing head under pressing force; the internal moisture content field model of tea brick is used to simulate moisture migration and spatial distribution of moisture content during the pressing process; pressing-related parameters include local density, local moisture content and local pressure; equipment control parameters include pressing force, pressing speed, holding time, mold temperature and demolding speed; the correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish the spatial mapping relationship between pressing-related parameters and equipment control parameters at each spatial node.

[0066] For example, a discrete element model of tea leaves and Gynostemma pentaphyllum mixed particles is constructed using discrete element software, and particle contact parameters are calibrated through angle of repose test and uniaxial compression test; a finite element model of mold and press head is established, a structured mesh is divided and constraints are applied; a moisture content field model inside the tea brick is established, and the moisture diffusion equation of porous media is used to simulate moisture migration during the pressing process; the three-dimensional geometric model is meshed, and each mesh node is used as a spatial anchor point to record pressing-related parameters such as local density and local moisture content at that node, and the mapping relationship between them and equipment control parameters is calibrated through simulation or data-driven methods and stored in a digital twin model database.

[0067] For example, a discrete element model of tea-Gynostemma pentaphyllum mixed particles is established, including: obtaining the particle size distribution, particle density, elastic modulus, friction coefficient, and cohesive force parameters of tea particles and Gynostemma pentaphyllum particles; generating a set of tea particles and a set of Gynostemma pentaphyllum particles in discrete element simulation software according to the particle size distribution, and mixing them in a virtual mold according to a preset ratio; setting contact models for tea particles and Gynostemma pentaphyllum particles respectively, the contact models including the normal contact stiffness, tangential contact stiffness, rolling friction coefficient, and interparticle bonding strength; calibrating the parameters of the established discrete element model, and adjusting the contact parameters until they are consistent by comparing the simulated angle of repose with the measured angle of repose, thus obtaining the discrete element model of tea-Gynostemma pentaphyllum mixed particles.

[0068] In this embodiment, measured particle size distribution data of tea leaves and Gynostemma pentaphyllum are imported into the discrete element simulation software, and physical property parameters such as particle density, elastic modulus, friction coefficient, and cohesion are set. The two types of particles are mixed and stacked in a virtual mold according to a preset ratio. Contact models are specified for the two types of particles, and normal and tangential contact stiffness, rolling friction coefficient, and bonding strength are set. The angle of repose is simulated and compared with the measured angle of repose in the laboratory. If the deviation exceeds the limit, the contact parameters are adjusted until they are consistent, thus completing the establishment of the discrete element model of the tea leaf-Gynostemma pentaphyllum mixed particles.

[0069] For example, establishing a mold-indenter finite element model includes: obtaining the geometric dimensions, material properties, and assembly relationships of the mold and indenter, where material properties include elastic modulus, Poisson's ratio, and yield strength; meshing the mold and indenter using finite element preprocessing software and setting contact elements on the contact surface between the mold and indenter; applying mold fixing constraints and indenter motion constraints, and applying pressing force load to the top of the indenter; setting material nonlinearity options and geometric nonlinearity options to establish a mold-indenter finite element model for calculating the stress distribution and deformation of the mold and indenter during the pressing process.

[0070] In this embodiment, a three-dimensional geometric model of the mold and the pressure head is obtained, imported into finite element preprocessing software, and material properties are defined and the elastic modulus, Poisson's ratio, and yield strength are set. The mold and the pressure head are meshed separately, and contact pairs are defined on their contact surfaces and friction coefficients are set. A fixed constraint is applied to the bottom surface of the mold, a motion constraint is applied to the pressure head along the pressing direction, and a pressing force load is applied to the top of the pressure head. The material nonlinearity and geometric nonlinearity solution options are enabled to complete the establishment of the mold-pressure head finite element model, which is used to calculate the stress distribution and deformation of the mold and the pressure head during the pressing process.

[0071] For example, establishing a moisture content field model inside a tea brick includes: treating the tea brick blank as a porous medium, establishing a convection-diffusion equation for the change of moisture content with time and space during the pressing process, wherein the moisture migration rate is related to the local pressure gradient, temperature gradient, and moisture content gradient; obtaining the initial moisture content, saturated moisture content, and moisture diffusion coefficient of the tea-Gynostemma pentaphyllum mixture; setting the mold wall as the moisture discharge boundary and the contact surface between the press head and the tea blank as the dynamic pressure boundary; and using the finite volume method to discretize and solve the convection-diffusion equation to obtain the change of moisture content of each spatial node inside the tea brick with time during the pressing process, thus forming a moisture content field model inside the tea brick.

[0072] In this embodiment, the tea brick blank is discretized into a porous three-dimensional mesh, and the initial moisture content, saturated moisture content, and moisture diffusion coefficient of the tea-Gynostemma pentaphyllum mixture are read. The mold wall is set as the moisture discharge boundary, and the contact surface between the press head and the tea blank is set as the dynamic pressure boundary. The finite volume method is used to discretize the moisture content convection-diffusion equation, and the moisture content of each mesh node is solved iteratively at each time step. By gradually advancing the time, the moisture content change of each node is recorded, forming the moisture content field distribution inside the tea brick during the pressing process.

[0073] For example, establishing a correlation model between the pressing parameters of the tea brick blank and the equipment control parameters includes: dividing the three-dimensional geometric model of the tea brick blank into multiple spatial nodes, each spatial node corresponding to a local location; using a discrete element model of tea-Gynostemma pentaphyllum mixed particles and a finite element model of mold-pressing head, simulating multiple combinations of different equipment control parameters, and extracting the pressing parameters at each spatial node, including local pressure, local density, and local moisture content; using the equipment control parameters as input and the pressing parameters at each spatial node as output, establishing a surrogate model at the spatial node level, the surrogate model adopting a radial basis function neural network; combining the surrogate models of each spatial node according to the spatial node position to form a correlation model with the spatial nodes of the three-dimensional geometric model as anchor points, representing the spatial mapping relationship between the pressing parameters and the equipment control parameters at each spatial node.

[0074] In this embodiment, the three-dimensional geometric model of the tea brick blank is divided into several spatial nodes, and the spatial coordinates of each node are recorded. Then, under multiple different combinations of equipment control parameters, the discrete element model of the tea-Gynostemma pentaphyllum mixture particles and the finite element model of the mold-pressing head are jointly invoked for simulation. The local pressure, local density, and local moisture content at each spatial node are extracted from the simulation results. Next, using the equipment control parameters as input features and the pressing correlation parameters of each spatial node as output labels, a radial basis function neural network surrogate model is trained for each spatial node. The surrogate model adopts a radial basis function neural network. The radial basis function neural network adopts a single hidden layer structure with 20-50 neurons in the hidden layer. The radial basis function is a Gaussian function, and the width parameter is determined according to the average distance between training samples. The output layer adopts linear weighted summation. The least squares method is used to determine the weight of the output layer during training. Finally, the surrogate models of all spatial nodes are organized according to their spatial positions to form a correlation model with the spatial nodes of the three-dimensional geometric model as anchor points, so that the spatial distribution of the pressing correlation parameters of each node inside the tea brick blank can be predicted given the equipment control parameters.

[0075] Step S106: The digital twin model is calibrated using the initial mapping relationship.

[0076] The digital twin model contains multiple sub-models and undetermined internal parameters. When it is built, there are idealized assumptions, which need to be calibrated to make it consistent with the actual production process. Calibration uses the initial mapping relationship as a priori constraint or reference benchmark to adjust the undetermined parameters in the digital twin model, so that the simulation output of the model under the same raw material and process inputs can approach the historical molding quality index, thereby improving the prediction accuracy of the digital twin model.

[0077] For example, a calibration dataset is selected from historical batches. Using the raw material characteristic parameters and press process parameters of each batch as input, the discrete element model, finite element model, moisture content field model, and correlation model in the digital twin model are run sequentially to obtain predicted values ​​of the molding quality indicators. The predicted values ​​are compared with the initial mapping relationship output or historical measured quality indicators to construct a deviation function between the two. Using a genetic algorithm or particle swarm optimization algorithm, the contact parameters of the discrete element model, the material parameters of the finite element model, the moisture content diffusion coefficient, and the weights of the radial basis function neural network in the correlation model are adjusted. The deviation function is iteratively minimized until the deviation converges to below a preset threshold, thus completing the calibration of the digital twin model.

[0078] For example, the characteristic parameters of the current batch of raw materials and the actual press process parameters are obtained, and the corresponding predicted values ​​of molding quality indicators are calculated using the initial mapping relationship. The press process parameters are input into the digital twin model for simulation to obtain the spatial distribution of pressing-related parameters and the simulation results of moisture content field of the tea brick blank. The molding quality indicators predicted by the digital twin model are calculated based on the simulation results and compared with the predicted values ​​of molding quality indicators calculated by the initial mapping relationship to calculate the deviation. With the goal of minimizing the deviation, the particle contact parameters, moisture diffusion coefficient and mold-press head contact stiffness in the digital twin model are adjusted using the parameter inversion method until the output of the digital twin model is consistent with the output of the initial mapping relationship within the allowable deviation range, thus completing the calibration of the digital twin model.

[0079] In this embodiment, the characteristics of the current batch of raw materials and the actual press process parameters are obtained. First, the predicted values ​​of the molding quality indicators are calculated based on the initial mapping relationship. Then, the same set of process parameters are input into the digital twin model, and discrete element, finite element, and moisture content field simulations are run to obtain the internal pressure, density, and moisture content distribution of the tea brick body. The molding quality indicators predicted by the digital twin model are then calculated. The two are compared to obtain the deviation. With the goal of minimizing the deviation, a genetic algorithm or gradient descent method is used to invert and adjust the particle contact parameters, moisture diffusion coefficient, and mold-press head contact stiffness. The simulation and parameter updates are repeated until the output of the digital twin model and the output of the initial mapping relationship are consistent within the allowable deviation range, thus completing the calibration.

[0080] Step S108: Obtain the characteristic parameters and target quality distribution of the current batch of raw materials. Using the calibrated digital twin model, with the optimization objective of minimizing the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution, and with the allowable adjustment range of the equipment control parameters as the constraint, solve for the optimal combination of equipment control parameters.

[0081] Among them, the pressing correlation parameters of each spatial node are the local pressure, local density and local moisture content of each local position of the tea brick blank predicted by the digital twin model; the spatial deviation is the difference between the pressing correlation parameters of each spatial node and the target value of the corresponding node in the target quality distribution; the optimization objective is to minimize the comprehensive measure of the spatial deviation of all spatial nodes; the allowable adjustment range of the equipment control parameters is the adjustable upper and lower limits of each control variable of the press; the optimal combination of equipment control parameters is the set values ​​of pressing force, pressing speed, holding time, mold temperature and demolding speed that minimize the spatial deviation.

[0082] For example, the raw material characteristic parameters and target quality distribution of the current batch are read, and the target quality distribution is mapped to each spatial node of the digital twin model. Using equipment control parameters as decision variables, the weighted sum of deviations between the pressing-related parameters of each spatial node and the target value is used as the optimization objective function, with the upper and lower limits of the equipment control parameters as constraints. An iterative search is performed using a genetic algorithm or particle swarm optimization algorithm: each time a set of equipment control parameters is generated, the calibrated digital twin model is called to predict the pressing-related parameters of each spatial node, the spatial deviation objective function value is calculated, and the parameter combination is updated according to the optimization algorithm. This iteration is repeated until the objective function converges or the preset number of iterations is reached, outputting the optimal equipment control parameter combination to guide actual pressing production.

[0083] For example, the characteristic parameters and target quality distribution of the current batch of raw materials are obtained, where the target quality distribution is the expected density and expected moisture content values ​​of each spatial node of the tea brick. The characteristic parameters of the current batch of raw materials are input into the calibrated digital twin model. Using the equipment control parameters as decision variables, an optimization objective function is established. The optimization objective function is the weighted norm of the spatial deviation between the pressing correlation parameters of each spatial node and the target quality distribution. The allowable adjustment range constraints of the equipment control parameters are set, including the upper and lower limits of pressing force, pressing speed, holding time, mold temperature, and demolding speed. The equipment control parameters are iteratively searched within the allowable adjustment range. In each iteration, the digital twin model is called to calculate the pressing correlation parameters of each spatial node and evaluate the optimization objective function value until the optimization objective function value converges to the preset threshold. The optimal combination of equipment control parameters is then output.

[0084] In this embodiment, the characteristic parameters and target quality distribution of the current batch of raw materials are read, and the target quality distribution is mapped to each node of the digital twin model according to spatial coordinates. Using equipment control parameters as decision variables and the weighted deviation norm of the suppression correlation parameters of each node and the target value as the optimization objective, upper and lower limit constraints are set for each control parameter. Initialize the particle swarm. The position vector of each particle represents a set of equipment control parameters. Each dimension of the equipment control parameter set corresponds to pressing force, pressing speed, holding time, mold temperature, and demolding speed, respectively. The initial values ​​of each dimension are randomly generated within the allowable adjustment range, and each particle is assigned an initial velocity vector. For each particle, the equipment control parameter set corresponding to its position vector is input into the calibrated digital twin model. The digital twin model calculates the local density and local moisture content of each spatial node of the tea brick body. Obtain the expected density value and expected moisture content value of each spatial node in the target quality distribution. For each spatial node, calculate the first deviation between the local density and the expected density value of the node, and the second deviation between the local moisture content and the expected moisture content value of the node. Assign preset first weights and second weights to the first deviation and the second deviation, respectively. The first weight and the second weight are determined according to the degree of influence of the uniformity of tea brick density and moisture content on the quality of the finished product, and the sum of the two is 1. For example, the first weight is 0.6 and the second weight is 0.4. Calculate the weighted average. The spatial node deviation is calculated; the weighted deviations of all spatial nodes are summed and squared to obtain the optimization objective function value corresponding to the particle; the current optimization objective function value of each particle is compared with its individual historical best objective function value. If the current value is smaller, the current position of the particle is updated to its individual best position; the current optimization objective function value of all particles is compared with the global historical best objective function value. If the current value is smaller, the current position of the corresponding particle is updated to the global best position; the velocity vector of each particle is updated according to its individual best position and global best position, and the position vector of the particle is updated according to the updated velocity vector; boundary constraint processing is performed on each component of the updated position vector, truncating the components exceeding the upper limit of the allowable adjustment range to the upper limit value, and truncating the components below the lower limit of the allowable adjustment range to the lower limit value; the above steps are repeated until the preset maximum number of iterations is reached or the change of the global best objective function value for several consecutive iterations is less than the preset convergence threshold, and the combination of equipment control parameters corresponding to the global best position is output. The process is iterated repeatedly until the objective function converges to a preset threshold, and the optimal combination of equipment control parameters is output to guide actual pressing production. The upper and lower limits are determined based on the rated parameters on the nameplate of the pressing equipment and the production process specifications of Fu brick tea. For example, the upper and lower limits of pressing force are 5MPa to 25MPa, the upper and lower limits of pressing speed are 1mm / s to 10mm / s, the upper and lower limits of holding time are 30s to 300s, the upper and lower limits of mold temperature are 20℃ to 80℃, and the upper and lower limits of demolding speed are 1mm / s to 5mm / s.

[0085] Step S110: Send the optimal combination of equipment control parameters to the compression molding equipment for execution.

[0086] Among them, the pressing and forming equipment is a brick tea press and its control system; "downloading" refers to writing the optimal parameter combination into the controller of the pressing and forming equipment through the communication interface, replacing the current control parameters, so that the equipment can perform the pressing action according to the optimized parameters.

[0087] For example, the optimal combination of equipment control parameters obtained from the solution is encapsulated into control commands according to a communication protocol and sent to the programmable logic controller (PLC) of the pressing and molding equipment via industrial Ethernet or fieldbus. After verifying the integrity of the commands, the controller updates the set values ​​of pressing force, pressing speed, holding time, mold temperature, and demolding speed, and executes them in the next pressing cycle. The equipment feeds back the actual values ​​of each parameter, and after confirming that they are consistent with the set values, it enters the formal production pressing stage.

[0088] For example, the pressing force, pressing speed, holding time, mold temperature, and demolding speed in the optimal equipment control parameter combination are converted into control command values ​​that the pressing and molding equipment can execute. Each control command value is sent to the programmable logic controller of the pressing and molding equipment through the equipment control interface. The programmable logic controller controls the press hydraulic system, mold heating system, and demolding mechanism to act according to the command values. During the execution of the pressing and molding equipment, the actual operating parameters of the equipment are collected in real time and compared with the sent control command values. If a deviation occurs, fine-tuning is performed through the feedback control loop.

[0089] In this embodiment, the pressing force, pressing speed, holding time, mold temperature, and demolding speed in the optimal equipment control parameter combination are converted into corresponding control command values ​​according to the equipment communication protocol and sent to the programmable logic controller (PLC) of the pressing and molding equipment via an industrial Ethernet interface. Upon receiving the command, the PLC controls the press hydraulic system to adjust the overflow valve pressure according to the pressing force command, adjusts the proportional valve flow according to the speed command, controls the mold heating system to adjust the heating power according to the temperature command, and controls the demolding mechanism to adjust the demolding ejection speed according to the speed command. During execution, the actual pressing force, pressing speed, holding time, and mold temperature are collected in real time by pressure sensors, displacement sensors, and temperature sensors and continuously compared with the sent command values. If the deviation exceeds a preset threshold, the feedback control loop fine-tunes the corresponding actuator until each parameter stabilizes within the allowable range of the command values.

[0090] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0091] Based on the same inventive concept, this application also provides a digital twin parameter optimization system for pressing Pingli Gynostemma pentaphyllum Fu brick tea, used to implement the aforementioned method for optimizing digital twin parameters in pressing Pingli Gynostemma pentaphyllum Fu brick tea. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the digital twin parameter optimization system for pressing Pingli Gynostemma pentaphyllum Fu brick tea provided below can be found in the limitations of the digital twin parameter optimization method for pressing Pingli Gynostemma pentaphyllum Fu brick tea described above, and will not be repeated here.

[0092] In one exemplary embodiment, such as Figure 2 As shown, a digital twin parameter optimization system for pressing Pingli Gynostemma pentaphyllum Fu brick tea is provided, including:

[0093] The parameter mapping module 202 is used to obtain the raw material characteristic parameters, press process parameters and molding quality indicators during the brick tea pressing process, and to construct the initial mapping relationship between the raw material characteristic parameters, press process parameters and molding quality indicators.

[0094] The digital modeling module 204 is used to establish a three-dimensional geometric model of the tea brick blank and divide the three-dimensional geometric model into multiple spatial nodes; to establish a digital twin model of the brick tea pressing and forming equipment, the digital twin model including a discrete element model of tea material-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, a moisture content field model inside the tea brick, and a correlation model between the pressing correlation parameters of the tea brick blank and the equipment control parameters; wherein, the correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish the spatial mapping relationship between the pressing correlation parameters and the equipment control parameters at each spatial node;

[0095] Model calibration module 206 is used to calibrate the digital twin model using the initial mapping relationship;

[0096] The parameter optimization module 208 is used to obtain the characteristic parameters and target quality distribution of the raw materials in the current batch. Using the calibrated digital twin model, the optimization objective is to minimize the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution. The optimal combination of equipment control parameters is solved with the allowable adjustment range of equipment control parameters as the constraint.

[0097] The equipment control module 210 is used to send the optimal combination of equipment control parameters to the compression molding equipment for execution.

[0098] The various modules in the aforementioned digital twin parameter optimization system for pressing Pingli Gynostemma pentaphyllum brick tea can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0099] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memories can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least relational and non-relational databases. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0100] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for optimizing digital twin parameters in the pressing and molding of Pingli Gynostemma pentaphyllum Fu brick tea, characterized in that, The method includes: Obtain the raw material characteristic parameters, press process parameters, and molding quality indicators during the brick tea pressing process, and construct an initial mapping relationship between the raw material characteristic parameters, the press process parameters, and the molding quality indicators; A three-dimensional geometric model of the tea brick blank is established, and the three-dimensional geometric model is divided into multiple spatial nodes. A digital twin model of the brick tea pressing and forming equipment is established. The digital twin model includes a discrete element model of tea material-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, a moisture content field model inside the tea brick, and a correlation model between the pressing correlation parameters of the tea brick blank and the equipment control parameters. The correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish a spatial mapping relationship between the pressing correlation parameters and the equipment control parameters at each spatial node. The digital twin model is calibrated using the initial mapping relationship; Obtain the characteristic parameters and target quality distribution of the current batch of raw materials. Using the calibrated digital twin model, with the optimization objective of minimizing the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution, and with the allowable adjustment range of the equipment control parameters as the constraint, solve for the optimal combination of equipment control parameters. The optimal combination of equipment control parameters is sent to the compression molding equipment for execution.

2. The method according to claim 1, characterized in that, The initial mapping relationship between the raw material characteristic parameters, the press process parameters, and the molding quality indicators includes: Obtain raw material characteristic parameters, press process parameters, and corresponding molding quality index data recorded in historical production batches. The raw material characteristic parameters include tea moisture content, Gynostemma pentaphyllum ratio, particle size distribution, and mixing uniformity. The press process parameters include pressing force, pressing speed, holding time, mold temperature, and demolding speed. The molding quality indexes include uniformity of tea brick density distribution and uniformity of moisture content. The obtained raw material characteristic parameters, press process parameters and molding quality index data are normalized and abnormal samples are removed. Using raw material characteristic parameters and press process parameters as input features, and molding quality indicators as output labels, a multiple linear regression model is used to train an initial mapping model to obtain the initial mapping relationship between the raw material characteristic parameters, the press process parameters, and the molding quality indicators.

3. The method according to claim 1, characterized in that, The establishment of the discrete element model of the tea-Gynostemma pentaphyllum mixture includes: The particle size distribution, particle density, elastic modulus, friction coefficient, and cohesive force parameters of tea granules and Gynostemma pentaphyllum granules were obtained. In the discrete element simulation software, a set of tea particles and a set of Gynostemma pentaphyllum particles are generated according to the particle size distribution, and then mixed in a virtual mold according to a preset ratio. Contact models are set for tea granules and Gynostemma pentaphyllum granules respectively. The contact models include the normal contact stiffness, tangential contact stiffness, rolling friction coefficient and interparticle bonding strength. The parameters of the established discrete element model were calibrated. By comparing the simulated and measured repose angles, the contact parameters were adjusted until they were consistent, thus obtaining the discrete element model of the tea-Gynostemma pentaphyllum mixed particles.

4. The method according to claim 1, characterized in that, The establishment of the mold-indenter finite element model includes: Obtain the geometric dimensions, material properties, and assembly relationship of the mold and the indenter, wherein the material properties include elastic modulus, Poisson's ratio, and yield strength; Finite element preprocessing software was used to mesh the mold and the pressure head, and contact elements were set on the contact surface between the mold and the pressure head. Apply mold fixing constraints and indenter movement constraints, and apply pressing force load to the top of the indenter; By setting material nonlinearity and geometric nonlinearity options, a finite element model of the mold-pressing head is established to calculate the stress distribution and deformation of the mold and pressing head during the pressing process.

5. The method according to claim 1, characterized in that, The establishment of the internal moisture content field model of the tea brick includes: Treating the tea brick blank as a porous medium, a convection-diffusion equation is established to show the change of moisture content with time and space during the pressing process. The moisture migration rate is related to the local pressure gradient, temperature gradient and moisture content gradient. The initial moisture content, saturated moisture content, and moisture diffusion coefficient of the tea-Gynostemma pentaphyllum mixture were obtained. Set the mold wall as the water discharge boundary, and set the contact surface between the pressure head and the tea blank as the dynamic pressure boundary; The convection-diffusion equation is solved discretically using the finite volume method to obtain the change in moisture content of each spatial node inside the tea brick over time during the pressing process, thus forming a moisture content field model inside the tea brick.

6. The method according to claim 1, characterized in that, The establishment of the correlation model between the pressing parameters of the tea brick blank and the equipment control parameters includes: The three-dimensional geometric model of the tea brick blank is divided into multiple spatial nodes, and each spatial node corresponds to a local position. Using the aforementioned discrete element model of tea-Gynostemma pentaphyllum mixed particles and finite element model of mold-pressing head, simulations were performed for multiple sets of different equipment control parameter combinations. Pressing correlation parameters at each spatial node were extracted, including local pressure, local density, and local moisture content. Using equipment control parameters as input and suppression correlation parameters at each spatial node as output, a spatial node-level surrogate model is established. The surrogate model adopts a radial basis function neural network. The proxy models of each spatial node are combined according to the location of the spatial node to form an association model that uses the spatial nodes of the three-dimensional geometric model as anchor points and represents the spatial mapping relationship between the suppression association parameters and the equipment control parameters at each spatial node.

7. The method according to claim 1, characterized in that, The calibration of the digital twin model using the initial mapping relationship includes: Obtain the characteristic parameters of the current batch of raw materials and the actual press process parameters, and use the initial mapping relationship to calculate the corresponding predicted values ​​of molding quality indicators; The press process parameters are input into the digital twin model for simulation to obtain the spatial distribution of pressing-related parameters and the simulation results of moisture content field of tea brick blank; The molding quality index predicted by the digital twin model is calculated based on the simulation results, and compared with the predicted value of the molding quality index calculated by the initial mapping relationship to calculate the deviation. With the goal of minimizing the deviation, the particle contact parameters, moisture diffusion coefficient, and mold-indenter contact stiffness in the digital twin model are adjusted using the parameter inversion method until the output of the digital twin model is consistent with the output of the initial mapping relationship within the allowable deviation range, thus completing the calibration of the digital twin model.

8. The method according to claim 1, characterized in that, The process utilizes the calibrated digital twin model, with the optimization objective of minimizing the spatial deviation between the suppression correlation parameters of each spatial node and the target quality distribution, and with the allowable adjustment range of the equipment control parameters as constraints, to solve for the optimal combination of equipment control parameters, including: Obtain the characteristic parameters and target mass distribution of the raw materials for the current batch, wherein the target mass distribution is the expected density value and expected moisture content value of each spatial node of the tea brick; Input the characteristic parameters of the current batch of raw materials into the calibrated digital twin model, use the equipment control parameters as decision variables, and establish an optimization objective function. The optimization objective function is to assign preset density deviation weights and preset moisture content deviation weights to the first deviation between the local density and the expected density value and the second deviation between the local moisture content and the expected moisture content value of each spatial node, respectively, and then calculate the square root of the weighted sum of squares as the weighted L2 norm of the spatial deviation. Set the allowable adjustment range constraints for the equipment control parameters, including the upper and lower limits of pressing force, pressing speed, holding time, mold temperature, and demolding speed; Within the allowable adjustment range, the device control parameters are iteratively searched. In each iteration, the digital twin model is called to calculate the suppression correlation parameters of each spatial node and evaluate the optimization objective function value until the optimization objective function value converges to the preset threshold, and the optimal combination of device control parameters is output.

9. The method according to claim 1, characterized in that, The step of sending the optimal equipment control parameter combination to the compression molding equipment for execution includes: The pressing force, pressing speed, holding time, mold temperature and demolding speed in the optimal equipment control parameter combination are respectively converted into control command values ​​that the pressing and molding equipment can execute; The control command values ​​are sent to the programmable logic controller of the pressing and molding equipment through the equipment control interface. The programmable logic controller controls the press hydraulic system, mold heating system and demolding mechanism to operate according to the command values. During the pressing and molding process, the actual operating parameters of the equipment are collected in real time and compared with the control command values ​​issued. If a deviation occurs, fine-tuning is performed through the feedback control loop.

10. A digital twin parameter optimization system for pressing Pingli Gynostemma pentaphyllum Fu brick tea, characterized in that, The system includes: The parameter mapping module is used to obtain the raw material characteristic parameters, press process parameters and forming quality indicators during the brick tea pressing process, and to construct the initial mapping relationship between the raw material characteristic parameters, the press process parameters and the forming quality indicators. The digital modeling module is used to establish a three-dimensional geometric model of the tea brick blank and divide the three-dimensional geometric model into multiple spatial nodes; it also establishes a digital twin model of the brick tea pressing and forming equipment, which includes a discrete element model of tea material-Gynostemma pentaphyllum mixed particles, a finite element model of mold-pressing head, a moisture content field model inside the tea brick, and a correlation model between the pressing correlation parameters of the tea brick blank and the equipment control parameters; wherein, the correlation model uses the spatial nodes of the three-dimensional geometric model as anchor points to establish a spatial mapping relationship between the pressing correlation parameters and the equipment control parameters at each spatial node; A model calibration module is used to calibrate the digital twin model using the initial mapping relationship; The parameter optimization module is used to obtain the characteristic parameters and target quality distribution of the raw materials in the current batch. Using the calibrated digital twin model, the optimization objective is to minimize the spatial deviation between the compression correlation parameters of each spatial node and the target quality distribution. The optimal combination of equipment control parameters is solved with the allowable adjustment range of equipment control parameters as the constraint. The equipment control module is used to send the optimal combination of equipment control parameters to the compression molding equipment for execution.

Images