Multi-source emission and quality collaborative optimization method for mixing and injection molding integrated production line

Abstract

The invention provides a multi-source emission and quality collaborative optimization method for a mixing and injection molding integrated production line. The method is particularly suitable for a mixing and injection molding integrated production line, and aims to synergistically optimize the multi-source environmental emission and the product quality of the mixing and injection molding integrated production line. Through establishment of a quantitative model and multi-target optimization solution, process parameter optimization of emission reduction and quality guarantee targets is realized. By adjusting key process parameters of a production line, reduction of carbon emission, VOCs emission and PM2.5 emission is completed on the premise that the product percent of pass or quality requirements are guaranteed. The invention provides a decision support method which is close to the actual production condition of a production line for reducing the comprehensive environmental emission on the premise of ensuring the product quality of the mixing injection molding production line through systematic modeling and an optimization framework.

Description

Technical Field

[0001] This invention relates to the field of green and intelligent manufacturing technology, and in particular to a method for synergistic optimization of multi-source emissions and quality in an integrated mixing and injection molding production line. Background Technology

[0002] Integrated mixing and injection molding technology is one of the core processes in modern polymer material processing and molding. It achieves efficient one-step manufacturing of materials by integrating mixing and granulation with injection molding. Currently, to improve production efficiency, the industry commonly intervenes in the production process by adjusting process parameters. Existing optimization technologies mainly focus on traditional production indicators, emphasizing improving product yield, reducing direct energy consumption of equipment, or controlling individual production costs. In terms of process control and prediction, some existing technologies are beginning to introduce data-driven methods, such as using neural network algorithms to establish mapping models between process parameters and product quality to assist production decisions. Furthermore, improvements to production lines are mostly concentrated on upgrading hardware or macro-management systems at the workshop level. However, with the increasing demands for green and intelligent manufacturing, the aforementioned existing technologies have revealed the following significant defects and shortcomings in practical applications:

[0003] A. Existing optimization methods often focus on one aspect while neglecting others, primarily addressing a single or a few traditional objectives. In actual production, the mixing and injection molding process not only generates carbon emissions but also produces volatile organic compounds (VOCs) and particulate matter due to polymer thermal degradation and additive volatilization. Current technologies fail to consider these three multi-source environmental emission indicators within the same optimization framework as product qualification rate. This leads to the neglect of other pollutant emissions when pursuing high quality or low energy consumption, making it difficult to minimize the overall environmental burden while ensuring product quality.

[0004] B. Currently widely used quality prediction models, such as neural networks, are black-box models. While they improve prediction accuracy to some extent, their internal mechanisms are opaque, their interpretability is poor, and their computational processes are complex. These shortcomings make it difficult to transform such models into well-structured, computationally efficient explicit mathematical functions, preventing them from being directly and quickly invoked as deterministic constraints by multi-objective optimization algorithms. This significantly limits the real-time response capability and computational efficiency of optimization algorithms in practical engineering applications.

[0005] C. Existing technological improvements remain at the level of partial hardware upgrades or general management systems, with insufficient integration of hardware and software. For complex systems like integrated mixing and injection molding production lines, which are characterized by multiple variables and strong coupling, existing technologies lack a systematic software approach that can deeply integrate mechanistic models, data mining, and multi-objective optimization algorithms. This results in the inability to fully exploit the green production potential of the production line under dynamic operating conditions, and the difficulty in automatically searching for the optimal combination of process parameters that meets both quality constraints and multi-dimensional environmental requirements within complex process windows.

[0006] Therefore, developing a multi-source emission and quality synergistic optimization method for an integrated mixing and injection molding production line is of great significance. Summary of the Invention

[0007] The purpose of this invention is to provide a method for synergistic optimization of multi-source emissions and quality in an integrated mixing and injection molding production line, so as to solve the problems existing in the prior art.

[0008] The technical solution adopted to achieve the purpose of this invention is as follows: a method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line, comprising the following steps:

[0009] S1) Establish quantitative emission calculation functions and optimization objective functions. For the integrated mixing and injection molding production line, establish separate calculation functions for carbon emissions, VOCs emissions, and particulate matter emissions. Construct an optimization objective function based on these three types of emission calculation functions.

[0010] S2) Determine the decision variables and constraints. Select the process parameters for production line adjustment as decision variables. Determine the feasible domain of the decision variables based on the equipment operating conditions, which will serve as constraints for the optimization process.

[0011] S3) Establish a functional relationship between process parameters and product qualification rate, which can be used as a product qualification rate prediction function.

[0012] S4) Perform multi-objective optimization. With the optimization objectives of minimizing carbon emissions, minimizing VOC emissions, and minimizing particulate matter emissions, and based on the feasible region of the decision variables, a multi-objective optimization algorithm is used to perform optimization calculations to obtain an initial Pareto optimal solution set containing multiple combinations of process parameters.

[0013] S5) Substitute each set of process parameters in the initial Pareto optimal solution set into the product qualification rate prediction function to calculate the corresponding predicted qualification rate. Eliminate process parameter combinations with predicted qualification rates lower than the preset standard value to obtain the final optimal parameter scheme.

[0014] S6) Outputs the optimal process parameter scheme after screening, for use in production line control.

[0015] Furthermore, in step S1), based on energy consumption monitoring and carbon flow analysis, the total carbon emission rate D is calculated using the following model. carbon :

[0016] D carbon =E total ×EF grid

[0017] In the formula, EF grid E is the carbon emission factor of the power grid. total Let E be the total energy consumption of the mixing and injection molding process. total It is the sum of the energy consumption of each step:

[0018] E total =E comp +E inj +E hold +E cool +E base

[0019] In the formula, E comp Energy consumption for mixing and plasticizing Einj For injection energy consumption, E hold To maintain pressure and energy consumption, E cool For cooling energy consumption, E base Based on basic energy consumption.

[0020] Furthermore, the energy consumption calculation formulas for each step are as follows:

[0021] Energy consumption for compounding and plasticizing:

[0022] Injection energy consumption:

[0023] Pressure holding energy consumption:

[0024] In the formula, n screw V is the screw speed. inj For injection speed, P hold To maintain the pressure, t comp t inj t hold These are the plasticizing, injection, and holding times, respectively. screw Let η be the cross-sectional area of ​​the screw. motor and η system These are the motor efficiency and the system efficiency, respectively.

[0025] Furthermore, in step S1), an empirical-mechanistic model for calculating VOCs emissions related to melt temperature is established based on the Arrhenius equation:

[0026]

[0027] In the formula, D VOCs k represents the VOCs emission rate. VOCs Here, A is the material-specific coefficient, Ea is the pre-exponential factor, R is the apparent activation energy, and T is the ideal gas constant. melt t is the absolute temperature of the melt. exposure This represents the cumulative time the material spends in the molten state.

[0028] Furthermore, in step S1), a particulate matter emission calculation model related to mechanical motion parameters is established:

[0029]

[0030] Among them, D PM2.5 For PM2.5 emission rate, k PM Based on the emission factor, n ref With V ref These are the reference rotational speed and reference speed, respectively; α and β are empirical exponents; N cycle n is the number of production cycles. screw V is the screw speed. inj Injection speed.

[0031] Further, in step S2), the decision variable is defined as a vector x = [T, V, P, N]. The constraints include the upper and lower limits of each decision variable:

[0032]

[0033]

[0034]

[0035]

[0036] In the formula, T is the melt temperature, v is the injection speed, P is the holding pressure, and n is the screw speed.

[0037] Furthermore, in step S3), the pass rate function is used as one of the optimization objective functions to construct the objective function.

[0038]

[0039] In the formula, This indicates the carbon emission rate per unit or per batch of products. This indicates the VOCs emission rate per unit or per batch of products. This indicates the particulate matter emission rate per unit or per batch of products. This indicates the pass rate.

[0040] Furthermore, in step S3), the establishment of the product qualification rate prediction function specifically includes:

[0041] Multiple sets of process parameters and their corresponding actual yield observation data were collected, and a multiple linear regression model was constructed based on the least squares method.

[0042]

[0043] Where Q is the product qualification rate, N is the screw speed, T is the melt temperature, V is the injection speed, P is the holding pressure, and β0~β4 are regression coefficients.

[0044] Furthermore, in step S4), the multi-objective optimization algorithm employs a non-dominated sorting genetic algorithm with an elitist strategy. Step S4 specifically includes the following sub-steps:

[0045] S4.1) Randomly generate the initial parent population within the feasible region of the decision variables.

[0046] S4.2) Using the carbon emission, VOCs emission, and particulate matter emission calculation functions established in S1), calculate the three objective function values ​​for each individual in the population.

[0047] S4.3) Perform non-dominated sorting and crowding calculation on the population, and generate offspring populations through selection, crossover and mutation operations.

[0048] S4.4) Merge the parent and offspring populations and select a new generation of populations based on non-dominance level and crowding.

[0049] S4.5) Repeat the evaluation and evolution operations until the preset maximum number of iterations is reached, and output the Pareto optimal solution set.

[0050] The technical effects of this invention are beyond doubt:

[0051] A. It achieves a synergistic win-win situation between multi-source environmental emissions and product quality, overcoming the limitations of single-objective optimization. It integrates environmental indicators from three dimensions—carbon emissions, VOCs emissions, and particulate matter emissions—with product qualification rate indicators into a single system for evaluation. Through quantitative models and the setting of constraints, it ensures that product quality standards are strictly met while reducing overall environmental impact.

[0052] B. Through mechanistic analysis and data regression, a clear functional relationship between process parameters, multi-source emissions, and compliance rate was constructed. This explicit mathematical model is not only computationally efficient and can be directly used by optimization algorithms, but also has good physical interpretability, making process adjustment more transparent and controllable, and effectively reducing trial-and-error costs.

[0053] C. A non-dominated sorting genetic algorithm with an elitist strategy is used to solve the problem. The output is not a single fixed solution, but a set of Pareto optimal process parameter solutions containing different trade-offs. This gives production managers a great deal of decision-making space, enabling them to flexibly select the most suitable process scheme from the solution set according to different production scenarios, achieving the best balance between production efficiency and environmental benefits.

[0054] D. The screening mechanism eliminates the interference of secondary factors, reduces the dimensionality of the model and the difficulty of solving it, and makes the optimization method more adaptable to the actual working conditions of the production line, making it easier to deploy and promote in the industrial field. Attached Figure Description

[0055] Figure 1 Architecture diagram of a multi-source emission and quality synergistic optimization method for integrated mixing and injection molding production lines;

[0056] Figure 2 A flowchart for energy consumption classification in an integrated mixing and injection molding production line;

[0057] Figure 3 A flowchart for collaborative optimization of an integrated mixing and injection molding production line. Detailed Implementation

[0058] The present invention will be further described below with reference to embodiments, but it should not be construed that the scope of the present invention is limited to the following embodiments. Various substitutions and modifications made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention should be included within the scope of protection of the present invention.

[0059] Example 1:

[0060] See Figure 1 This embodiment provides a method for synergistic optimization of multi-source emissions and quality in an integrated mixing and injection molding production line, including the following steps:

[0061] S1) Establish quantitative emission calculation functions and optimization objective functions. For the integrated mixing and injection molding production line, establish separate calculation functions for carbon emissions, VOCs emissions, and particulate matter emissions. Construct an optimization objective function based on these three types of emission calculation functions.

[0062] S2) Determine the decision variables and constraints. Select the process parameters for production line adjustment as decision variables. Determine the feasible domain of the decision variables based on the equipment operating conditions, which will serve as constraints for the optimization process.

[0063] S3) Establish a functional relationship between process parameters and product qualification rate, which can be used as a product qualification rate prediction function.

[0064] S4) Perform multi-objective optimization. With the optimization objectives of minimizing carbon emissions, minimizing VOC emissions, and minimizing particulate matter emissions, and based on the feasible region of the decision variables, a multi-objective optimization algorithm is used to perform optimization calculations to obtain an initial Pareto optimal solution set containing multiple combinations of process parameters.

[0065] S5) Substitute each set of process parameters in the initial Pareto optimal solution set into the product qualification rate prediction function to calculate the corresponding predicted qualification rate. Eliminate process parameter combinations with predicted qualification rates lower than the preset standard value to obtain the final optimal parameter scheme.

[0066] S6) Outputs the optimal process parameter scheme after screening, for use in production line control.

[0067] Example 2:

[0068] The main content of this embodiment is the same as that of Embodiment 1, wherein, see [link / reference]. Figure 2 For the integrated mixing and injection molding production line, the carbon mass of the materials is conserved before and after processing, and the process carbon can be included in the overall energy-driven traceability. Therefore, carbon emissions are mainly calculated based on energy consumption. For this production line, the driving force of each part mainly comes from electricity consumption. In step S1), based on energy consumption monitoring and carbon flow analysis, the total carbon emission rate D is calculated using the following model. carbon :

[0069] D carbon =E total ×EF grid

[0070] In the formula, EF grid E is the carbon emission factor of the power grid. total Let E be the total energy consumption of the mixing and injection molding process. total It is the sum of the energy consumption of each step:

[0071] E total =E comp +E inj +E hold +E cool +E base

[0072] In the formula, E comp Energy consumption for mixing and plasticizing Einj For injection energy consumption, E hold To maintain pressure and energy consumption, E cool For cooling energy consumption, E base Based on basic energy consumption.

[0073] Since the total energy consumption is the sum of the energy consumption of each process step, and the energy consumption of each process step is related to the setting of process parameters, an energy consumption model for each process step is further constructed. The calculation formula for the energy consumption of each process step is as follows:

[0074] Energy consumption for compounding and plasticizing:

[0075] Injection energy consumption:

[0076] Pressure holding energy consumption:

[0077] In the formula, n screw V is the screw speed. inj For injection speed, P hold To maintain the pressure, t comp t inj t hold These are the plasticizing, injection, and holding times, respectively. screw Let η be the cross-sectional area of ​​the screw. motor and η system These are the motor efficiency and the system efficiency, respectively.

[0078] VOCs mainly originate from polymer thermal degradation and additive volatilization. VOCs are generated from multiple sources, such as additives in raw materials and auxiliary materials. Additives like solvents, plasticizers, and mold release agents volatilize at processing temperatures of approximately 150-250℃. Polymer thermal processing and decomposition also produce thermal decomposition at certain temperatures, and the coating process in post-processing also generates volatilization. Small-molecule organic compounds added to improve processing or product performance are one of the main sources of VOC emissions. During normal processing or overheating, when the thermal decomposition temperature is exceeded, polymer chains break or oxidize, producing low-molecular-weight organic compounds.

[0079] VOCs are mainly generated from multiple sources, and the specific source substances and key temperatures are shown in the table below:

[0080] Table 1 Sources of VOCs Emissions

[0081]

[0082] In step S1), an empirical-mechanistic model for calculating VOCs emissions related to melt temperature is established based on the Arrhenius equation:

[0083]

[0084] In the formula, D VOCs k represents the VOCs emission rate. VOCs Here, A is the material-specific coefficient, Ea is the pre-exponential factor, R is the apparent activation energy, and T is the ideal gas constant. melt t is the absolute temperature of the melt. exposure This represents the cumulative time the material spends in the molten state.

[0085] Typical particulate matter PM2.5 mainly consists of tiny particles generated by material friction and degradation. In step S1), a particulate matter emission calculation model related to mechanical motion parameters is established:

[0086]

[0087] Among them, D PM2.5 For PM2.5 emission rate, k PM Based on the emission factor, n ref With V ref These are the reference rotational speed and reference speed, respectively; α and β are empirical exponents; N cycle n is the number of production cycles. screw V is the screw speed. inj Injection speed.

[0088] Example 3:

[0089] This embodiment is similar in main content to Embodiment 1 or 2. However, through mechanistic analysis and production data mining, it focuses on evaluating the sensitivity of each parameter to multi-source emissions and product quality, as well as its process controllability in the production line. Finally, a subset of decision variables consisting of four key process parameters is selected. In step S2), the decision variables are defined as vector x = [T, V, P, N]. The constraints include the upper and lower limits of each decision variable:

[0090]

[0091]

[0092]

[0093]

[0094] In the formula, T is the melt temperature, v is the injection speed, P is the holding pressure, and n is the screw speed.

[0095] Example 4:

[0096] The main content of this embodiment is the same as any one of embodiments 1 to 3. The optimization process also needs to meet the constraint of the pass rate. Two schemes are designed here:

[0097] Option 1: In step S3), the pass rate function is used as one of the optimization objective functions to construct the objective function.

[0098]

[0099] In the formula, This indicates the carbon emission rate per unit or per batch of products. This indicates the VOCs emission rate per unit or per batch of products. This indicates the particulate matter emission rate per unit or per batch of products. This indicates the pass rate.

[0100] Option 2: Maintain the original three emissions as the objective function. Then, for the optimized set of process parameters, use a pass rate prediction function to predict the achievable pass rate, thereby selecting a subset of process parameter combinations that meet the pass rate requirements. In step S3), collect multiple sets of process parameters and their corresponding actual pass rate observations, and construct a multiple linear regression model based on the least squares principle.

[0101] The prediction function Q is the functional relationship between the product's yield rate and process parameters. Since there is currently no readily available and accurate functional expression for integrated mixing and injection molding production lines, a regression model is used to derive the prediction function. The specific method for establishing this function is as follows:

[0102] The quantitative relationship between the process parameters and the product qualification rate was established using a multiple linear regression method. This method, based on the principle of least squares, aims to fit a predictive linear mathematical model using a set of experimental process parameter data.

[0103] Step 1: Define the regression model

[0104] The independent variables are set as: screw speed (N), melt temperature (T), injection speed (V), and holding pressure (P). The dependent variable is the product qualification rate (Q), which is a continuous value between 0 and 1 (e.g., 95% is denoted as 0.95). The following multiple linear regression model is established:

[0105]

[0106] Where β0 is the regression constant, β1 to β4 are the partial regression coefficients of the corresponding independent variables, and ε is the random error term. The goal of the model is to estimate an optimal set of coefficients that minimizes the overall error between the model's predictions and the actual observations.

[0107] Step 2: Data Acquisition and Matrix Representation

[0108] The system collects m sets of sample data (m is much larger than the number of independent variables, 4) through injection molding experiments or high-fidelity simulations such as Moldex3D. Each set of data contains a specific set of data. and its corresponding pass rate observation value Q i .

[0109] Transforming data into matrix form is fundamental to efficient computation. Definition:

[0110] Design matrix X: a The matrix, where each row represents the independent variable value of a sample, with a constant term of 1 added to the first column. That is... .

[0111] Response vector y: an m×1 column vector containing the pass rates of all observations. .

[0112] Coefficient vector β: A 5×1 column vector to be determined. .

[0113] Step 3: Parameter Estimation

[0114] Using mathematical modeling software, the least squares method is employed for parameter estimation. This method solves for the optimal coefficient vector by minimizing the residual sum of squares (RSS). ( (The estimated value). The sum of squared residuals is defined as:

[0115]

[0116] By taking the derivative and setting it to zero, we obtain the famous normal equation. When When a matrix is ​​of full rank (i.e., invertible), there exists a unique solution:

[0117]

[0118] Solving this equation will yield the specific regression constant term. Partial regression coefficients of various process parameters to .

[0119] Step 4: Model Acquisition and Application

[0120] Substituting the obtained coefficients into the initial model yields the deterministic function used for prediction:

[0121]

[0122] Where Q is the product yield rate, N is the screw speed, T is the melt temperature, V is the injection speed, P is the holding pressure, and β0~β4 are regression coefficients. This function can be used as a prediction module and integrated into subsequent optimization algorithms.

[0123] It is important to note that both Scheme 1 and Scheme 2 aim to simultaneously reduce emissions and improve product qualification rates during the optimization process. The choice between the two methods can be made based on the actual data. During the optimization iteration process, the deterministic function, when substituted into Scheme 1, serves as the optimization objective in the calculation. When substituted into Scheme 2, the deterministic function serves as the key indicator for selecting the optimization results. (Appendix) Figure 1 and attached Figure 3 The drawing is mainly based on the logic of Scheme 2.

[0124] Example 5:

[0125] See Figure 3 This embodiment is essentially the same as any one of embodiments 1-4. NSGA-II (Non-dominated Sorting Genetic Algorithm II) is widely used due to its excellent convergence performance and solution set distribution when handling multi-objective optimization problems. Addressing the complex trade-offs among multiple emission objectives in this study, NSGA-II can effectively explore the non-dominated solution space. It demonstrates superior convergence and distribution of solutions when handling continuous variable multi-objective problems, providing decision-makers with a series of balanced optimization process parameter configurations. See also... Figure 3 In step S4), the multi-objective optimization algorithm employs a non-dominated sorting genetic algorithm with an elitist strategy. Step S4 specifically includes the following sub-steps:

[0126] S4.1) Randomly generate an initial parent population within the feasible region of the decision variables. The algorithm first randomly samples within the decision space of the problem to generate an initial parent population of size N. This initialization process aims to ensure a uniform distribution of individuals in the solution space, providing a diverse starting point for subsequent evolutionary searches. Fitness evaluation involves calculating all objective function values ​​for each individual in the initial population (i.e., a specific set of parameter combinations). In this study, corresponding to the optimization objectives of the three major emission values, each individual will be evaluated and assigned three independent objective function values, thus completing the mapping from the decision space to the objective space.

[0127] S4.2) Using the carbon emission, VOCs emission, and particulate matter emission calculation functions established in S1), calculate the three objective function values ​​for each individual in the population.

[0128] S4.3) Perform non-dominated sorting and crowding calculation on the population, and generate offspring populations through selection, crossover and mutation operations.

[0129] S4.4) Merge the parent and offspring populations and select a new generation of populations based on non-dominance level and crowding.

[0130] S4.5) Repeat the evaluation and evolution operations until the preset maximum number of iterations is reached, and output the Pareto optimal solution set.

[0131] Each solution represents a set of process parameters that achieve different trade-offs among the three emission types while meeting quality constraints, and includes corresponding predicted compliance rates. Engineers can select the final implementation plan from the solution set based on the specific environmental policy priorities.

Claims

1. A method for synergistic optimization of multi-source emissions and quality in an integrated mixing and injection molding production line, characterized in that, Includes the following steps: S1) Establish quantitative emission calculation functions and optimization objective functions; for the integrated mixing and injection molding production line, establish carbon emission calculation functions, VOCs emission calculation functions, and particulate matter emission calculation functions respectively; construct optimization objective functions based on the three types of emission calculation functions; S2) Determine the decision variables and constraints; The process parameters for production line adjustments were selected as decision variables; The feasible domain range of decision variables is determined based on the equipment operating conditions, serving as constraints for the optimization process; S3) Establish a functional relationship between process parameters and product qualification rate, and use it as a product qualification rate prediction function; S4) Perform multi-objective optimization solution; With the optimization objectives of minimizing carbon emissions, minimizing VOCs emissions, and minimizing particulate matter emissions, a multi-objective optimization algorithm is used to perform optimization calculations based on the feasible domain range of decision variables, and an initial Pareto optimal solution set containing multiple combinations of process parameters is obtained. S5) Substitute each set of process parameters in the initial Pareto optimal solution set into the product qualification rate prediction function to calculate the corresponding predicted qualification rate; remove process parameter combinations whose predicted qualification rate is lower than the preset standard value to obtain the final optimal parameter scheme. S6) Outputs the optimal process parameter scheme after screening, for use in production line control.

2. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that: In step S1), based on energy consumption monitoring and carbon flow analysis, the total carbon emission rate D is calculated using the following model. carbon : D carbon =E total ×EF grid In the formula, EF grid E is the carbon emission factor of the power grid. total Let E be the total energy consumption of the mixing and injection molding process. total It is the sum of the energy consumption of each step: AND total =E comp +E inj +E hold +E cool +E base In the formula, E comp Energy consumption for mixing and plasticizing Einj For injection energy consumption, E hold To maintain pressure and energy consumption, E cool For cooling energy consumption, E base Based on basic energy consumption.

3. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 2, characterized in that, The formulas for calculating energy consumption in each step are as follows: Energy consumption for compounding and plasticizing: Injection energy consumption: Pressure holding energy consumption: In the formula, n screw V is the screw speed. inj For injection speed, P hold To maintain the pressure, t comp t inj t hold These are the plasticizing, injection, and holding times, respectively. screw Let η be the cross-sectional area of ​​the screw. motor and η system These are the motor efficiency and the system efficiency, respectively.

4. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that: In step S1), an empirical-mechanistic model for calculating VOCs emissions related to melt temperature is established based on the Arrhenius equation: In the formula, D VOCs k represents the VOCs emission rate. VOCs Here, A is the material-specific coefficient, Ea is the pre-exponential factor, R is the apparent activation energy, and T is the ideal gas constant. melt t is the absolute temperature of the melt. exposure This represents the cumulative time the material spends in the molten state.

5. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that: In step S1), a particulate matter emission calculation model related to mechanical motion parameters is established: Among them, D PM2.5 For PM2.5 emission rate, k PM Based on the emission factor, n ref With V ref These are the reference rotational speed and reference speed, respectively; α and β are empirical exponents; N cycle n is the number of production cycles. screw V is the screw speed. inj Injection speed.

6. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that, In step S2), the decision variables are defined as vectors x = [T, V, P, N]; the constraints include the upper and lower limits of each decision variable. In the formula, T is the melt temperature, v is the injection speed, P is the holding pressure, and n is the screw speed.

7. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that, In step S3), the pass rate function is used as one of the optimization objective functions to construct the objective function; In the formula, This indicates the carbon emission rate per unit or per batch of products. This indicates the VOCs emission rate per unit or per batch of products. This indicates the particulate matter emission rate per unit or per batch of product. This represents the pass rate multiplied by -1.

8. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that, In step S3), the establishment of the product qualification rate prediction function specifically includes: Multiple sets of process parameters and their corresponding actual yield observation data were collected, and a multiple linear regression model was constructed based on the least squares method. Where Q is the product qualification rate, N is the screw speed, T is the melt temperature, V is the injection speed, P is the holding pressure, and β0~β4 are regression coefficients.

9. The method for multi-source emission and quality synergistic optimization of an integrated mixing and injection molding production line according to claim 1, characterized in that, In step S4), the multi-objective optimization algorithm employs a non-dominated sorting genetic algorithm with an elitist strategy; step S4 specifically includes the following sub-steps: S4.1) Randomly generate the initial parent population within the feasible region of the decision variables; S4.2) Using the carbon emission, VOCs emission, and particulate matter emission calculation functions established in S1), calculate the three objective function values ​​for each individual in the population; S4.3) Perform non-dominated sorting and crowding calculation on the population, and generate offspring populations through selection, crossover and mutation operations; S4.4) Merge the parent and offspring populations and select a new generation of populations based on non-dominance level and crowding. S4.5) Repeat the evaluation and evolution operations until the preset maximum number of iterations is reached, and output the Pareto optimal solution set.

Images